THE PRIMARY PREVENTION OF CARDIOVASCULAR DISEASE IN WOMEN WITH AN EMPHASIS ON PHYSICAL ACTIVITY: A SOCIAL MARKETING APPROACH
CEINWEN SAWYER
A submission presented in fulfilment of the requirements of Cardiff Metropolitan University for the degree of Doctor of Philosophy
Cardiff School of Sport Cardiff Metropolitan University October 2012
DECLARATION I certify that this work has not been previously accepted in substance for any degree, and is not being concurrently submitted in candidature for any other degree. I further certify that the whole of this work is the result of my individual efforts, except where otherwise stated. All quotations from books and journals have been acknowledged and a reference section is included. This work is submitted in fulfilment of the requirements of Cardiff Metropolitan University for the degree of Doctor of Philosophy
I hereby give consent for my dissertation, if accepted, to be available for photocopying, and for inter-library loan, and for the title and summary to be made available to outside organisations.
Signed:………………………………………………………………………………………………….. (candidate)
Date:...........................................................................................................................................
Certificate of supervising tutor in respect of the student's submitted work
I am satisfied that this work is the result of the above-named student's own efforts. Signed:………………………………………………………………….............................................. (tutor)
Date:…………………………………………………………………………………………………….
i
ACKNOWLEDGEMENTS
I am indebted to Dr Stephen-Mark Cooper, Dr Bill Davies, and Professor Non Thomas for their support and friendship.
I am grateful to all the women who willingly gave their time to take part in my study. Their enthusiasm and humour was infectious. I was so fortunate to have them as my participants.
ii
DEDICATION
To my girls Mari Louise, Amy Georgina, and Zara Rose Love is all we need
To Non
iii
LIST OF CONTENTS PAGE NUMBER DECLARATION ACKNOWLEDGEMENTS DEDICATIONS LIST OF TABLES LIST OF FIGURES ACRONYMS ABSTRACT
i ii iii xii xiv xv xvii
CHAPTER I: INTRODUCTION TO THE THESIS 1.1
WOMEN AND CARDIOVASCULAR DISEASE
1
1.2
ATHEROGENIC CARDIOVASCULAR DISEASE
1
1.3
PHYSICAL ACTIVITY AND CARDIOVASCULAR DISEASE
4
1.4
A RATIONALE FOR PREVENTING CARDIOVASCULAR DISEASE
6
1.5
A SOCIAL MARKETING APPROACH TO HEALTH
7
1.6
AIMS OF THE THESIS
8
1.7
ORGANISATION OF SUPPORTING LITERATURE AND PREFERRED
9
USE OF TERMINOLOGY
CHAPTER II: STUDY ONE KNOWLEDGE AND PERCEPTIONS OF CARDIOVASCULAR DISEASE, SOURCES OF HEALTH INFORMATION, SCREENING BEHAVIOURS, AND RISK FACTOR STATUS OF WOMEN AGED BETWEEN 25 AND 65 YEARS
2.1.1
ABSTRACT
10
CARDIOVASCULAR DISEASE: A SIGNIFICANT HEALTH ISSUE FOR
11
WOMEN 2.1.2
DECLINING MORTALITY RATES
11
2.1.3
WOMEN‘S KNOWLEDGE OF CARDIOVASCULAR DISEASE
11
2.1.4
AIM OF STUDY ONE
12
2.2.1
INTRODUCTION TO THE REVIEW OF LITERATURE
13
2.2.2
MAJOR RISK FACTORS FOR CARDIOVASCULAR DISEASE
13
iv
2.2.3
SEX DIFFERENCES AND SYMPTOMS OF CARDIOVASCULAR
18
DISEASE: WHY DO THEY MATTER? 2.2.4
WOMEN‘S EXPOSURE TO CARDIOVASCULAR HEALTH
20
INFORMATION 2.2.5
INFLUENCES ON WOMEN‘S HEALTH BEHAVIOUR
21
2.2.6
REACHING A WIDE AUDIENCE THROUGH MEDIA CHANNELS
22
2.2.7
STAGE BASED TAILORING OF HEALTH INFORMATION
23
2.2.8
USING TECHNOLOGY
25
2.2.9
COMPETING HEALTH ISSUES
25
2.2.10 SURVEYS OF WOMEN‘S KNOWLEDGE AND AWARENESS OF
26
CARDIOVASCULAR DISEASE 2.2.11 SUMMARY
28
2.2.12 OBJECTIVES OF STUDY ONE
28
2.3.1
RATIONALE FOR THE SURVEY
29
2.3.2
THE PARTICIPANTS
30
2.3.3
SAMPLING
31
2.3.4
DEVELOPMENT OF THE QUESTIONNAIRE
32
2.3.5
KNOWLEDGE AND AWARENESS OF CARDIOVASCULAR DISEASE
33
2.3.6
SECTION ONE: AGE, MARITAL AND EMPLOYMENT STATUS
34
2.3.7
SECTION ONE: BODY MASS INDEX
34
2.3.8
SECTION ONE: ABDOMINAL ADIPOSITY
35
2.3.9
RISK OF CARDIOVASCULAR DISEASE AND LONG TERM ILLNESS
35
2.3.10 SECTION ONE: CHOLESTEROL SCREENING
36
2.3.11 SECTION ONE: KNOWLEDGE AND AWARENESS OF HIGH BLOOD
36
PRESSURE 2.3.12 SECTION ONE: SMOKING STATUS
36
2.3.13 SECTION ONE: PHYSICAL ACTIVITY STATUS
36
2.3.14 SECTION TWO: KNOWLEDGE OF CARDIOVASCULAR DISEASE AND
37
SOURCES OF INFORMATION 2.3.15 SECTION THREE: PERCEPTIONS OF RISK
39
2.3.16 PEER REVIEW
39
2.3.17 THE PILOT STUDY
39
2.3.18 PROCEDURE
40
2.3.19 ETHICS
41
2.3.20 QUESTIONNAIRE ANALYSIS AND PRESENTATION OF THE
41
RESULTS
v
2.4.1
RESPONSE RATE
42
2.4.2
DEMOGRAPHICS OF THE SAMPLE
42
2.4.3
BODY MASS INDEX (Q 5 AND 6)
44
2.4.4
ABDOMINAL ADIPOSITY (Q 7)
47
2.4.5
SCREENING BEHAVIOUR AND KNOWLEDGE OF HEALTHY
48
CHOLESTEROL AND BLOOD PRESSURE MEASURES (Q 8-11) 2.4.6
SMOKING STATUS (Q 12)
50
2.4.7
PHYSICAL ACTIVITY STATUS (Q13)
50
2.4.8
PERCEPTIONS OF PHYSICAL FITNESS (Q 26)
51
2.4.9
PREFERRED SOURCES OF INFORMATION (Q 14)
52
2.4.10 KNOWLEDGE AND PERCEPTIONS OF CARDIOVASCULAR DISEASE
53
2.4.11 CARDIOVASCULAR DISEASE, BREAST CANCER AND CAUSE OF
54
DEATH IN THE UK (Q17) 2.4.12 KNOWLEDGE OF DIABETES (Q17)
54
2.4.13 KNOWLEDGE OF THE CAUSES OF CARDIOVASCULAR DISEASE
54
(Q 18) 2.4.14 THE INITIATION OF CARDIOVASCULAR DISEASE (Q 19)
56
2.4.15 KNOWLEDGE OF THE SYMPTOMS OF MYOCARDIAL INFARCTION
56
(Q 20) 2.4.16 KNOWLEDGE OF STROKE INJURY (Q 21)
57
2.4.17 KNOWLEDGE OF STROKE SYMPTOMS (Q 22)
57
2.4.18 KNOWLEDGE ABOUT HIGH BLOOD PRESSURE AND PHYSICAL
58
ACTIVITY (Q 9 and Q 24) 2.4.19 CHOLESTEROL AND HEALTHY FUNCTIONING OF THE BODY (Q 24)
58
2.4.20 OVERWEIGHT, OBESITY AND RISK OF DIABETES (Q 24)
59
2.4.21 PERCEPTIONS OF RISK OF CARDIOVASCULAR DISEASE (Q 27)
59
2.5.1
GENERAL CONCLUSIONS TO STUDY ONE
60
2.5.2
PHYSICAL ACTIVITY
60
2.5.3
OVERWEIGHT AND OBESITY
61
2.5.4
SCREENING FOR HIGH BLOOD PRESSURE AND CHOLESTEROL
61
2.5.5
SOURCES OF HEALTH INFORMATION
62
2.5.6
GAPS IN WOMEN‘S KNOWLEDGE OF CARDIOVASCULAR DISEASE
63
2.5.7
PERCEPTIONS OF RISK
64
2.5.8
SUMMARY
64
vi
CHAPTER III: STUDY TWO PERCEIVED AND MEASURED AEROBIC FITNESS, AND PHYSICAL ACTIVITY AND OBESITY STATUS IN WOMEN AGED BETWEEN 25 AND 65 YEARS
3.1.1
ABSTRACT
66
RECOMMENDATIONS FOR PREVENTING CARDIOVASCULAR
67
DISEASE 3.1.2
WOMEN‘S MEASURED AND PERCEIVED AEROBIC FITNESS
67
3.1.3
THE VALUE OF ASSESSING AEROBIC FITNESS AND PHYSICAL
68
ACTIVITY 3.1.4
THE RISK OF OVERWEIGHT AND OBESITY
69
3.1.5
AIMS OF STUDY TWO
70
3.2.1
INTRODUCTION TO THE REVIEW OF LITERATURE
71
3.2.2
WOMEN‘S PERCEPTIONS OF THEIR AEROBIC FITNESS
71
3.2.3
AEROBIC FITNESS AND RISK OF CARDIOVASCULAR DISEASE
72
3.2.4
CLINICAL ASSESSMENT OF AEROBIC FITNESS
78
3.2.5
OVERWEIGHT AND OBESITY: DETERMINING THE RISK OF
79
CARDIOVASCULAR DISEASE 3.2.6
BODY MASS INDEX, PHYSICAL ACTIVITY AND RISK OF
81
CARDIOVASCULAR DISEASE 3.2.7
SUMMARY
82
3.3.1
STUDY PROCEDURES
83
3.3.2
ETHICAL CONSENT AND UNDERPINNING RATIONALE
83
3.3.3
ELIGIBILITY CRITERIA
84
3.3.4
RECRUITMENT OF PARTICIPANTS
84
3.3.5
INFORMED CONSENT
85
3.3.6
ANONYMITY AND CONFIDENTIALITY
85
3.3.7
THE TESTING VENUE
86
3.3.8
PRE-EXERCISE SCREENING
86
3.3.9
MEASURING AEROBIC FITNESS
87
3.3.10 SUBMAXIMAL EXERCISE TESTING
87
3.3.11 RATIONALE FOR THE USE OF THE CHESTER STEP TEST
88
3.3.12 VALIDITY AND RELIABILITY OF THE CHESTER STEP TEST
89
3.3.13 EQUIPMENT, PRE-TEST CONDITIONS AND PROCEDURE FOR
89
COMPLETION OF GRAPHICAL DATA SHEET
vii
3.3.14 PERCEPTIONS OF FITNESS
90
3.3.15 ANALYSIS OF MEASURED AND PERCEIVED SCORES OF AEROBIC
91
FITNESS 3.3.16 PHYSICAL ACTIVITY LEVELS
94
3.3.17 AEROBIC FITNESS SCORES
94
3.3.18 OVERWEIGHT AND OBESITY STATUS
94
3.3.19 BODY MASS INDEX AND PHYSICL ACTIVITY STATUS
95
3.3.20 PRESENTATION OF THE RESULTS AND DISCUSSION
95
3.4.1
CHARACTERISTICS OF THE PARTICIPANTS
95
3.4.2
WOMEN‘S PERCEPTIONS OF THEIR AEROBIC FITNESS
96
3.4.3
COMPARING MEASURED AND PERCEIVED AEROBIC FITNESS
96
SCORES 3.4.4
AEROBIC FITNESS AND CARDIOVASCULAR DISEASE RISK
99
3.4.5
PHYSICAL ACTIVITY LEVELS
102
3.4.6
BODY MASS INDEX AND RISK OF CARDIOVASCULAR DISEASE
103
3.4.7
ABDOMINAL ADIPOSITY AND RISK OF CARDIOVASCULAR DISEASE
104
3.4.8
COMBINED BODY MASS INDEX AND PHYSICAL ACTIVITY LEVEL
105
3.5.1
CONCLUSION TO STUDY TWO
106
3.5.2
WOMEN‘S ABILITY TO ESTIMATE THEIR AEROBIC FITNESSS
106
3.5.3
WOMEN‘S AEROBIC FITNESS AND PHYSICAL ACTIVITY
106
3.5.4
MEASURED OVERWEIGHT AND OBESITY
107
3.5.5
COMMUNICATING RISK
107
CHAPTER IV: STUDY THREE PERCEPTIONS OF EXERCISE BENEFITS AND BARRIERS IN WOMEN AGED BETWEEN 25 AND 79 YEARS
4.1.1
ABSTRACT
109
THE DETERMINANTS OF WOMEN‘S PHYSICAL ACTIVITY
110
BEHAVIOUR 4.1.2
THEORIES AND MODELS OF HEALTH BEHAVIOUR
111
4.1.3
A NEED FOR FURTHER RESEARCH
112
4.1.4
AIMS OF STUDY THREE
113
4.2.1
INTRODUCTION TO THE REVIEW OF LITERATURE
114
4.2.2
FRAMING PHYSICAL ACTIVITY BEHAVIOUR WITHIN A SOCIAL
114
viii
MARKETING PARADIGM 4.2.3
CONTEXTUALISING PERCEIVED EXERCISE BENEFITS AND
115
BARRIERS WITHIN BEHAVIOURAL THEORIES 4.2.4
THE HEALTH BELIEF MODEL
116
4.2.5
THE TRANSTHEORETICAL MODEL
116
4.2.6
SOCIAL COGNITIVE THEORY
119
4.2.7
WOMEN‘S PERCEIVED AND REAL EXERCISE BENEFITS AND
120
BARRIERS 4.2.8
YOUNG ADULTS PERCEPTIONS OF EXERCISE BENEFITS AND
120
BARRIERS 4.2.9
OLDER PEOPLE‘S PERCEPTIONS OF EXERCISE BENEFITS AND
122
BARRIERS 4.2.10 OBESE WOMEN‘S PERCEPTIONS OF EXERCISE BENEFITS AND
123
BARRIERS 4.2.11 SUMMARY
125
4.3.1
METHODOLOGY: THE STUDY PROCEDURES
126
4.3.2
THE PARTICIPANTS
126
4.3.3
ETHICS
127
4.3.4
INFORMED CONSENT
127
4.3.5
ANONYMITY AND CONFIDENTIALITY
128
4.3.6
VENUES FOR DATA COLLECTION
128
4.3.7
THE EXERCISE BENEFITS AND BARRIERS SCALE
128
4.3.8
SCORING THE EXERCISE BENEFITS AND BARRIERS SCALE
129
4.3.9
DATA ANALYSIS
130
4.3.10 AIM ONE
130
4.3.11 AIM TWO
131
4.3.12 AIM THREE
131
4.3.13 JUSTIFICATION FOR THE STATISTICAL ANALYSIS
132
4.3.14 BODY MASS INDEX
133
4.3.15 PHYSICAL ACTIVITY STATUS
133
4.4.1
PRESENTATION OF THE RESULTS AND DISCUSSION
134
4.4.2
CHARACTERISTICS OF WOMEN AGED BETWEEN 25 AND 79
134
YEARS 4.4.3
CHARACTERISTICS OF WOMEN AGED BETWEEN 25 AND 65 YEARS
ix
134
4.4.4
TOTAL MEAN SORES FOR PERCEIVED EXERCISE BENEFITS AND
135
BARRIERS IN WOMEN AGED BETWEEN 25 AND 65 YEARS 4.4.5
PERCEIVED EXERCISE BENEFITS OF WOMEN AGED BETWEEN 25
136
AND 65 YEARS 4.4.6
PERCEIVED EXERCISE BARRIERS OF WOMEN AGED BETWEEN 25
137
AND 65 YEARS 4.4.7
PERCEIVED EXERCISE BENEFITS OF WOMEN AGED BETWEEN 25
139
AND 45 YEARS 4.4.8
PERCEIVED EXERCISE BARRIERS OF WOMEN AGED BETWEEN 25
140
AND 45 YEARS 4.4.9
PERCEIVED EXERCISE BENEFITS OF WOMEN AGED BETWEEN 46
141
AND 65 YEARS 4.4.10 DIFFERENCES IN PERCEIVED EXERCISE BENEFITS AND
143
BARRIERS IN WOMEN AGED BETWEEN 25 AND 45 YEARS AND WOMEN AGED BETWEEN 46 AND 65 YEARS 4.4.11 CHARACTERISTICS OF WOMEN AGED BETWEEN 66 AND 79
144
YEARS 4.4.12 PERCEIVED EXERCISE BENEFITS AND BARRIERS IN WOMEN
144
AGED BETWEEN 66 AND 79 YEARS 4.4.13 PERCEIVED EXERCISE BENEFITS OF WOMEN AGED BETWEEN 66
145
AND 79 YEARS 4.4.14 PERCEIVED EXERCISE BARRIERS IN WOMEN AGED BETWEEN 66
147
AND 79 YEARS. 4.5.1
ISSUES PERTAINING TO PERCEIVED EXERCISE BENEFITS AND
149
BARRIERS IN WOMEN AGED BETWEEN 25 AND 79 YEARS 4.5.2
KEY PERCEPTIONS OF BENEFITS AND BARRIERS
149
4.5.3
OLDER WOMEN AND SOCIAL INTERACTION
151
4.5.4
HEALTH BENEFITS
151
4.5.5
CONCLUSION: A BASIS FOR DISCUSSION
152
CHAPTER V: CONCLUSION TO THE THESIS
5.1
INVESTIGATIVE AIMS OF THE THESIS
153
5.2
CONCLUSIONS DRAWN FROM THE RESEARCH FINDINGS
156
5.3
GENERAL OBSERVATIONS
156
5.4
INTERVENTION AT AN INDIVIDUAL LEVEL
156
x
5.5
A ROLE FOR FITNESS INSTRUCTORS
158
5.6
DISSEMINATING INFORMATION TO THE TARGET AUDIENCE
159
5.7
PERCEIVED EXERCISE BENEFITS AND BARRIERS
161
5.8
OVERWEIGHT AND OBESITY
162
5.9
SUMMARY
163
5.10
GENERAL LIMITATIONS
164
5.11
RECOMMENDATIONS FOR FUTURE RESEARCH
165
REFERENCES
167
APPENDICES
181
xi
LIST OF TABLES CHAPTER II: STUDY ONE TABLES Table 1.
Characteristics of the pilot study sample population.
40
Table 2.
Age breakdown of women aged between 25 and 65 years.
43
Table 3.
Marital status of women aged between 25 and 65 years.
43
Table 4.
Employment status of women aged between 25 and 65 years.
44
Table 5.
Percentage of women at risk of long-term illness in women aged between 25 and 65 yrs (n = 699) according to NICE guidelines. (2006).
48
CHAPTER III: STUDY TWO
TABLES Table 6.
Studies of aerobic fitness and risk of cardiovascular disease.
74
Table 7.
Aerobic fitness in METs according to age categories.
76
Table 8.
Characteristics of the participants.
96
Table 9.
Categorization according to norms of the Chester Step test.
100
Table 10.
Percentage of women at risk according to aerobic fitness thresholds.
101
Table 11.
MET values for low, intermediate and high aerobic fitness.
101
Table 12.
Women‘s physical activity status categorized as fully active, partially active and sedentary.
102
Table 13.
Body mass index categories.
103
Table 14.
Risk of cardiovascular disease or other long term health problems based on combined body mass index and waist circumference.
104
Table 15.
Body mass index and physical activity status.
105
xii
CHAPTER IV: STUDY THREE
TABLES Table 16.
Breakdown of venues and number of participants.
126
Table 17.
Total scores for the Exercise Benefits and Barriers Scale for women aged between 25 and 65 years.
135
Table 18.
Mean, standard deviation and t-values for benefits and barriers standardised scores for women aged between 25 and 65 years.
135
Table 19.
Ranked order of perceived exercise benefits of women aged between 25 and 65 years.
136
Table 20.
Ranked order of perceived exercise barriers of women aged between 25 and 65 years.
137
Table 21.
Ranked order of perceived exercise benefits of women aged between 25 and 45 years.
139
Table 22.
Ranked order of perceived exercise barriers of women aged between 25 and 45 years.
140
Table 23.
Ranked order of perceived exercise benefits in women aged between 46 and 65 years.
141
Table 24.
Ranked order of perceived exercise barriers of women aged between 46 and 65 years.
142
Table 25.
Means, standard deviation and t-values for benefits and barriers variables as a function of age.
143
Table 26.
Mean, standard deviation and t-values for benefits and barriers standardised scores for women aged between 66 and 79 years.
145
Table 27.
Ranked order of perceived exercise benefits of women aged between 66 and 79 years
146
Table 28.
Ranked order of perceived exercise barriers of women aged between 66 and 79 years
148
CHAPTER V CONCLUSION TO THE THESIS TABLES
Table 29.
Investigative aims of the thesis
153
Table 30.
Main findings of three studies
154
xiii
LIST OF FIGURES
CHAPTER II: STUDY ONE Fig 1.
Body mass index of women aged between 25 and 65 years.
45
Fig 2.
Body mass index status of women in g1 and g2.
45
Fig 3.
Abdominal adiposity in women aged between 25 and 45 years and between 46 and 65 years.
47
Fig 4.
Participants who had accessed screening for hypertension and high cholesterol.
49
Fig 5.
Women‘s preferred sources of information about cardiovascular disease.
52
Fig 6.
Participants‘ perceptions of how well they were informed about cardiovascular disease.
53
Fig 7.
Women‘s perceptions of the causes of cardiovascular disease.
55
Fig 8.
Women‘s perceptions of their cardiovascular disease risk.
59
CHAPTER III: STUDY TWO
Fig 9.
Bland and Altman plot showing the comparison of agreement between the n = 58 participants Mean and Perceived scores.
97
Fig 10.
Normality of errors or residuals.
98
xiv
ACRONYMS AHA
American Heart Association
ACSM
American College of Sports Medicine
BACPR
British Association of Cardiovascular Prevention and Rehabilitation
BHF
British Heart Foundation
BMI
Body Mass Index
BBPA
British Blood Pressure Association
BHS
British Hypertension Society
CHD
Coronary Heart Disease
CMO
Chief Medical Officer
CST
Chester Step Test
CVD
Cardiovascular Disease
DoH
Department of Health
EBBS
Exercise Benefits and Barriers Scale
ESC
European Society of Cardiology
EU
European Union
GP
General Practitioner
HBM
Health Belief Model
HR
Heart Rate
IDF
International Diabetic Federation
MET
Metabolic Equivalent
NAO
National Audit Office
NHS
National Health Service
NICE
National Institute for Health and Clinical Excellence
NOF
National Obesity Forum
SCT
Social Cognitive Theory
xv
TC
Total Cholesterol
TM
Transtheoretical Model
UK
United Kingdom
WAG
Welsh Assembly Government
WC
Waist Circumference
WHO
World Health Organisation
WHR
Waist To Hip Ratio
WSFF
Women‘s Sport and Fitness Foundation
xvi
ABSTRACT
The policies of the Welsh Assembly Government (WAG) regarding the health of the Welsh population are underpinned by a social marketing approach. This is where the individual is supported in their efforts to take personal responsibility for their health. Atherogenic cardiovascular disease (CVD) is a prevalent health problem for women who can take preventative steps through attention to major modifiable risk factors for the disease. There is a paucity of information about the experience of preventing CVD from the perspective of Welsh women and this was the stimulus for this project involving a profiling of apparently healthy women aged between 25 and 79 years living in the Vale of Glamorgan. The thesis consisted of three exploratory studies the first of which utilised a 27-item questionnaire developed by the primary investigator to ascertain knowledge of CVD, preferred sources of health information, physical activity levels, prevalence of overweight and obesity, smoking status, perceptions of risk, and health screening behaviours of 724 women aged between 25 and 65 years. The second study of women aged between 25 and 65 years (n = 58) utilised a Chester Step test to measure women‘s aerobic fitness facilitating comparison with thresholds of CVD risk identified in the literature, and comparison of selfperceptions of aerobic fitness with measured aerobic fitness. Additionally, measured bodyweight and stature enabled estimation of body mass index and this together with measured waist circumference allowed comparison with risk thresholds identified in the literature. The third study investigated and compared perceptions of exercise benefits and barriers of women aged between 25 and 79 years (n = 128) utilising an Exercise Benefits and Barriers Scale (Sechrist et al., 1987). Participants for the three studies were obtained by convenience sampling and this took place in various localities where women were known to meet for leisure or employment. The results of study one highlighted gaps in women‘s knowledge base of CVD, a concerning prevalence of overweight and obesity, low levels of physical activity, poor uptake of screening for cholesterol and blood pressure, and misperceptions of personal CVD risk. Preferred sources of information about CVD were magazines and television and only 10.0% of women reported discussing CVD prevention with their General Practitioner. The results of study two suggested that women were moderately accurate regarding self-perception of their aerobic fitness and 75.8% reported partially meeting recommendations for physical activity. Fifty-seven percent were overweight or obese and 50.0% were over thresholds advised by the National Institute for Health and Clinical Excellence (2006) for risk of CVD and other long-term illness. Nineteen percent had an increased risk of all-cause mortality and cardiovascular events based on failure to achieve an aerobic threshold of 7.9 METs suggested by Kodama et al. (2009). The results of study three demonstrated that women in the sample perceived more exercise benefits than barriers. Physical performance and psychological outcome benefits were the most agreed with exercise benefits in women aged between 25 and 65 years but in women aged between 66 and 79 years social interaction was the most important. Preventative health benefits were not ranked highly across the age spectrum. The most agreed with barrier in women aged between 25 and 79 years was that of exercise as tiring or fatiguing, and restrictions caused by time and family responsibilities also ranked highly. The overall results indicated that interventions to increase exercise participation in women must account for possible negative perceptions of exercise as tiring and fatiguing and efforts to increase participation should focus on enjoyment. Furthermore, account should be taken of the time restrictions faced by women, and exercise opportunities should be easily accessible and convenient. The prevention of CVD entails attention to major modifiable risk factors. The overall results of this thesis suggested that women might more readily take responsibility for CVD prevention if gaps in their knowledge base were addressed, they were supported in efforts at maintaining
xvii
a healthy body-weight, were engaged in talking with health professionals about prevention, and if more attention was paid to their perceptions of exercise benefits and barriers. Further research could capitalise on these findings adding to what is known about women and the prevention of CVD, particularly with regard to interventions to increase physical activity and for the management of body-weight.
xviii
1.1 WOMEN AND CARDIOVASCULAR DISEASE
Cardiovascular disease (CVD) is a collective term for diseases of the heart and circulatory system inclusive of ischemic heart disease, cardiomyopathy, valvular disease, heart failure, arrhythmias, peripheral artery disease, and stroke (British Association of Cardiovascular Prevention and Rehabilitation [BACPR], 2006). Whilst some CVD is an inevitable consequence of familial disposition, the risk of CVD can also be exacerbated by some forms of illness such as diabetes or myocarditis (Fagan, 2005). The emphasis of the present thesis was on atherogenic CVD where modifiable risk factors for the disease predispose to injury of the innermost lining of arteries and contribute to atheroma development (BACPR, 2006). In atherogenic CVD there is potential for the individual to ameliorate factors associated with an increased risk of developing the disease (BACPR, 2006).
The personal motivation for this thesis arose from the experience of the primary investigator working with women for over 20 years in the fitness industry. It was clear that many women were ill informed about CVD. Moreover, many were overweight or obese and were not sufficiently active to benefit their cardiovascular health. Obesity and inactivity are associated with the development of CVD, which causes approximately 22.0% of female deaths in the UK before the age of 75 years (British Heart Foundation [BHF], 2008). In 2007 in Wales, 6,000 women died from coronary heart disease (CHD) and circulatory disease; almost 2,000 more than cancer BHF (2010). Moreover, apart from premature mortality, CVD can cause disability and poor quality of life because it is associated with physical, emotional and economic suffering (BHF, 2010). 1.2 ATHEROGENIC CARDIOVASCULAR DISEASE
Atherogenic CVD is caused by a combination of risk factors exerting an adverse influence on the cardiovascular system over time and, that relate strongly to lifestyle and modifiable physiological factors (De Backer et al., 2003). The risk of atherosclerotic CVD results from exposure to genetic, biological, behavioural, and environmental ‗risk factors‘ (Gordon et al., 2006). In a seminal paper by Ross (1999; p116), the process of atherosclerosis was described as a disease of endothelial dysfunction: ―resulting from elevated and modified low density lipoprotein, cigarette smoking, hypertension, diabetes mellitus, genetic alterations; elevated plasma homocysteine concentrations; infectious micro organisms, and combinations of these or other factors‖.
1
Exposure to vascular irritants results in injury to endothelial cells and initiates a cascade of inflammatory mechanisms resulting in changes to normal homeostatic function (Ross, 1999). The injured endothelium provides a surface for the adherence of blood platelets and monocytes, initiating a process of plaque formation, and this has the effect of narrowing the vessel lumen and restricting blood flow (Gordon et al., 2006).
Atherosclerosis is not
restricted to arteries of the heart but can manifest in other parts of the circulatory system. Approximately 50.0% occurs in the form of CHD, although hypertension, stroke, and peripheral artery disease are also manifestations of atherogenic disease (BHF, 2008).
For clinical integrity, a risk factor for CVD must satisfy certain criteria. These are a consistent strength of association resulting from the findings of multiple prevention studies, the risk factor precedes the disease over time, and there is a slope where higher levels of the risk factor are associated with a higher possibility of disease (Gordon et al., 2006). Within the literature, risk factors are categorised as ‗modifiable‘ and ‗non-modifiable‘ and as ‗traditional‘ or ‗classical‘, ‗novel‘ or ‗emerging‘, this latter terminology reflecting the current evolving body of knowledge. The traditional risk factors amenable to modification consist of smoking cigarettes, physical activity, hypertension, plasma lipids, diabetes, glucose levels, and obesity (Ross, 1999; Stangl et al., 2002; De Backer et al., 2003; BHF, 2010). In adition to modifiable risk factors there are non-modifiable factors that exert a powerful influence on subsequent development of disease and these are age, heredity and ethnicity (Thompson et al., 2002). Some sources broaden the list for example, the American Heart Association (AHA, 2010) additionally include stress, contraceptive medication, alcohol, and illegal substances. Although the same risk factors are associated with the development of CVD in men and women, it is the significance and relative weighting of these factors that are different in women, for example, as seen in changes to the lipid profile occurring at menopause (Stangl et al., 2002). Because the average lifetime risk of CVD for women is very high, preventing CVD is important for all women (Mosca et al., 2007). Therefore, the management of risk factors through a lifestyle conducive to lowering risk, or where necessary, through pharmacology is crucial to prevention.
It is accepted that atherosclerosis generally results from exposure to a combination of risk factors and not a single cause (Gordon et al., 2006). This combination exerts an adverse influence on the cardiovascular system over time (De Backer et al., 2003). In order to mimimise the level of personal risk it could be safely assumed that a cardiovascular health enhancing lifestyle would be one where the individual avoided smoking, was habitually physically active, maintained body-weight within healthy parameters and took steps through screening to ensure cholesterol and blood pressure were monitored. In which case, most 2
women could be pro-active in preventing CVD if they were motivated to do so, if their personal circumstances allowed, and if they were knowledgeable about prevention. Unfortunately, the evidence would indicate that many Welsh women do not adhere to health enhancing behaviours and might therefore be at increased risk. For example, successive Welsh Health Surveys have reported that many of the female population are overweight or obese (53.0%), more than 50.0% are insufficiently active, and the prevalence of smoking cigarettes is approximately 21.0% in women over the age of 16 years. It is also unfortunate that women are less likely to identify their risk factors for CVD compared to men, and to participate in screening programmes (Stramba-Badiale et al., 2009).
Women differ to men in that risk of CVD is heightened with the advent of menopause with its associated metabolic changes (Stangl et al., 2002). Prior to menopause the potential risk factors of total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) are usually lower in women than men although high-density lipoprotein cholesterol (HDL-C) is higher in women. High-density lipoprotein cholesterol fractions and LDL-C fractions are more predictive of elevated risk (Stangl et al., 2002). Raised plasma triglyceride (TG) levels also contribute to the atherosclerotic process, and risk associated with raised TG is further elevated when occurring in combination with elevated LDL-C and, or lowered HDL-C levels (D‘Agostino et al., 2000). The risk associated with Lipoprotein (a) is also postulated to play a role in atherogenesis but there is conflicting evidence and because it cannot be modified by changes to lifestyle behaviour, and when LDL-C is lowered to target levels, Lp (a) loses its atherogenic effect (Maher et al., 1995). A further factor is Homocystein which is considered as a non-lipid risk associated with endothelial dysfunction, enhanced platelet aggregation and vascular smooth muscle proliferation, possibly predisposing to CHD, peripheral artery disease, stroke and venous thromboembolism (Boushey et al., 1995). However the clinical usefulness of measuring homocysteine as an indicator of risk is uncertain (Gordon et al., 2006).
Hypertension is an asymptomatic risk factor for CVD characterised by consistently elevated blood pressure resulting in an increased workload for the heart, ultimately increasing the risk of vascular endothelial dysfunction (BACPR, 2006). The myocardial and vascular risk from hypertension rises consistently with elevated blood pressure levels and acts idependently of other risk factors (Gordon et al., 2006). It is of concern that hypertension is rarely an isolated phenomena and is frequently accompanied by other risk factors such as overweight and obesity or lipid abnormalities that in combination significantly increase level of CVD risk (Stangl et al., 2002).
3
The presence of diabetes exerts a profound influence on risk particularly for CHD and is associated with more severe consequences for women than for men (Lee et al., 2000; Hu et al., 2001). Obesity pre-disposes to diabetes type 2 and is defined as a body mass index of 30 kg m2 or more. Abdominal adiposity confers a particularly elevated risk when in conjunction with the presence of other risk factors. ‗Metabolic syndrome‘ is the term given to a clustering of these risk factors, the core components consisting of central obesity, insulin resistance, dyslipidemia and hypertension (De Backer et al., 2003).
With the acceptance of atherosclerotic CVD as an inflammatory process, markers of heightened inflammation have been considered for their potential predictive value of future cardiovascular events. High sensitivity C-Reactive Protein (hs CRP) is a systemic inflammatory marker that has been investigated to determine its role in atherogenesis (Kuller et al., 1996; Ridker et al., 1998). Whilst other inflammatory markers have also been investigated, for example, serum amyloid A, IL-6, and sICAM-1, according to Ridker et al. (2000) hs CRP offers the best predictive value. The role of coagulation factors in heighteneing cardiovascular risk has also been demonstrated such that fibrinogen is accepted to play a role in atherothrombosis and is implicated in an inflammatory process of the vascular wall. However, although fibrinogen is accepted as a risk factor for CHD, in current clinical practice it is not always assessed as part of routine risk factor assessment (BACPR, 2006).
1.3 PHYSICAL ACTIVITY AND CARDIOVASCULAR DISEASE
Physical inactivity is an independent risk factor for CVD (BACPR, 2006). Conversely, because habitual physical activity has multiple protective effects on the cardiovascular system, primary preventative cardiovascular health care includes physical activity in order to potentially obviate the need for future interventional cardiology. Habitual aerobic physical activity of sufficient duration and intensity stimulates the cardiovascular system. This results in enhanced structural and functional efficiency, and the underlying mechanisms of adaptations provides a rationale for the outcome of aerobic fitness and the consequent protection from CVD. With sufficient stimulus during training, over a period of time a number of structural, functional and metabolic adaptations occur in skeletal muscle, the heart and the circulation that impact positively on CVD risk (BACPR, 2006). Central adaptations consist of an increased maximum cardiac output, resulting from an increase in maximal stroke volume, an increase in left ventricular mass and chamber size, an increase in total blood volume and a reduction in total peripheral resistance. Peripheral changes include an improvement in capilliarisation around trained muscle, an increased myoglobin concentration, an increased 4
number and size of mitochondria together with increased oxidative enzyme activity within mitochondria (McArdle et al., 2007). These training induced adaptations result in an increase in fitness associated with a more favourable cardiovascular profile with a decreased risk of premature CVD mortality and morbidity (Blair et al., 1989; Blair et al., 1996; Gulati et al., 2003; Kodama et al., 2009).
The physiological adaptations associated with habitual physical activity positivley impact on risk factors for CVD so that blood pressure, TC and LDL-C are reduced, HDL-C is increased, glucose metabolism is improved, there is increased sensitivity to insulin and a favourable influence on obesity (Thompson et al., 2003). Additionally, Feuerstein et al. (2003) point out that habitual physical activity influences inflammatory pathways in atherogenesis and that in more active individuals physical activity is associated with lower hs CRP levels compared to sedentary controls. Bassuk and Manson (2003) make the powerful case that physical activity slows the initiation and progression of atherosclerosis via favourable effects on the risk factors of body-weight, blood pressure, insulin sensitivity, glycemic control, the lipid profile, fibrinolysis, endothelial function and inflammation.
The role of physical activity in the prevention of CVD remains however an emerging field of knowledge as identified by Bowles and Laughlin (2011), pointing out that the mechanisms underlying the effectiveness of physical activity in preventon and treatment have not been completely established. Bowles and Laughlin (2001) explain that whilst the independence of physical inactivity as a risk factor is supported in the literature, only 35.0% of the beneficial effects of physical activity can be attributed to favorable changes in known risk factors and therefore 65.0% of the established beneficial effect remains to be confirmed. They propose several additional mechansisms through which physical activity is thought to be protective and identify current avenues of research. These include the direct effects of physical activity on the vascular wall and the contribution of nonvascular adaptations, such as improved ischemic tolerance of the myocardium and reduced thrombogenecity, the role of smooth muscle in the initiation and progression of atherosclerotic lesions, and the role of traininginduced collateralization, alterations in thrombogenecity, and activity induced adaptations to the myocardium. These avenues of
current
research will undoubtedly enhance
understanding of the mechanisms through which habitual physical activity exerts a protective effect on the cardiovascular system and strengthen the case for an active lifestyle.
5
1.4 A RATIONALE FOR PREVENTING CARDIOVASCULAR DISEASE In 2007, the European Society of Cardiology ([ESC], Graham et al., 2007), issued a rationale for the prevention of CVD based on the reasoning that it is a major cause of death and disability in Europe. Furthermore, that CVD advances asymptomatically, and is well advanced before diagnosis, death often occurs suddenly, and the mass of CVD relates strongly to lifestyle and modifiable risk factors, although modification of risk factors can reduce mortality and morbidity.
From a medical perspective, the objective of CVD prevention is to reduce the incidence of first or recurrent clinical events and the focus is on preventing disability and premature death (De Backer et al., 2003). However, in general terms research about prevention in women has lagged behind that focussing on men because historically there was a misperception that CVD was more likely to affect men than women (Barrett-Connor, 1995; Stangl et al., 2002). In 1997, Barrett-Connor drew attention to the need for increased recognition of CVD as a health problem for women based on a re-evaluation of statistical information regarding morbidity and mortality. This demonstrated that CVD was also a health problem for women, but one experienced at a later age. Moreover, risk of CVD was underestimated in women because of a widespread belief that they were protected by their hormonal status, and misperceptions were only challenged when the results of large epidemiological studies became available highlighting the issue of CVD in women (Mosca et al., 2007). Consequently, there was a drive to heighten knowledge and awareness of CVD held by health care providers and the public. For example, the ‗Go Red for Women‘ campaign (AHA, 2003), and an initiative of the ESC called ‗Women at Heart‘ (2005) signalled the recognition of CVD as a significant health issue for women. In 2009, the European Heart Network and the ESC drew attention to the need for research specifically focussing on women in a publication titled ‗Red Alert on Women‘s Hearts‘ (Stramba-Badiale et al., 2009). Most atherogenic CVD is preventable (Mosca et al., 2007). However, who is responsible for preventing it is open to debate. Whilst there is sound argument for individuals taking responsibility for their health, not all will be equally empowered to do so. Some individuals will be disadvantaged by an adverse hereditary profile, or by acquired chronic illness that limits ability to lead a healthy lifestyle, which although theoretically is health preserving, can be more challenging for some than others. This was recognised by De Backer et al. (2003), and pertained to people who are socially and economically disadvantaged, or who cannot exert much control in their employment, who have a stressful family life, or who live alone without social support. Moreover, disadvantage might specifically refer to women who,
6
because of past misperceptions of CVD as a ‗male disease‘, might lack knowledge and awareness crucial to taking preventative action.
1.5 A SOCIAL MARKETING APPROACH TO HEALTH Taking personal responsibility for health is a theme consistent with objectives of the Welsh Assembly Government (WAG) as made public in ‗Well Being in Wales‘ (WAG, 2001), ‗Health Challenge Wales‘ including, ‗Change4Life‘ (WAG, 2004), ‗Creating an Active Wales‘ (WAG, 2009) and the ‗National Obesity Pathway‘ (WAG, 2010). These policies and campaigns demonstrated an orientation to a social marketing approach whereby business marketing principles and techniques were applied to health-related issues. Moreover, the underpinning social marketing approach reflected the views expressed in a previous publication by the Department for Health titled ‗Securing good health for the whole population‘ in which it was suggested that individuals should be encouraged to take more responsibility for their own health because future health care provision would be economically unsustainable (Wanless, 2004).
Maibach et al. (2002) defined social marketing as a process consisting of efforts to create voluntary exchange between a marketing organisation and members of a target market, or ‗audience‘ based on mutual fulfilment of self-interest. Within the health arena, the marketing organisation would refer to those individuals concerned with promoting good health and with treatment of ill health, and the target audience would refer to individuals, communities and populations. The aim would be to engage all members of the population in the quest for good health, with the individual in the ‗driving seat‘. Therefore, a health oriented social marketing approach identifies and profiles a target audience with the intention of developing interventions to facilitate the audience reaching an identified goal (US Department of Health and Human Services, 1999). It would involve research to generate clear understanding with regard to identified behaviours and attributes of the target audience in order to understand their perceived needs, perceived benefits and barriers, and to develop opportunities to intervene (Maibach et al., 2002). For example, this might be to increase physical activity, to increase the uptake of screening for hypertension, or to facilitate body-weight management. The terms ‗targeted‘ and ‗tailored‘ appear frequently in the health intervention literature, sometimes interchangeably. However, Marcus (2000) differentiated between them explaining that a targeted approach is one where a particular sub-group of the population is defined, based on one or more demographic characteristics, and there is an assumption that this audience is similar enough to communicate a health message. A tailored approach is more 7
specific and health messages are individualised. Following identification of the audience and confirmation of their characteristics, the intervention would consist of a ‗technique‘ or ‗approach‘ for supporting voluntary behaviour change (King et al., 2006). For example, the use of cognitive behavioural theory to underpin smoking cessation or physical activity programmes.
An application of social marketing to the prevention of CVD would necessitate profiling the identified audience in order to describe their behaviours and relevant characteristics. For example, physical activity levels, or overweight and obesity status might give cause for concern. In relation to the prevention of CVD in Welsh women, there is a paucity of information about what they know about CVD, their sources of health information or their personal perceptions of risk. Furthermore, although Welsh Health Surveys continue to report women‘s low physical activity levels and concerning levels of obesity, little is known about Welsh women‘s perceived and measured aerobic fitness, or their perceived exercise benefits and barriers. However, these are factors for consideration in the development of a profile focussing on CVD prevention. Therefore, in light of the current interest in women‘s experience of CVD and the identified need for research from the perspective of women, the objective of the present thesis was to develop a prevention-focussed profile that would provide a timely discussion platform for health care practitioners and fitness instructors regarding effective, targeted intervention programmes.
1.6 AIMS OF THE THESIS The thesis adopted an investigative approach consisting of three studies. The voluntary participants were drawn from the local community of the Vale of Glamorgan and were asymptomatic women aged between 25 and 65 years (studies one and two), and in study three, women were aged between 25 and 79 years. The first study (chapter two) consisted of a cross-sectional survey with the aim of investigating women‘s knowledge and awareness of CVD, preferred sources of health information, self-reported physical activity levels, overweight and obesity, smoking status, and health screening behaviours (n = 724). The age range of 25 to 65 years was commensurate with age categories of adulthood and middle age, as described by Spirduso et al. (2005). The aim of the second study (chapter three) was to compare women‘s measured aerobic fitness with their perception of their fitness, to evaluate their aerobic fitness against thresholds for risk of CVD as found in the literature, and to determine the prevalence of overweight and obesity (n = 58). Participants completed a Chester Step test (Sykes, 2005), a seven-day physical activity question, and stature, body-
8
weight and WC were measured by the researcher. The aim of the final study (chapter four) was to investigate women‘s perceptions of exercise benefits and barriers (n = 128). Participants completed an Exercise Benefits and Barriers Scale (Sechrist et al., 1987) a seven-day physical activity recall question, and stature, body-weight and WC were measured by the researcher. The results of the three studies were synthesised to fulfil the overall aim of the thesis, which was to construct a profile in order to contribute to knowledge about women in relation to CVD prevention.
1.7 ORGANISATION OF SUPPORTING LITERATURE AND PREFERRED USE OF TERMINOLOGY The three studies adopted different methodological approaches to fulfil the stated aims. In light of this, the search for relevant literature was broad in the first instance and then specific to the issues pertinent to each study. The scope of the search for relevant literature was narrowed to focus on adult women and key terms included ‗cardiovascular disease‘, ‗physical activity‘, ‗exercise‘, ‗overweight and obesity‘, ‗aerobic fitness‘, ‗perceived fitness‘ and ‗perceived exercise benefits and barriers‘. The primary investigator utilised current literature pertaining to her involvement in professional organisations such as the British Association of Cardiac Prevention and Rehabilitation, and the American College of Sports Medicine. A search for literature then consisted of a hand search of journals and electronic searches, for example using MEDLINE, Google Scholar and the Health of Wales Information Services (HOWIS).
Relevant scientific statements and policy documents provided sources of
literature and these were not limited to the United Kingdom. For example, electronically available literature of the ESC was utilised.
Within the literature, there was some difference in the use of key terms. For example, physiologically oriented academic papers more often use the term ‗body-mass‘. Papers with a more health-oriented perspective often use the term ‗body weight‘. For clarity of expression throughout the thesis, the primary investigator has used terms commonly used in health and fitness literature.
9
ABSTRACT The aim of study one was to investigate women‘s knowledge of cardiovascular disease (CVD), personal perceptions of risk, screening behaviour for blood pressure and cholesterol, physical activity levels, obesity and smoking status, and to elicit preferred sources of information about CVD. The study took the form of a cross-sectional survey with the intention of developing a profile of women aged between 25 and 65 years, and which involved comparison of women aged between 25 and 45 years, with women aged between 46 and 65 years. A 27-item questionnaire developed by the primary investigator was distributed to venues in the locality of the study. One thousand six hundred questionnaires were distributed; 724 questionnaires were returned by post, or collection. The response rate was 56.0%. The results highlighted five issues of concern: (i) a high prevalence of overweight and obesity (46.0%), (ii) concerning levels of physical activity (54.0%), (iii) misperceptions of personal cardiovascular health risk, (iv) poor up-take of screening services for blood pressure (53.0%) and cholesterol (36.0%), and (v) gaps in women‘s knowledge base of CVD. Smoking prevalence was 15.0%. Women had sourced their information about CVD predominantly from the media, including magazines and newspapers (42.0%) and television (25.0%). Few women had sourced their information from talking to a doctor (10.0%), or from reading health-promotion literature (11.2%). There was a tendency to underestimate personal CVD risk in women who were overweight or obese, and in those women classed as inactive. Knowledge of physical inactivity as a contributory cause of CVD was high (94.0%), although the results revealed a discrepancy between women‘s knowledge and their selfreported levels of physical activity. The findings provide a basis for health messages to address gaps in women‘s knowledge base, and confirm a need for targeted intervention regarding physical activity, and overweight and obesity.
10
2.1.1 CARDIOVASCULAR DISEASE: A SIGNIFICANT HEALTH ISSUE FOR WOMEN Cardiovascular disease (CVD) is a significant health issue for women as evidenced by the number of women it affects. In the UK in 2008, there were an estimated 406,000 women over the age of 45 years living with heart failure, 850,000 women over 35 years who had experienced angina, with approximately 43,000 new cases of angina diagnosed in women each year (British Heart Foundation, [BHF] 2008). In addition to personal consequences, CVD has economic implications with an estimated £26 billion annual financial expenditure to the UK economy in direct and indirect costs (BHF, 2008).
2.1.2 DECLINING MORTALITY RATES Declining mortality rates of coronary heart disease (CHD) in men and women living in England and Wales were found by Unal et al. (2004), with approximately 42.0% of the decline attributed to medical and surgical intervention, and 58.0% to a reduction in risk factors, primarily smoking, blood pressure and cholesterol. Despite the decline, Unal et al. (2004) observed adverse trends for physical activity and obesity. There is evidence from successive Welsh Health Surveys (Welsh Assembly Government [WAG], 2004/5; 2007; 2009) that obesity and physical inactivity continue to have a high prevalence within the population of women in Wales and this is concerning because of the strong association with development of type 2 diabetes and CVD (Chief Medical Officer, [CMO] 2004; Bassuk and Manson, 2005).
2.1.3 WOMEN’S KNOWLEDGE OF CARDIOVASCULAR DISEASE The scale of the problem caused by inactivity, and overweight and obesity, in combination with other major modifiable risk factors gives rise to questions about what women know about CVD, and how pro-active they are in preventing it. For example, whether or not they know about healthy measures of cholesterol or blood pressure, whether they take steps to undertake screening and, in light of a high prevalence of overweight and obesity, whether women perceive themselves as at risk of CVD. Women can however access numerous sources of health information about CVD including the media and charitable organisations, for example, from the BHF and Diabetes UK. Additionally, they can access information from a WAG internet site called ‗Health Challenge Wales‘ (2004). This signposts members of the public to health information and includes a publication titled ‗Women and Cardiovascular
11
Disease‘ informing women of the disease and how to prevent it. There is however, a paucity of current information about Welsh women‘s knowledge and awareness of CVD. Although knowledge alone is insufficient to guarantee that women adopt and maintain healthy behaviours, ignorance about CVD might be disempowering and, erroneous perceptions of risk might contradict beliefs or opinions held by health professionals leading to uneasy communication. Therefore, exposure of the strengths and weaknesses of the knowledge held by women is pertinent to the development of cogent cardiovascular health messages. Furthermore, women might have over optimistic beliefs about their personal risk of developing CVD and, over optimism or a lack of knowledge and awareness would be cause for concern in a population identified by the European Society of Cardiology (ESC, 2008), as a high-risk region of the European Union (EU). The ESC have pointed out that women hold misperceptions about CVD and have a lack of awareness, although the results of a study conducted by the American Heart Association involving American women was used as a basis to suggest misperceptions in European women. Whilst it might be safe to assume some similarities between American and European women, the knowledge and awareness of women within regions of the EU might differ in many respects. There is therefore, an opportunity to contribute to what is known about women‘s experience of CVD through research to investigate their knowledge, awareness, and perceptions.
2.1.4 AIM OF STUDY ONE The aim of the present study was to investigate knowledge of CVD, screening behaviour for blood pressure and cholesterol, physical activity levels, perceptions of fitness and risk of CVD, obesity and smoking status, and preferred sources of information about CVD in a sample of women living in the Vale of Glamorgan, UK. The intention was to develop a profile of women aged between 25 and 65 years with a comparison of women aged between 25 and 45 years with women aged between 46 and 65 years. .
12
2.2.1 INTRODUCTION TO THE REVIEW OF LITERATURE The first section of the review discusses women‘s major modifiable risk factors focusing on overweight and obesity, smoking cigarettes, high blood pressure and high cholesterol, and physical inactivity. The effect of aging on the risk of CVD is also included in this section. There is discussion about the role of knowledge in preventing CVD, how health messages diffuse into a community, and of where women might access sources of cardiovascular health information. The chapter concludes by highlighting the issue of priorities in women‘s health screening services and by reviewing studies of women‘s knowledge and perceptions of CVD.
2.2.2 MAJOR RISK FACTORS FOR CARDIOVASCULAR DISEASE Obesity is an accepted major modifiable risk factor for the development of CVD and was defined by the World Health Organisation (WHO, 2004; p6) as ―a condition of abnormal or excessive fat accumulation in adipose tissue, to the extent that health may be impaired‖. According to the National Obesity Forum (NOF, 2009), there are factors that predispose to obesity, for example, genetic influence, environmental barriers, Prader-Willi syndrome or hypothyroidism. From a physiological perspective however, the cause of obesity is an energy imbalance linked to inactivity and an abundance of energy rich food (Mc Ardle et al., 2007). Even if not causing immediate health problems, obesity is a chronic condition pre-disposing to future ill health and associated with metabolic complications including dyslipidemia, high blood pressure, diabetes mellitus and vascular inflammation; each potentially increasing cardiovascular risk (US Obesity Task Force, 1998; WHO 2003; Yusof et al., 2004; Balkau et al., 2007). Obesity increases the risk of developing a number of diseases relative to the nonobese population and in the UK the four most common health problems associated with obesity are high blood pressure, coronary heart disease (CHD), type 2 diabetes, and certain cancers (National Audit Office [NAO], 2010). Based on data from the Health Survey for England (2007), the NAO calculated that overweight women were more likely to have high blood pressure than were women of normal body-weight. Moreover, women with a waist circumference (WC) of 88 cm or more, were more than twice as likely to have high blood pressure compared with women with a WC of less than 88 cm (43.0% to 19.0% respectively). It is a matter of concern that according to the National Heart Forum (2010), the prevalence of overweight and obesity continues to rise in the UK, and it is expected that this in turn, will lead to an increase in the incidence and prevalence of type 2 diabetes and this can increase the risk of CVD. 13
Measures to define overweight and obesity are necessary because they can be used for evaluation against thresholds for increased risk of disease. Although it has some limitations and should be interpreted cautiously, the calculation of body mass index (BMI) is a widely accepted method to define overweight and obesity and is calculated by dividing a person‘s mass in kilograms by the square of their height in metres (Health and Social Care Information Centre, 2010). In adults, a BMI of 25.0 to 29.9 kg m-2 is categorised as ‗overweight‘, and a BMI of 30.0 kg m-2 or more, as ‗obese‘. Guidelines from the National Institute for Health and Clinical Excellence (NICE, 2006) stated that risk of long-term health problems including CVD could be identified using a combination of BMI and waist circumference (WC). An optimum combination would consist of a BMI less than 25.0 kg m-2 and a WC measuring less than 80 cm. Welsh Health Surveys do not however request a measure of waist circumference and there is a paucity of information with regard to this measure in Welsh women. Recognising the cardiovascular risk of abdominal adiposity, a campaign by Diabetes UK (2006) titled ‗Measure Up‘ recommended that women in the UK should have a WC of less than 80 cm (31.5 inches). Furthermore, in recognition of the prevalence of overweight and obesity in the Welsh population, the WAG launched an ‗All Wales Obesity Pathway‘ (WAG, 2009) whereby Health Boards, Local Authorities and other agencies would work together to intervene at an individual, community and population level. The obesity pathway provides guidance to health professionals with regard to the multiple interventions and strategies needed to make an impact on the problem of overweight and obesity and consists of four levels of interventions that are long-term and focused on healthy body-weight, not just obesity.
Cigarette smoking is a major modifiable risk factor for CVD with strong evidence for its harmful effects. The Royal College of Physicians (Britton et al., 2000) reported on nicotine addiction finding that the effects were mediated through sympathetic neural stimulation associated with increased circulating catecholomines. This results in increased heart rate, blood pressure and cardiac contractility, resulting in a restricted blood flow to coronary arteries with an elevated risk of ischaemia. Thirty-three successive reports of the USA Surgeon General have made explicitly clear that smoking is a danger to the health of men, women and children. Two reports have specifically focussed on health consequences of smoking for women, with the report published in 2001 cataloguing the extensive and disastrous effects in terms of the risk of cancer and CVD, and confirming smoking as a major cause of CHD in women.
14
Smoking cigarettes is a major risk factor for the development of hypertension; considered a powerful independent contributor to cardiovascular morbidity and mortality (Kannel and Higgins, 1990). Hypertension is asymptomatic and can proceed to cause injury to the vasculature unless treated. It was pointed out by Stangl et al. (2002) that hypertension as an isolated risk factor is rare and more frequently occurs in the presence of other risk factors, and that the number of risk factors in addition to hypertension imposes a greatly increased risk of CVD in women. It is recommended by the British Blood Pressure Association (BBPA, 2010) that blood pressure is measured every five years, and that all adults should aim for an optimal blood pressure measure given as at, or less than 120/80 mm Hg, with normal blood pressure as at, or less than 130/85 mm Hg. Continuous campaigning by the BBPA aims to raise public awareness of the importance of screening for hypertension because of the potential to prevent stroke, myocardial infarction and heart failure.
Cholesterol screening is important because abnormalities are not apparent unless clinically evaluated and abnormal levels are associated with heightened CVD risk (Stangl et al., 2002). Whilst cholesterol is necessary for healthy functioning, above recommended levels can increase risk through its deposition in the arterial wall contributing to the development of atheroma. Prior to menopause, total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) are usually lower in women compared to men, although high-density lipoprotein cholesterol (HDL-C) is generally higher among pre-menopausal women (Stangl et al., 2002). Following the menopause, TC and LDL-C increase in women particularly between the ages of 55 and 65 years with an increase in cardiovascular risk, although occurring approximately a decade later than in men (Stangl et al., 2002). Clinical assessment by health professionals can determine sub-fractions of cholesterol to aid in decision making regarding any need for treatment.
The BHF continues to mount educational initiatives to inform the public of the desired healthy measure of TC given as less than 5.0 mMol L-1 in apparently healthy individuals. Whilst there is a potential for familial hypercholesterolemia requiring intervention through pharmacology, it is advised by ‗Health Challenge Wales‘ (WAG, 2004) that in general, women can take preventative
measures
through
accessing
screening
through
their
local
General
Practitioner‘s (GP) surgery or at pharmacies. Moreover, they can pay attention to reducing the total amount of fat eaten and participate in habitual physical activity. It is a point of interest that women who deliberately seek out information about reducing the risk of abnormally high cholesterol or hypertension might be highly motivated to do so, or have been recommended to read publications such as those available from ‗Health Challenge Wales‘ (WAG, 2004). However, less motivated women, or women with little or no knowledge about 15
the advantage of screening might be inadequately prepared to implement preventative selfcare actions to minimise risk, such as attention to nutrition and physical activity.
With regard to screening for heightened risk of CVD, a recent innovation of the National Health Service (NHS) was the ‗Health Check Programme‘ (2010) for individuals aged between 40 and 74 years. It has the purpose of ensuring people in this age range are screened to determine their risk of CHD, stroke, kidney disease and diabetes. It aims to help patients reduce or manage their risk by providing individually tailored advice. The health checks consist of questions to patients on their health and diet, exercise habits and family medical history, stature and body-weight measurements, a blood test for cholesterol and for glucose levels, and measurement of blood pressure. A follow up personal assessment sets out the individual's level of risk and recommends what could be done to reduce risk including body-weight management, smoking cessation and physical activity. Unfortunately, at the time of writing it was established by the primary researcher that although this health check had been implemented in the Gwent area of Wales since 2005, no such programme was operating in the locality of the present study1. 1
Habitual aerobic activity of sufficient duration and intensity is associated with a lower risk of CVD. Conversely, a sedentary lifestyle and low levels of aerobic fitness are risk factors for all-cause mortality, CHD, other forms of CVD, and stroke in both men and women (Blair, et al., 1989; Blair et al., 2001; Williams, 2001; Kodama et al., 2009). According to European guidelines on CVD prevention, a lack of physical activity contributes to the early onset and progression of the disease (Graham et al., 2007). Physical activity does not have to be physically punishing in order to accrue health benefits. For example, Duncan et al. (1991) investigated walking as an activity that could promote favourable changes in cholesterol status in 102 sedentary women aged between 20 and 40 years. The women walked 4.8 kilometres on five days a week for 24 weeks. Compared with sedentary controls, walkers showed improvements in maximal oxygen uptake (p < 0.0001) and in their lipoprotein profile. The authors concluded that vigorous physical activity is not necessary for women to obtain meaningful improvements in their lipoprotein profile, and that walking at intensities that do not have a major impact on aerobic fitness might nonetheless produce favourable changes in the cardiovascular risk profile. The Women‘s Health Initiative Observation Study (2002), a large prospective study, compared walking with vigorous exercise and hours spent sitting, as predictors of the 1
Established through personal communication with a representative of the Vale of Glamorgan Health Board
16
incidence of coronary and total cardiovascular events in post-menopausal women aged between 50 and 79 years (n = 73,743). A physical activity score was calculated for each participant based on total self-reported activity, walking, vigorous exercise and hours spent sitting. The results demonstrated a strong, graded, inverse association with the risk of both coronary and cardiovascular events irrespective of race, ethnic group and BMI. The higher the physical activity score, the fewer coronary events and lower total number of cardiovascular events. The age–adjusted relative risks of coronary events in increasing quartiles of activity were 1.00, 0.73, 0.69, 0.68 and 0.47 respectively demonstrating that the most sedentary women had twice the risk of a coronary event compared with the most active. Walking briskly for at least 2.5 hours per week was associated with a 30.0% reduction in cardiovascular events over 3.2 years of follow up. A strength of this study lies in the emphasis on walking as an appropriate mode of activity. With low physical activity levels in Welsh women it is important to consider how they can translate physical activity health messages into their everyday life. Walking is cost effective, offers a certain degree of flexibility in a busy life, and is easily adapted to meet basic fitness principles. Unfortunately, walking as an effective mode of exercise does not feature strongly in commercial attempts to support women in being more active. Women are noticeably targeted in commercial fitness videos and by health clubs but the more simple cost effective value of outdoor walking to meet health guidelines is one that can also address women‘s physical activity needs.
Importantly, the Chief Medical Officer (CMO, 2004) reviewed the evidence for the relationship between physical activity and CVD finding that physical activity is a major independent protective factor against CHD in men and women, and that inactive and unfit people have double the risk of dying from CHD compared to active or fit peers. Furthermore, finding that physical activity significantly reduces the risk of stroke, and positively mediates risk factors of blood pressure, lipid profile and insulin resistance. In guidelines from the American College of Sports Medicine (ACSM) and the American Heart Association (AHA) issued in 2007 it was recommended that physical activity should consist of five sessions a week of 30 minutes or more of moderately intense aerobic activity, or three sessions of 20 minutes of vigorous activity, or a combination of both (Haskell et al., 2007). The CMO for the United Kingdom confirmed this prescription for physical activity (Donaldson, 2009). Whilst there are numerous options for mode of physical activity to meet the recommended prescription, walking in particular is a relatively easy physical activity option for older women unless they are at risk of falls when they would require specialist exercise programming (Sivan et al., 2010).
17
Older age is a non-modifiable risk factor for CVD and the mechanisms through which age exerts its influence for increased risk consist of sedentary living and the normal developmental structural and functional changes that occur with ageing (Lakatta and Levy, 2003). Lakatta and Levy (2003) conceptualised ageing as either ‗successful‘ or ‗unsuccessful‘. Successful ageing implies an absence of limiting subclinical disease, whilst unsuccessful ageing results from an interaction between vascular and cardiac ageing, and disease pathophysiology. Lakatta and Levy (2003) note that unsuccessful ageing is not synonymous with having clinically obvious disease, but a person regarded as having aged unsuccessfully would clinically be rated poorly for measures of cardiovascular structure and function, and considered to be at higher risk of CVD. Priebe (2000) suggested that medical or physical fitness best reflects the quality of cardiovascular biological age, and a measure of this is exercise tolerance. Poor exercise tolerance indicates severity of underlying disease with a low aerobic capacity indicative of preoperative short and long term cardiac risk. According to Blair (2007; p898) ―The precise effect of regular aerobic training on the agerelated decline in aerobic capacity remains unresolved‖. It is clear however, that there is an age related decline in aerobic fitness with increasing age but the decline in fitness occurs nearly twice as fast in sedentary men and women compared with individuals who maintain regular exercise training (Blair, 2007). Of concern is that recent evidence from Welsh Health Surveys (WAG, 2004/2005; 2007; 2009) and the Women‘s Sport and Fitness Foundation (WSFF, 2007) demonstrated that in the UK, decreasing levels of physical activity in women accompany increasing age, potentially placing inactive older women at higher risk of CVD.
2.2.3 SEX DIFFERENCES AND SYMPTOMS OF CARDIOVASCULAR DISEASE: WHY DO THEY MATTER? Recent years have seen an acknowledgement of significant physiological sex differences in cardiovascular structure and function, and as a consequence, in women‘s experience of CVD. According to Blair (2007), sex differences manifest in differing symptoms of disease, co-morbidities and disease outcomes, and an increasing recognition of the impact of biological sex differences might necessitate a paradigm shift in health education messages that have traditionally adopted a gender-neutral approach. Recognition of the need to take account of sex-based differences and their impact on health has been recognised by the ESC (2006), advising that understanding of sex differences might improve clinical management of CVD, and possibly generate gender-specific diagnostic and therapeutic options. An example of sex-based difference is the experience of more atypical symptoms of ischaemia in women, including back pain, a burning sensation in the chest with abdominal discomfort, nausea, and unusual fatigue; moreover, women rather than men frequently 18
experience symptoms in the weeks prior to a myocardial infarction but may be unaware of the significance (Douglas and Ginsburg, 1996). McSweeney et al. (2003) investigated symptoms experienced in 515 women subsequently diagnosed with acute myocardial infarction. Pre-event symptoms were reported by 95.0% of women with the most frequent being unusual fatigue (70.7%); sleep disturbance (47.8%), shortness of breath (42.1%), indigestion (39.4%) and anxiety (35.5%). Seventy-eight percent of women reported experiencing one or more symptoms in the month prior to a myocardial infarction. McSweeney et al. (2003) acknowledged that it was unclear whether prior symptoms predicted future CHD events or whether location, frequency and intensity of symptoms were more important than the number of symptoms; furthermore, whether or at what level, prior symptoms were predictive. The findings indicated that the predominant acute symptom of extreme chest pain in men may not typify the experience of women and that prior symptoms can serve as an alert in the weeks before a cardiovascular event that women should be aware of, and which require urgent medical attention. Of concern were the findings of a prospective study of 3,779 patients in the Euro Heart Survey of Stable Angina evaluating the presentation of, and management of stable angina (Daly et al., 2005). The average age of patients was 61 years, and 42.0% were female. The study revealed marked differences in use of clinical investigations, with evaluation of biochemical risk factors suboptimal. Women were less likely to be referred for medical evaluation, in particular for exercise stress testing, less likely to be referred for revascularisation and less likely to be prescribed preventative therapy on initial assessment. It was therefore unsurprising that the ESC (2006), in a policy statement on CVD in women concluded that there was robust evidence that women with suspected stable angina were under-investigated and under-treated and that there was a need for further research to elicit the reasons why an adverse prognosis was observed in women with these conditions.
A delay in medical investigation of angina and other symptoms obviously disadvantages women at risk of a myocardial infarction. Owen-Smith et al. (2002) investigated five-year mortality in women with angina symptoms known, or not known by their GP to have ischaemic heart disease. Chest pain was common among the studied population, but so was a lack of diagnosis. Owen-Smith et al. (2002) speculated that this was due to nonpresentation of illness by women rather than a failure by GPs to diagnose. This raises questions about why women would not actively seek diagnosis of chest pain, and suggests that they do not recognise its potential seriousness, or in an attempt at self-diagnosis attribute chest pain to other less malign causes. If women‘s knowledge of the symptoms of a myocardial infarction or stroke is poor, they will possibly delay seeking life-saving medical diagnosis and treatment. If this is the case, there is a need for gender-oriented health 19
information. However, a recent past education campaign titled ‗Doubt Kills‘ (BHF, 2006/2007) published in posters and newspapers portrayed a middle-aged man with an obviously painful, tightened belt around his chest. The portrayal was accompanied by a health message to call an ambulance immediately if chest pain was experienced. Whether women in general extrapolated this information as relevant to them is questionable and it is pertinent to consider whether women would benefit from health messaging that is specifically tailored to their needs.
2.2.4 WOMEN’S EXPOSURE TO CARDIOVASCULAR HEALTH INFORMATION There is a paucity of information about women‘s knowledge of CVD although attempts to communicate information to the Welsh public are not new. In 1981, the Health Education Council identified CHD as a health priority resulting in a co-ordinated five-year programme called ‗Heartbeat Wales‘ (DoH, 1985). A range of communication strategies were adopted in the campaign including the use of media, smoking cessation programmes and exercise initiatives. From a community perspective, this programme encouraged the setting up of smoke-free areas and other work place initiatives. The programme recognised the ineffectiveness of relying heavily on the provision of information as a means to change health behaviour, and of the need to take account of how behaviour is shaped by the social and physical environment. Since the inception of ‗Heartbeat Wales‘, there have been other initiatives to enhance knowledge about health issues including CVD, as currently evidenced in ‗Health Challenge Wales‘ where one publication is devoted specifically to CVD in women and where there is general advice about leading a healthy lifestyle (WAG, 2004).
In addition to formalised channels of communication health knowledge is known to cascade through communities via individuals who are perceived to be knowledgeable and therefore influential (Rogers, 1995). For example, a GP communicating messages about the importance of physical activity to patients who are sedentary contributes to positive innovation within a community. Additionally, knowledge of health issues including CVD can cascade through a network of social contacts, friends and family. Therefore, within a given demographic locality and culture, collective consciousness of health issues can come from diverse channels. In light of the conscious raising efforts of the BHF and of ‗Health Challenge Wales‘ (2004), in combination with health messages from media channels of communication, it is probable that Welsh women can easily access cardiovascular health messages from at least one, and probably more sources.
20
2.2.5 INFLUENCES ON WOMEN’S HEALTH BEHAVIOUR The persuasion of people to change an identified behaviour in favour of another is a complex issue not only because of ethical implications but because people do not live in a vacuum, whereby they remain free of outside influence and demands. Whilst women in the locality of the study live in a culture where there is relative ease of access to health information, they have a degree of choice over whether to act on advice. Moreover, should an individual decide that information about making a behaviour change would be beneficial to them there is no guarantee that they would have the motivation or other personal resources to follow through with behaviour change. In addition to knowing about situations requiring urgent action, knowing about CVD also includes knowing about the major modifiable risk factors and behaviours considered cardio-protective such as habitual physical activity, eating a healthy diet or non-smoking. Knowledge alone however, is not a guarantee that individuals will adopt and maintain healthy behaviours over the long-term. Maibach et al. (2002) considered the usefulness of education as a means to influence the behaviour of a target audience pointing out that it could be successful under certain conditions where the audience perceives the benefits of a recommended behaviour as attractive, and where barriers to that behaviour are perceived, as minimal. For example, it might cause minimal inconvenience to attend a GP surgery for blood pressure measurement once every five years but exercising five times weekly, and habitually maintaining that behaviour might be perceived as more difficult.
With regard to women knowing about the benefits of physical activity Prochaska and Sallis (2006) pointed out that only very motivated individuals, or those who are strongly encouraged, will seek out information. Furthermore, the challenge for the communication of health messages regarding physical activity is to reach the proportion of the population who have no interest or awareness, and to continue support for those who are already active. Prochaska and Sallis (2006) recommended repeated presentation of information and suggestions for behaviour change strategies using a variety of communication channels. These can include personal communication, print media including information pamphlets, newsletters, advertisements, exercise diaries, and electronic media including radio, telephone and the internet. Prochaska and Sallis (2006) pointed out that whilst traditional health messaging adopted a ‗one size fits all‘ approach, more recent efforts have focused on the development of targeted communication directed at a specific segment of the population; furthermore recommending use of a communications model to frame the health message. This would take account of the source of the message, for example, a celebrity or other model who endorses the targeted health behaviour, the delivery channel, for example,
21
electronic or personal communication; the characteristics of the target audience such as age and gender, and the context; for example, a worksite physical activity programme.
A condition of learning pertains to the audience personally identifying with the health message and developing perceptions of personal susceptibility. According to Thornton (2003) personal perceptions of health-risk result from multiple sources including experiences over time, social networks, behavioural norms, exposure to the media, perceived susceptibility to disease and a desire to take personal responsibility for health. Steenkist et al. (2004) investigated the ideas, fears, and expectations of coronary risk in 22 patients from GP surgeries in Amsterdam. In-depth semi-structured interviews were conducted to ascertain patients‘ perceptions of CVD risk factors. Patients‘ understanding of how to prevent CVD was considered insufficient and risk perceptions were unrealistic. There was an undue emphasis by patients on the importance of cholesterol as a risk factor, and less emphasis on the danger of smoking cigarettes. Furthermore, hypertension was accepted as a routine issue by patients with an expectation of a need for treatment. There were also communication barriers between the GP and patients. Some patient‘s perceptions were directly influenced by searching the internet. The findings indicated that discussion between GPs and patients about taking responsibility for CVD prevention might be challenging. Therefore, enhanced awareness about personal risk of CVD and awareness of a need to change behaviour did not guarantee the patient would be willing or able to make behaviour changes to reduce risk.
2.2.6 REACHING A WIDE AUDIENCE THROUGH MEDIA CHANNELS Television and newspaper advertisements can be utilised effectively to reach a wide audience as for example in infomercials about stroke awareness. Whether media coverage is always in the best interests of the public however, is questionable. According to Tones and Green (2006), public health has a tradition of using mass media to try to influence the behaviour of the public but with disappointing results. Furthermore, Naidoo and Earle (2007), warn of the potential for media health message fatigue, with people becoming desensitized to shock tactics or developing ‗risk factor phobia‘ with the likelihood of increasing anxiety about lifestyle or becoming deliberately defiant about health warnings.
Televised health messages appear to have three observable uses. The first involves programmes sponsored directly by television companies with the aim of attracting large viewing figures through a ‗health as entertainment‘ remit. The second consisting of television advertisements sponsored by the Department of Health (DoH), for example, those that encourage people to access vaccinations against various infectious diseases or to become 22
more physically active. The third consisting of advertising marketing ploys targeted at the viewing public and portraying certain products as a healthy choice. Where the medium of television focuses on health issues as entertainment, it is pertinent to question the degree to which programme content reflects best practice in terms of sound health knowledge. For example, the portrayal of obese people undertaking punitive exercise regimes focussing on fat loss may be considered entertaining but could be harmful to the obese person‘s psychological and physical well-being, and contrary to what most fitness professionals would advise for safe body-fat loss and cardiovascular health.
Models and theories exist that attempt to explain how people change their behaviour. According to social cognitive theory, the perceived confidence of the individual to perform the behaviour is an important determinant of that behaviour (Bandura, 1986). Influences on confidence include feelings associated with mastering the behaviour, verbal persuasion, physiological and affective responses and the vicarious experiences of others. This latter influence involves the observation of other people, who the individual perceives is similar to them. For example, middle-aged women watching other women of similar age walking briskly in an organised walking programme might draw comparison and decide that they are also capable of this behaviour. This modelling of the desired behaviour prompts other similar individuals to perform the behaviour because they feel confident they can perform it. This strategy is widely employed by many popular commercial body-weight management programmes, and is evident in televised advertisements and magazines featuring women of various ages who have apparently been successful in their efforts to lose body-weight by adherence to a prescribed dietary regime. The common characteristic of the target audience is overweight or obesity, and age, gender and ethnicity are accounted for in marketing strategies that use modelling to sell the product. This is not however, the strategy employed by the current ‗Change4Life‘ campaign (WAG, 2011) which chooses to use animated faceless cartoon characters of indeterminate age, to ‗nudge‘ the population towards healthier behaviours, but which potentially make it more challenging to identify with ‗the model‘ and possibly therefore, with the health message.
2.2.7 STAGE BASED TAILORING OF HEALTH INFORMATION Health messaging in general is beset by many challenges including the mode of communication, the construction of the message, and its relevance to the target audience. One solution has been to use stage-based interventions underpinned by the transtheoretical model of behaviour change (Prochaska and DiClemente, 1983). Processes of behaviour change consist of five behavioural and five cognitive strategies that a person uses as they 23
move through stages of changing their behaviour. These stages consist of ‗precontemplation‘ where there is no intention to change behaviour, ‗contemplation‘ where the individual is thinking about behaviour change, ‗preparation‘ where small steps are made towards change, ‗action‘ where the individual has adopted the desired behaviour and ‗maintenance‘ where the individual is performing the behaviour for six months or longer. Stage based interventions tailor health messages individually according to stage of behaviour. A study by Bock et al. (2001) examined predictors of exercise maintenance following completion of a physical activity intervention of six months duration. Sedentary adults (n = 194) were randomly assigned to either a treatment or control group. The treatment group received motivational, print-based material that was stage matched and focused on cognitive processes appropriate for each stage of change, and feedback reports that were individually tailored to psychological variables. The control group received standard self-help manuals developed by the AHA. An assessment-only follow-up was conducted six months after the end of the intervention at month 12. Participants were assessed for current physical activity participation, motivational readiness for physical activity, psychological constructs posited to influence participation in physical activity (e.g. self-efficacy), and current affect. At the 12-month follow-up, 42.0% of the treatment group were meeting recommended guidelines for leisure-time physical activity compared to only 25.0% of the control group.
The provision of educational material tailored to individual needs has proven successful in a work-based intervention with the aim of increasing physical activity levels among employees. Using the theoretical framework provided by the transtheoretical model (Prochaska and DiClemente, 1983), Petersen and Aldana (1999) evaluated the effect of a stage-based exercise intervention in a randomised trial of employed adults (n = 527). Employees were randomly assigned to one of three groups consisting of a stage based intervention group, a generic intervention group, or a control group. Those employees in the stage-based intervention received written messages tailored to their individual stage of change. Employees in the generic group received non-tailored materials based on information from the ‗Report of the Surgeon General‘ on physical activity (1999). Employees in the control group did not receive any materials. A seven-day physical activity recall questionnaire was used to estimate self-reported physical activity levels, and a ‗stage of readiness to change measure‘ was used to determine the stage of change that most accurately described each employee‘s intention to change. This process was repeated after six weeks. The results demonstrated a 13.0% increase in physical activity in the stage-based group; the generic group showed a 1.0% increase and the control group demonstrated an 8.0% decrease with significant differences between all groups (p < 0.05). The authors concluded that stagebased tailored messages appeared to be more effective at increasing short-term activity 24
levels than were generic messages or no information at all. Moreover, those employees who received the stage-based written materials were more likely than those receiving generic based information to report that they had read the information they received. Increases in physical activity were only monitored at the short time lapse of six weeks but the results suggested that when possible, tailored information is likely to be more effective than generic.
2.2.8 USING TECHNOLOGY Advances in technology have opened up possibilities to reach large numbers of people with both generic and tailored health messages using interactive internet sites. According to Marcus et al. (2000), an expert system contains a series of feedback sections based on the constructs considered necessary for behaviour change and framed on behavioural theory. These systems can include normative (compared with population norms) and ipsative (compared with previous scores) feedback on salient variables, and use of the system can provide individualised feedback. Whilst Marcus et al. (2000) point out the initial high costs of development, this interactive communication model might have potential as a further channel of communication with women in the age group of the present study. Pertinently, the use of interactive technology is under development by ‗Change4Life‘ so that by 2014 there will be more of a focus on interactive, practical tools and less on information-driven wordy leaflets (DoH, 2011).
2.2.9 COMPETING HEALTH ISSUES How women perceive their risk of CVD is interesting in light of a highly visible media focus in recent years on women‘s cancer issues. Some studies have demonstrated that many women believe they are more at risk from breast cancer than other illnesses (Fiandt et al., 1999; Wendt, 2005). Although this demonstrates the success of media and NHS efforts to highlight the important issue of breast cancer, it raises questions about women‘s level of knowledge and perceived susceptibility towards CVD. Media campaigns to heighten awareness of breast cancer have appealed to women through powerful marketing strategies, for example, the adoption of pink ribbon emblems and lapel broaches (Lee and Frayn, 2008). Whilst the health issues of breast and cervical cancer are important, this focus might have masked CVD as women‘s major health threat in terms of premature mortality and morbidity. In relation to breast cancer, prominent charity driven media campaigns have run in tandem with NHS screening programmes designed to detect early abnormalities. An example was the Welsh NHS television and billboard campaign that advertised screening for cervical cancer (2007,
25
2008). Whilst breast and cervical screening undoubtedly saves lives, there appears to be an imbalanced amount of attention to women‘s reproductive anatomy perpetuating a ‗bikini‘ approach to women‘s health that could be construed as misleading. Screening services for cervical cancer were established in the UK in 1988. Screening for CVD was only recently put onto a health agenda through the ‗National Service Framework for Coronary Heart Disease‘ (DoH, 2000), and more recently through a National Screening Committee Policy (UK National Screening Committee, 2007). This recommended to the four Chief Medical Officers in the UK, the introduction of a vascular risk management programme through which adults over 40 years of age should be offered risk assessment. Whilst it is futile to argue for a hierarchy of importance in screening services, it is pertinent to question why screening for CVD has traditionally merited less attention than for other medical conditions. Whether women recognise the potential of screening for hypertension and cholesterol abnormalities and understand the rationale behind risk factor management is unknown.
2.2.10
SURVEYS
OF
WOMEN’S
KNOWLEDGE
AND
AWARENESS
OF
CARDIOVASCULAR DISEASE The value of investigating women‘s knowledge and awareness of CVD lies in the potential to identify gaps in knowledge and then to intervene appropriately and effectively. In a smallscale descriptive study by Oliver-McNeil et al. (2002), 33 American women aged between 36 and 85 years newly diagnosed with CHD answered a questionnaire about risk factors for CVD. Only four women could identify high cholesterol, stress, or diabetes as risk factors. Only one woman identified menopause or age, and only two identified smoking. Furthermore, the risks identified were fewer and differed from those documented in the women‘s medical history. These results demonstrated a concerning low level of knowledge and awareness with the consequence that if women perceive their susceptibility is low based on a lack of information they could significantly underestimate their personal risk of CVD. This has implications for health behaviours necessary for the prevention of CVD and this poses a challenge for the promotion of healthy behaviours. Whilst the sample in the study by OliverMcNeil et al. (2002) was small, the results demonstrated a need for further investigation of women‘s knowledge and perceptions of cardiovascular risk. In a clinical investigation of women‘s awareness of CHD disease by the AHA (Mosca et al., 2004), the knowledge, awareness and perceptions relating to CVD risk was evaluated through means of a telephone survey (n = 1024). Nearly 50.0% of respondents identified CHD as the leading cause of death for all women, but 20.0% cited cancer in general, and 15.0% cited breast cancer, indicating a need for improved awareness of CVD. However, in 26
terms of the general population of American women the sample size was small, indicating a need for further research with larger sample sizes, in varying locations, and with women from various ethnic origin and age groups.
Ferris et al. (2005), conducted a telephone survey of women aged over 25 years to investigate knowledge and awareness of stroke (n = 1024). Increasing age was associated with less stroke awareness with only 26.0% of women aged over 65 years reporting as ‗well informed‘ compared with 63.0% of women aged between 25 and 34 years. This survey found overall knowledge of the warning signs of stroke to be low among all ethnic groups leading Ferris et al. (2005) to conclude that knowledge and awareness was suboptimal. Furthermore, although 69.0% of women were correctly able to identify atypical or unusual signs of a heart attack, this left little room for complacency. Respondents were asked to identify the causes of CHD and only a minority were able to name major risk factors; 36.0% cited smoking, 31.0% high cholesterol, 29.0% family history, 19.0% hypertension and 7.0% stated diabetes. This latter finding is of particular concern because of the greater propensity for women to develop CHD as a complication of diabetes (Stramba-Badiale et al., 2006).
The above studies related to North American populations. If however, knowledge and awareness of CVD are shaped by diverse factors including socio-cultural determinants, media, and public health policy there could be differences across geographical localities. The React Survey (Erhardt, et al., 2002) investigated perceptions of CVD risk across five European countries including the UK. The participants (n = 5104) were men and women aged between 40 and 70 years. They were interviewed using qualitative and quantitative oriented questions to ascertain health status, and knowledge of risk factors. Participants rated CHD (45.4%) and cancer (44.8%) as equal leading causes of death, although participants from the UK and Sweden were more likely to identify CHD as the leading cause of death in their country. When asked to assess personal risk of developing CHD however, only 6.0% of the sample believed they were at high risk of future CHD. This was higher for UK participants where the figure was 9.0% (p < 0.05). Smoking as a risk factor was correctly identified by 70.0%, as was high blood pressure by 65.0%, obesity and overweight by 62.0%, and stress by 58.0% of participants. Only 18.0% expressed concern about cholesterol levels. Erhardt et al. (2002) reported that only after prompting could participants make an association between high cholesterol and CHD, one in three participants were unaware of the link, and only one in three were aware of their target cholesterol level. When asked if they had discussed cholesterol levels with their GP, 69.0% of UK participants stated ‗they had not‘.
27
Information about knowledge, awareness and perceptions of CVD held by women in the UK is sparse. Incidence and prevalence rates of risk factors might provide much needed information for health professionals concerned with policy formation and treatment but do not provide insight into the knowledge and awareness of CVD held by women. Whilst knowledge alone does not guarantee that women will adhere to a healthy lifestyle, it is a move in the right direction in that it might facilitate their engagement in decision-making about their personal risk of CVD. The development of a profile of women‘s knowledge and perceptions, together with corroboration of their smoking, obesity and physical activity status, sources of information and screening behaviour, is therefore both timely and pertinent to the current climate of initiatives to improve the health of people in Wales.
2.2.11 SUMMARY The review discussed women‘s major risk factors for CVD in light of the potential for women to be pro-active in prevention. However, taking preventative steps requires knowledge and awareness of the complex nature of the disease and the behaviours necessary to lower risk. Women‘s exposure to health information about CVD comes from diverse sources including the media, information leaflets, internet sources and personal communication but there is a dearth of information regarding Welsh women‘s preferred sources, and about their knowledge base of CVD. Furthermore, little is known about woman‘s screening behaviour for cholesterol and blood pressure although accessing these health checks is an important proactive step women can take. There is information from Welsh Health Surveys (WAG) regarding the prevalence of overweight and obesity, and smoking,
and about physical
activity levels but not specific to the locality of the present study. How women perceive their risk of CVD is also unknown. In light of these issues, the objectives of the present study were as follows:
2.2.12 OBJECTIVES OF STUDY ONE i.
To determine the prevalence of overweight and obesity based on BMI and WC according to criteria of the WHO (2004) and Diabetes UK (2006).
ii.
To determine self-perceptions of body weight as ‗healthy‘ or ‗unhealthy‘ and to compare perceptions with BMI status.
28
iii.
To determine the prevalence of women at risk of long-term health problems including CVD based on combined BMI and WC guidelines issued by NICE (2006).
iv.
To determine the percentage of participants‘ who had voluntarily sought blood pressure and cholesterol screening in the past five years.
v.
To determine the prevalence of cigarette smoking, the percentage of smokers who reported as overweight or obese, and to compare the uptake of screening for high blood pressure and cholesterol between smokers and non-smokers.
vi.
To determine participants‘ physical activity levels in order to establish the degree of adherence with current recommendations (Haskell et al., 2007).
vii.
To ascertain participants‘ perceptions of their physical fitness.
viii.
To ascertain participants‘ preferred primary sources of cardiovascular health information and the percentage of participants who had discussed their risk of CVD with a health practitioner.
ix.
To determine participants general knowledge of CVD evaluated through scoring on 50 knowledge-based questionnaire items.
x.
To elicit participants perceptions of their risk of CVD compared to their peers.
2.3.1 RATIONALE FOR THE SURVEY To address the objectives of the study a cross-sectional survey was conducted with a sample of women aged between 25 and 65 years living in the Vale of Glamorgan, in the UK (n = 724).
Surveys can provide a broad overview of a situation or phenomenon in order to
compare and contrast across groups and assess specific variables pertinent to stated aims. They have potential to capture objective reality as it exists at a given time, and elicit relationships between variables (Creswell, 1994). Whilst surveys can involve various methods of data collection, the aim of a cross-sectional survey is to access data from a specified population at a single point in time thus providing a fractional description of an issue (Creswell, 1994). A single cross-sectional survey can also increase in potency when repeated longitudinally because this facilitates evaluation of trends and longitudinal changes
29
(Calnan, 2007). This method can facilitate both an immediacy of understanding as well as monitor changes over time.
Whilst surveys can be conducted through the medium of telephone calls, face-to-face interviews, focus groups and internet formats, the protocol adopted for the present study was a questionnaire that was distributed to varied places where women were likely to meet. These included places of employment, communal leisure centres, a slimming club, social clubs, and venues offering childcare. The advantage of a distributed questionnaire compared to other means of reaching targeted respondents lies in the potential to access a larger sample, thereby increasing the likelihood of a representative response rate.
There are necessary considerations in adopting this method of data collection. The questionnaire format and structure, and covering information detailing purpose, ethics and assurances of confidentiality need to be sufficiently attractive to motivate the respondent to act. The questionnaire must be meaningful to the respondent particularly where relying on voluntary participation with no reward. A disadvantage is that questionnaire items may veer towards eliciting stringent objective information at the expense of more qualitative interaction afforded by focus groups and so the decision concerning which method to employ must be cognizant of this and design the format accordingly (Saks and Allsop, 2007).
Although
questionnaire items can be left open for the respondent to share beliefs, feelings and attitudes, this may be inferior compared with direct opportunity to interact with the researcher to clarify issues of concern. Due to the intentions of the present study however, other means of achieving the aims, for example, through focus groups, would have been inappropriate to address some of the issues included. For example, one item had potential to cause embarrassment if included in a group situation whereby the participant was required to measure WC. For the purposes of the present study, a tape measure was inserted with each questionnaire. Requesting this measure in the setting of a focus group would have been intrusive, whereas a questionnaire could be completed anonymously.
2.3.2 THE PARTICIPANTS The participants were apparently healthy women aged between 25 and 65 years. They were invited to participate if they were not pregnant, or diagnosed with a chronic health problem, for example, heart disease or diabetes. Pregnant women were ineligible on the basis that their physical status invalidated their responses to some questionnaire items.
30
2.3.3 SAMPLING The intention was to access a sample of women across the age spectrum of the study (25 to 65 years). Access to a representative sample was constrained by data protection laws and ethical procedures for approaching volunteer participants. Therefore, convenience sampling was employed to heighten the probability of access to a representative sample of women. The sampling frame considered the complex social structures that typify women‘s lives and in light of this, the primary investigator sought to access women where they met during the course of their daily life. It was not feasible to attempt to access a sample using postal distribution based on postcode due to the problems inherent in acquiring information about where women live. Convenience sampling eased the recruitment process and was reliant on women volunteering to complete a questionnaire and return either by agreed collection or by post. This type of recruitment method has limitations in that there is reliance on the respondent to return the completed questionnaire and bias might be introduced through overrepresentation of one group with some respondents more motivated to complete the questionnaire. To minimise this effect, whilst dependent on voluntary completion the questionnaires were distributed to diverse locations and were targeted at women across the age group of the study. For example, account was taken of the fact that some women were engaged in child rearing and therefore were more likely to be found in mother and toddler groups as well as places of employment, and other locations. Additionally, older retired women might have opportunities for many social and cultural pursuits and consideration was given to the type of locations in which they might be approached. The researcher identified a number of possible locations for distribution of the questionnaire. These comprised of the following localities:
i.
Retail and other commercial businesses across the locality.
ii.
Community centres for exercise and recreation, including three council run leisure centres, a yacht club, social club, and golf club.
iii.
Crèche and nursery facilities.
iv.
School teaching and ancillary staff from local schools.
v.
Two public houses.
vi.
Lady members of a Royal British Legion club.
vii.
A ‗Slimming World‘ club.
The Vale of Glamorgan has four main localities where retail and other businesses and charity shops are heavily clustered. It was logistically possible to approach between 20 and 40 businesses in each locality and obtain permission to inform employees or volunteer workers 31
about the study. Between one and 20 questionnaires were distributed at each business premises depending on the number of employees. Only three retail outlets declined to take part, citing lack of time. The opportunity was therefore available to capture the responses of women working full-time or part-time and, who were representative of various age groups.
The researcher approached managers at four community leisure centres for permission to inform women customers of the survey. Forty questionnaires were distributed at each venue. Leisure centres cater for a wide age range and women who completed the questionnaire were attending general fitness classes. Employees of the leisure centres were also informed about the study and were invited to participate. Fifteen questionnaires were distributed to members of a yacht club. Fifteen questionnaires were distributed to a Royal British Legion club. The manageress of two children‘s nurseries was approached and she agreed to leave information about the survey for women using child-care services. Thirty questionnaires were left for potential participants at each venue. Women at these venues tended to be representative of the younger age group of the study (25 to 45 years). Thirty questionnaires were distributed in a Women‘s Institute meeting where potential participants were aged over 50 years and the majority were retired. It was possible to access regular female customers at two public houses and leave 40 questionnaires. A further 20 questionnaires were distributed to women attending a knitting circle. Permission was obtained from Head teachers to leave 50 questionnaires for teaching and ancillary staff at four local schools. Finally, the organisation ‗Slimming World‘ was contacted requesting permission to inform clientele of the survey. After a period of three months, the ethics committee of Slimming World, approved the request and gave permission for distribution of 300 questionnaires. Unfortunately, distribution and collection was only to be organised through the leaders of each local group and there was a total failure of class leaders to comply with instructions from senior management of Slimming World. This lack of cooperation limited the final response, which resulted in 731 returned questionnaires from 1600.
2.3.4 DEVELOPMENT OF THE QUESTIONNAIRE The wording and design of a questionnaire are essential if the researcher is to obtain reliable responses. Polgar and Thomas (2000) warn of potential pitfalls, for example, the inclusion of ambiguous, biased or leading questions. Whilst open-ended and closed-response are the two major formats adopted in questionnaires, the latter format was adopted as appropriate for meeting the aims of this study. During development of the questionnaire, account was taken of previous health surveys undertaken in Wales. These provide an ongoing source of information about the health, and health-related lifestyle of people living in Wales. The 32
original survey was established in 1985 and was known as the ‗Health in Wales Survey‘. The more recent Welsh Health Survey first took place in 1995 and provided baseline data for a range of health targets and included items specific to health status in relation to self-reported illness and disability, and lifestyle characteristics. A revised WHS commissioned by the WAG in 2003 was implemented in 2003/4, 2004/5, 2007 and 2009. The format adopted for these surveys consisted of a set of multiple choice questions mailed to over 50,000 of the male and female adult population over the age of 16 years generating an average response rate of 59.0% (WAG, 2007). Whilst comprehensive in their remit, these ongoing surveys do not determine knowledge of CVD risk factors, or personal perceptions of cardiovascular health leaving a gap in what is known about women‘s knowledge and awareness of CVD. Therefore, to address the aim of the present study a multiple-choice questionnaire was designed consisting of 27 items organised into three sections and which contained 50 knowledge-based questions. The resulting data was organised into categories consisting of responses from women aged between 25 and 65 years, women aged between 25 and 45 years, and women aged between 46 and 65 years. Participant information is presented in Appendix A, the questionnaire in Appendix B, and the consent form in Appendix C.
2.3.5 KNOWLEDGE AND AWARENESS OF CARDIOVASCULAR DISEASE Knowledge of CVD was assessed through multiple choice questions based on information that women were likely to have been exposed in their daily life. The following sources provide examples:
i.
British Hypertension Society campaigns suggesting blood pressure assessment, and offering general advice about blood pressure.
ii.
National Health Service campaigns and services for smoking cessation.
iii.
The ongoing ‗Measure Up‘ campaign by Diabetes UK (2006) suggesting waist measurement as a risk factor for diabetes and CVD.
iv.
British Heart Foundation campaigns, leaflets, posters and internet advice and support.
v.
The WAG initiative, ‗Health Challenge Wales‘ (2004) outreaching the Welsh population about lifestyle issues including physical activity.
33
Scoring of the 50 knowledge-based questions took into account the difficulty in deciding what score might indicate a ‗good‘ knowledge of CVD. Ideally, the majority of women would be in possession of sufficient basic information to score highly, but taking into account a possible overall range of scores, for the purpose of the present study, a score of 40 or more was regarded as a good score; 25-39 as average, and under 25 as poor. It was acknowledged by the primary investigator that this decision was subjective, however a score of 40 or more would represent 80.0% or more correct responses to knowledge questions, scoring 25 -39 would represent 50.0% or more correct responses and less than 25 would indicate a knowledge base of less than 50.0% of information currently in the public domain.
2.3.6 SECTION ONE: AGE, MARITAL AND EMPLOYMENT STATUS This section of the questionnaire required responses to items concerning age, employment and marital status to facilitate description of the overall sample and to further categorise into two age groups consisting of group 1 (25 to 45 years), or group 2 (46 to 65 years). Chisquare analysis was undertaken to determine any relationship between marital status and physical activity level and between employment status and physical activity.
2.3.7 SECTION ONE: BODY MASS INDEX Participants were requested to self-report body-weight and stature (height). These items were included to facilitate estimation of BMI for each participant. The use of self-report items requesting anthropometric measures has potential for inaccurate responses, or no response. Danubio et al. (2008) found that women tend to over-report their height and under-report their body-weight. However, it was logistically impossible for the primary investigator to achieve the objective for this section of the present study without the use of a self-report item. Therefore, participants were asked to self-report body-weight and stature in order to estimate BMI on the basis that in previous surveys sufficient statistical agreement had been found between researcher and respondent measurements to accord value to inclusion of these items (Heartbeat Wales, 1985; Welsh Health Survey, 2003/04). Based on categorisations of the WHO (2004) a BMI of less than 18.5 kg m-2 was categorised as underweight. A BMI between 18.5 and 24.9 kg m-2 was categorised as a normal healthy body-weight. A BMI between 25 and 29.9 kg m-2 was categorised as overweight, and a BMI at or more than 30 kg m-2 was categorised as obese.
34
2.3.8 SECTION ONE: ABDOMINAL ADIPOSITY Whilst there are recommendations for anthropometric measurement of abdominal adiposity, for ease of measurement and coherence across the sample, participants were requested to record their WC measured at the level of their naval. Following the pilot study described in section 2.3.16 of the present study, a paper tape measure was inserted into each folded questionnaire to increase the response rate. One paper tape measure from each batch of fifty was used to calibrate the metric measurement. Responses were compared against thresholds advised by the Diabetic Association (2006) for a WC less than 80 cm.
2.3.9 RISK OF CARDIOVASCULAR DISEASE AND LONG TERM HEALTH PROBLEMS Potential increased risk of chronic health problems including CVD was estimated by combined BMI and WC based on NICE guidelines (NICE, 2006). Categorisation was undertaken as follows:
i.
Women with a BMI between 25 and 29.9 kg m-2 and a WC of less than 80 cm were categorised as ‗at no increased risk‘.
ii.
Women with a BMI between 25 and 29.9 kg m-2 and a WC between 80 and 88 cm were categorised as ‗at increased risk‘.
iii.
Women with a BMI between 25 and 29.9 kg m-2 and with a WC of more than 88 cm were categorised as ‗at high risk‘.
iv.
Women with a BMI more than 30 kg m-2 and a WC of less than 80 cm were categorised as ‗at increased risk‘.
v.
Women with a BMI more than 30 kg m-2 and with a WC between 80 and 88 cm were categorised as ‗at high risk; and those with a WC more than 88 cm as ‗at very high risk‘.
35
2.3.10 SECTION ONE: CHOLESTEROL SCREENING The BHF (2008) advises that total cholesterol for women should be less than 5.0 mMol L-1. Participants were asked if they had sought to have their cholesterol assessed within the past five years, and if they could identify a healthy cholesterol measure from four options. Two options were over the recommended level, one was the advised correct level, and a fourth option allowed the respondent to state that they did not know a correct level. This enabled estimation of the percentage of participants who reported having their cholesterol measured, and their knowledge of a healthy cholesterol level.
2.3.11 SECTION ONE: KNOWLEDGE AND AWARENESS OF HIGH BLOOD PRESSURE The BHS (2004) recommended that all adults should have their blood pressure measured every five years. A questionnaire item sought to clarify whether the participant had taken steps to have their blood pressure measured in the past five years, and a further item required identification of a normal blood pressure measure from four options. This enabled estimation of the percentage of participants reporting having their blood pressure measured, and the participant‘s knowledge of what would be a healthy blood pressure measure.
2.3.12 SECTION ONE: SMOKING STATUS In light of current legislation to ban smoking in public places, smoking status was assessed by asking the participant to state if they smoked one or more cigarettes on a weekly basis. A participant was classified as a smoker with an affirmative response.
2.3.13 SECTION ONE: PHYSICAL ACTIVITY STATUS Physical activity status was assessed using a self-report measure although this method has a potential for error (Ainsworth, 2000). This is because of a high probability of contamination due to an inaccurate estimation of activity, which for most women might include household and care-duty tasks in addition to other more easily quantifiable activities such as leisuretime fitness pursuits. Therefore, the varied roles women play in their daily life makes objective assessment of physical activity challenging (Ainsworth, 2000). It was not logistically possible in the present study to use a more robust method, for example, the use of pedometers and therefore, a seven-day recall question was utilised. This consisted of three 36
sections designed to elicit information about frequency, intensity, and duration of physical activity. Examples were provided of activities representative of light, moderate and vigorous intensity. Furthermore, the participant was requested to state the frequency of their participation in the past seven days. This facilitated judgement about physical activity levels enabling estimates of women‘s physical activity status. The CMO (2004) advised that moderately intense activity be undertaken on five or more days a week, for 30 minutes or more. A recommendation of the ACSM and the AHA advised a flexible approach, also taking account of vigorous doses of activity, but on fewer days of the week (Haskell et al., 2007). In surveys that discount these recommendations there is potential to underestimate the percentage of women who meet guidelines for a physically active lifestyle.
Whilst recognising the potential fallibility of self-report estimation of physical activity the seven-day recall question was useful as a guide to the percentage of women complying with recommendations. The Welsh Health Survey (WAG, 2004/5) used a seven-day recall question and calculated the percentage of women who were complying with five sessions a week of either moderate, or a combination of moderate and vigorous activity. Those women who reported undertaking vigorous activity three times a week were discounted leading to a potential underestimation of the percentage of active women in Wales. The position adopted in the present study was to accept any of the following criteria as indicating adherence to current activity guidelines:
i.
Moderately intensity activity on five days a week for 30 minutes or more.
ii.
A combination of moderately and vigorously intense activity on five days a week for 30 minutes or more.
iii.
Vigorously intense activity on three days a week for a minimum of 20 minutes.
Those women indicating that they only undertook light activity were classed as sedentary.
2.3.14 SECTION TWO: KNOWLEDGE OF CARDIOVASCULAR DISEASE AND SOURCES OF INFORMATION
Section two of the questionnaire related to women‘s sources of CVD information, their perceptions of how knowledgeable they were about CVD, and their general knowledge of CVD.
37
Question 14 requested participants to identify the source considered as most influential in their knowledge of CVD, for example, television or women‘s magazines. This item included an option to give another source if contrary to those given. The participants were presented with three options relating to their perceptions of how well informed they were regarding CVD and these were; ‗well informed‘, ‗informed‘, or ‗not at all informed‘. Additionally, participants were asked if they had ever discussed CVD prevention with a GP or nurse.
A further eight questions focussed on general knowledge of CVD and consisted of 48 multiple-choice items. Together with the questions in section one referring to cholesterol and blood pressure, this gave a possible overall knowledge score of 50.
The aim was to
measure the following:
i.
General knowledge of CVD.
ii.
Symptoms of a myocardial infarction and of stroke.
iii.
Risk factors for peripheral artery disease.
Question 17 consisted of five statements focussed on CVD as the main cause of mortality in women and asked whether breast cancer or CVD causes more deaths. There were three response possibilities which consisted of ‗agreeing‘ with the statement, ‘disagreeing‘, or the participant stating ‗do not know‘. The intention was to assess knowledge but to eliminate the possibility of guesswork.
Knowledge of major risk factors (Q18) was assessed through a series of statements requiring affirmation of those thought to place an individual at risk of CVD. Correct responses to this question were smoking, a fatty diet, inactivity, stress, high cholesterol, high blood pressure, and overweight or obesity. The participant was asked to rate the given risk factor as ‗important‘, ‗not important‘ or to state ‗do not know‘.
Question 19 required the respondent to indicate at what stage of life CVD develops. The choices were ‗childhood‘, ‗adolescence‘, ‗young adulthood‘ or ‗middle age‘. Question 20 referred to symptoms of a myocardial infarction. This was termed as a ‗heart attack‘ because of its familiar usage. Correct responses would indicate a raised awareness, with women in a position to recognise symptoms and take action. Poor levels of awareness would indicate a need for gender oriented health messages. 38
2.3.15 SECTION THREE: PERCEPTIONS OF RISK Section three of the questionnaire evaluated women‘s perceptions of their cardiovascular risk, perceived level of physical fitness and healthy body-weight. Question 25 asked the participant to rate their weight as ‗underweight‘, a ‗healthy weight‘, a ‗little overweight‘, or ‗too overweight‘ to be healthy. The BMI for each participant was estimated enabling comparison to the participant‘s response. For example, a participant rating their weight as ‗healthy‘, but with a BMI of more than 25 kg m-2 would indicate an incorrect perception and a possible lack of knowledge of healthy body-weight parameters. Question 26 asked the participant to rate their present physical fitness status as ‗excellent‘, ‗very good‘, ‗reasonable‘, ‗poor‘ or ‗very poor‘. The response was compared with data from the seven-day recall question (Q 13). A comparison was undertaken between declared activity in the previous week and rating of physical fitness. A mismatch would indicate a misperception of fitness and possible lack of knowledge about current recommendations for physical activity. Perceived susceptibility to CVD was evaluated by asking the participant to rate vulnerability in comparison to peers (Q27). Four options were given to allow for strength and direction of belief consisting of ‗higher than others‘, ‗slightly more‘, ‗the same as for other women‘, or ‗less than other women‘.
2.3.16 PEER REVIEW Following questionnaire development, five people knowledgeable about survey design and women‘s cardiovascular health completed a peer review. This involved four academic colleagues and the co-ordinator for NHS cardiac rehabilitation services in the study locality. After discussion, the decision was made by the primary researcher to include the variable of emotional stress in question 18 of the questionnaire as a cause of CVD. This had potential for ambiguity due to possible misinterpretations of the concept by lay people and some health care professionals. The BHF internet site titled ‗Women and Heart Disease‘ (2009) however gives a lay explanation of the involvement of stress in CHD and because this information was available in the public domain, the decision was taken to include this item.
2.3.17 THE PILOT STUDY A pilot study was undertaken with a convenience sample of 15 apparently healthy women aged between 25 and 65 years. The mean (±SD) age was 43 years ± 12.4. There were
39
seven participants in age group 1 (25-45 yrs) and eight participants in age group 2 (46-65 yrs). Table 1 illustrates the characteristics of the sample with respect to age, marital status, ethnicity and employment.
Table 1, Characteristics of the pilot study sample population (n = 15)
Age (yrs) 25 25 28 30 37 40 41 48 49 49 49 53 54 60 64
Ethnicity
Marital status
White White White Chinese White Asian White White White White White White White White White
Single Single Separated Married Married Married Married Divorced Married Married Single Married Married Widowed Married
Employment Full time Full time Part time Full time Full time Full time Home maker Full time Part time Full time Full time Charity worker Part time Full time Retired
The pilot responses were analysed in terms of the clarity of each question and instruction. The participants were asked to comment on the clarity of questionnaire items. As a result, item seven, which required the participant to self-report their WC revealed an absence of responses because women did not know what their WC measured. To heighten the probability of responses to this question a paper tape measure was inserted into subsequent questionnaires to increase the possibility of item completion. For question 10, the pilot revealed lack of question clarity. This item was amended to ask the participant to indicate a measure of healthy blood pressure from four given options.
2.3.18 PROCEDURE Information about the study, consent forms, and the questionnaires were distributed over a four-month period across the study locality by identifying places where women met on a regular basis. Enquiries were made at each venue regarding the number of questionnaires that could be left for completion. Where possible, a follow up telephone call was made one week after distribution to each venue with the aim of increasing the response rate. A stamped addressed envelope was included with each questionnaire. 40
A further four months was allowed for responses to be received by the researcher through the mail, or collected. Where necessary a courtesy telephone call was made in advance to ensure legal and courteous entry to a given facility. Details of study eligibility were enclosed in an information sheet attached to each questionnaire.
2.3.19 ETHICS There are ethical issues related to the possible effects of the contents of a questionnaire (Polgar and Thomas, 2000). It cannot be assumed that no ill effects will result from a particular line of enquiry or if questions of a highly personal nature are included. The assurance of anonymity can protect the participant from potential feelings of embarrassment, but due consideration must be paid to the protection of the participant from psychological harm. For example, the questionnaire used in the survey requested the participant to disclose their height, weight and waist circumference and this had potential to cause women some concern. With this in mind, the information sheet made clear the voluntary nature of the study and assured anonymity and confidentiality. Before distribution of the questionnaires within the community, Cardiff School of Sport Ethical Sub-Committee approved the study.
The questionnaire used in the survey was scrutinised by four senior health professionals working in the field of cardiac rehabilitative care, and six women volunteers read and discussed the content in a group setting. No objections were raised by them about the sensitive nature of any questions. In accordance with accepted ethical guidelines, all participants received information detailing the purpose of the study and informing them about the researcher. Participants‘ were assured of confidentiality and furthermore, they were informed that all data was stored in a secure location. Potential participants were informed of their right to abstain from participating in the study.
2.3.20 QUESTIONNAIRE ANALYSIS AND PRESENTATION OF THE RESULTS The questionnaire was developed using SNAP Survey Software Version 8 for questionnaire design and analysis (SNAP, 2005). The results were obtained as absolute counts and percentages, and the filtering of subsets was undertaken where appropriate. Descriptive statistics enabled interpretation of the results. Graphs and tables were used to complement the discussion where appropriate. For ease of understanding and presentation, the results and discussion were combined as one section.
41
The sample was described in terms of age, employment status and ethnicity followed by presentation and discussion of the results for BMI and WC as indicators of overweight and obesity. This section included discussion of women‘s perceptions of their body-weight status as ‗healthy‘ or ‗unhealthy‘ and a focus on the physical activity profile of women who reported as ‗overweight‘ or ‗obese‘. This was followed by the results pertaining to screening behaviour for cholesterol and blood pressure and participant‘s knowledge of healthy measures. The results for smoking status were discussed in light of women‘s knowledge and perceptions of the risk of CVD associated with smoking cigarettes. The results for self-reported physical activity status were given as a percentage of women who reported that they complied with physical activity recommendations, and the percentage classed as inactive (Haskell et al., 2007). The results regarding women‘s preferred sources of information about CVD were presented before the results for the 50 knowledge questions and discussion of knowledgebased items highlighted the strengths and weaknesses in women‘s general knowledge of CVD. Finally, the results regarding women‘s perceptions of their risk of CVD facilitated discussion about the general accuracy of these perceptions.
2.4.1 RESPONSE RATE One thousand six hundred questionnaires were distributed and 731 questionnaires were returned to the researcher giving a response rate of 55.6%. Seven questionnaires were discarded due to incomplete responses. There were 361 responses (49.9%) for age group 1 (25-45 yrs), and 363 (50.1%) responses (49.9%), and for age group 2 (46-65 yrs).
2.4.2 DEMOGRAPHICS OF THE SAMPLE The mean (± SD) age of the overall sample was 45 years ± 11.4 years. The distribution of age is shown in Table 2.
42
Table 2, Age breakdown of women aged between 25 and 65 years (n = 724)
Age
Percentage of total
25-30 yrs 31-35 yrs 36-40 yrs 41-45 yrs 46-50 yrs 51-55 yrs 56-60 yrs 61-65 yrs
13.7% 10.3% 12.2% 13.6% 12.4% 14.6% 13.6% 9.6%
Number
(n = 99) (n = 75) (n = 89) (n = 98) (n = 90) (n = 106) (n = 98) (n = 69)
Marital and employment status vary across the population of women in the UK, as does ethnicity. However, women living in the locality of the study were predominantly White British. This was reflected in the breakdown of ethnicity with 97.9% of women reporting as White British, 0.7% of women reported mixed ethnicity, 0.6% reported as Asian, 0.3 % reported as Black or Black British, and 0.6% reported as Chinese or other ethnic group.
The majority of the sample were married or in a long-term relationship as shown in Table 3. The majority of women were in full-time employment as shown in Table 4. Table 3, Marital status of women aged between 25 and 65 years (n = 724)
Marital status
Percentage of total
Number
Single Married Widowed Divorced Separated Other long term relationship
13.4% 59.8% 3.3% 9.4% 1.5% 12.6%
(n = 97) (n = 433) (n = 24) (n = 68) (n = 11) (n = 91)
43
Table 4, Employment status of women aged between 25 and 65 years (n = 724)
Employment
Percentage of total
Number
Self employed Employed full time Employed part time Work related training scheme Retired Waiting to start a job Unemployed Care for home and family Full time student Other
9.5% 45.2% 26.2% 0.6% 9.5% 0.2% 0.7% 5.8% 1.4% 0.9%
(n = 69) (n = 327) (n = 190) (n = 4) (n = 69) (n = 2) (n = 5) (n = 42) (n = 10) (n = 6)
2.4.3 BODY MASS INDEX (Q 5 and 6) Body mass index was categorised according to criteria of the WHO (2004) as ‗underweight‘, ‗normal weight‘, ‗overweight‘ or ‗obese‘. Two percent of the sample (n = 14) did not report either one, or both measures of stature or body-weight therefore it was not possible to estimate their BMI. Two women stated on their questionnaire that they did not know their body-weight and did not want to know because it might cause them to feel miserable.
Figure 1. illustrates the prevalence of overweight and obesity within the overall sample. Fourteen women (2.0%) were ‗underweight' with a BMI less than 18.5 kg-2 and 365 women (51.6%) had a BMI in the normal range (18.5-24.9 kg m-2). Three hundred and twenty-eight women (46.3%) were ‗overweight‘ (25.0-29.9 kg m-2) or ‗obese‘ (≥ 30.0 kg m-2). The mean BMI (± SD) for the overall sample was 25.5 ± 4.9 kg m-2. According to criteria of the WHO (2004), this is categorised as in a pre-obese range with a heightened risk of co-morbidities.
44
60
percentage (%)
50 40 30 20 10 0 Underweight
Normal weight
Overweight
Obese
BMI status according to criteria of the WHO (2004)
Figure 1, BMI status of women aged between 25 and 65 yrs according to criteria of the WHO (2004) (n = 707).
In group 1 (n = 355), 145 women (40.8%) aged between 25 and 45 years were categorised as ‗overweight‘ or ‗obese‘. The mean (± SD) BMI was 25.0 ± 4.5 kg m-2. In group 2 (n = 352) 183 women (51.9%) aged between 46 and 65 years were categorised as ‗overweight‘ or ‗obese‘. The mean (± SD) BMI was 26.1 ± 5.0 kg m-2. The mean BMI for both age groups was above that required for categorisation in a ‗normal‘ range (18.5-24.9 kg m-2) suggesting cause for concern .The BMI status of age groups 1 (g1) and 2 (g2) are illustrated in Figure 2.
70
Percentage (%)
60 50 40 30
g 1 (25-45 yrs)
20
g2 (46-65 yrs)
10 0 Underweight
Normal weight
Overweight
Obese
BMI status according to criteria of the WHO (2004)
Figure 2, BMI status of women in g1 (25-45 yrs) and g2 (46-65 yrs) (n = 707)
45
The Welsh Health Surveys (WAG, 2003/04 and 2007) reported a prevalence of overweight and obesity of 49.0% and 51.0% respectively in women aged 16 years and older. The results of the present study give a prevalence of 46.3% in women aged between 25 and 65 years although this might well be an underestimation because it is accepted that women tend to over report their height and under report their body-weight leading to an underestimation of their BMI (Danubio et al., 2008). The prevalence of overweight and obesity in the present study is concerning because of strong epidemiological evidence of the associated predisposition toward metabolic complications of dyslipidemia, high blood pressure, diabetes, and vascular inflammation, leading to an increase in cardiovascular risk (US Obesity Task Force 1998; WHO, 2003). However, women who are unaware of, or who underestimate the health risks associated with their overweight or obesity are not in a position to make informed decisions regarding body-weight management suggesting a need for health information to address this issue.
Previous Welsh Health Surveys have found an increase in BMI with age. For example, the 2003/04 survey reported BMI by sex and age and demonstrated an increase in obesity from 7.0% of the population of women aged between 16 and 24 years, increasing to 24.0% in women aged between 55 and 64 years. The effort of maintaining a BMI within a healthy range is arguably more difficult with the ageing process because it is characterised by natural changes in body composition and metabolic rate (Spirduso et al., 2005; World Health Organisation, 2011). These unavoidable ageing processes would, if combined with excessive calorie intake and insufficient physical activity provide a reasonable explanation for the agerelated increase in BMI as illustrated in the results of the present study. Women in the present study were asked to rate their weight in terms of ‗underweight for health‘, ‗about the right weight‘, ‗a little overweight for health‘, or ‗too overweight for health‘. Twenty-seven percent of women from the overall sample reported incorrect perceptions either underestimating or overestimating the health risk associated with their body-weight status. Five hundred and seven women (72.6%) correctly judged their body-weight as either ‗healthy‘ or ‗unhealthy‘. Fifty-nine women (8.5%) with a BMI in the ‗overweight‘ or ‗obese‘ classification underestimated their body-weight in relation to their heath perceiving it to be in a healthy range. Of the 328 women categorised as ‗overweight‘ or ‗obese‘, two women failed to rate their risk of CVD, and 126 (38.4%) correctly asserted that their risk was ‗slightly more‘ or ‗higher than other women‘. However, 202 overweight or obese women (61.6%) perceived their risk of CVD to be ‗the same‘, or ‗less than other women‘ suggesting that they lacked awareness of the health risks associated with their body-weight. From an overall perspective, the results suggested that many women in the sample were unable to make an accurate 46
judgement about their body-weight as healthy or unhealthy and might benefit from information about this issue.
2.4.4 ABDOMINAL ADIPOSITY (Q 7) Self-reported WC in the present study was evaluated in relation to a cut-off point of 80 cm and 88 cm to indicate increased risk of chronic health conditions including CVD and type 2 diabetes according to criteria of the Diabetic Association (2006). The overall prevalence of abdominal obesity was 62.9% with 445 women reporting a WC of 80 cm or more. Figure 3 illustrates abdominal obesity classed through WC for group 1 (25-45 yrs) and group 2 (46-65 yrs).
80 70 Percentage (%)
60 50 40
Above 80 cm (31.5 inches)
30
Under 80 cm (31.5 inches)
20 10 0 g1 (25-45 yrs)
g2 (46-65 yrs)
Abdominal adiposity
Figure 3, Abdominal adiposity g1 (25-45 yrs) and g2 (46-65 yrs) (n = 707)
In age group 1 (25-45 yrs) there were 209 women (29.5%) and in group 2 (46-65 yrs) there were 236 women (33.3%) with a WC more than 80 cm. Moreover, 248 (35.0%) women from the overall sample reported a WC of more than 88 cm and according to the National Obesity Forum (2009) would be categorised as at significant risk of diabetes and other metabolic complications. It is undetermined whether women in the UK have knowledge of the health risks associated with abdominal adiposity although a campaign by Diabetes UK titled ‗Measure up‘ (2006) has been visible in the media and in pharmacies clearly stating the advised thresholds. It is a pertinent point that during the pilot study women stated that they were guessing their WC indicating that they might benefit from targeted information about the risks from abdominal adiposity. In guidelines issued by NICE (2006), the risk of ill health 47
including CVD can be estimated by taking into account a combination of BMI and WC. If BMI is between 25 kg m-2 and 29.9 kg m-2 and the WC is less than 80 cm, this is categorised as ‗at no increased risk‘. If however, the WC is between 80 and 88 cm there is an ‗increased risk‘, and more than 88 cm is categorised as at ‗high risk‘ with a BMI of more than 30 kg m-2 regardless of WC categorised as at ‗very high risk‘ of health problems including CVD. Table 5 illustrates the percentage of women categorised at risk of long-term illness aged between 25 and 65 years with a breakdown of results for women aged between 25 and 45 years and between 46 and 65 years. Table 5, Percentage of women at risk of CVD or other long term illness in women aged between 25 and 65 yrs (n = 699) according to NICE guidelines (2006).
Women aged between 25 and 65 yrs (n = 699)
No risk Increased risk High risk Very high risk
59.0% 15.0% 12.5% 13.5%
Total percentage at risk of CVD or other long term illness
41.0%
Age group1 (25-45 yrs) (n = 346)
Age group 2 (46-65 yrs) (n = 353)
No risk Increased risk High risk Very high risk
No risk Increased risk High risk Very high risk
Total percentage at risk
64.0% 12.1% 15.2% 8.3%
35.6%
Total percentage at risk
54.0% 17.5% 15.0% 13.5%
46.0%
This result demonstrating an overall prevalence of 41.0% of participants categorised ‗at risk‘ of long-term health problems, including CVD is concerning and suggestive of a need for interventions to address overweight and obesity in this sample.
2.4.5 SCREENING BEHAVIOUR AND KNOWLEDGE OF CHOLESTEROL AND BLOOD PRESSURE MEASURES (Q 8-11)
HEALTHY
Screening for lipid abnormalities and hypertension are relatively simple procedures accessed through a visit to a GP or recognised pharmacies across the UK. The primary investigator 48
attempted to ascertain the percentage of the adult population who seek out screening of cholesterol and high blood pressure but no evidence was found from the DoH. However, the results of the present study illustrated in Figure 4 demonstrated that neither blood pressure nor cholesterol screening rated highly in women‘s repertoire of protective health behaviours.
70
Percentage (%)
60 50 40 30 Blood pressure
20
Cholesterol
10 0 Had accessed screening
Had not accessed screening Percentage of women who had accesed screening for hypertension and high cholesterol (25-65 yrs)
Figure 4, Participants who had accessed screening for hypertension and high cholesterol (n = 724) Two hundred and sixty-four women (36.5%) from the overall sample had accessed cholesterol screening. There were 81 women (22.4%) in age group 1 (25-45 yrs), and 183 (50.6%) in age group 2 (46-65 yrs). Three hundred and eighty-one women (52.7%) had accessed screening for high blood pressure; 154 women (42.8%) in group 1 (25-45 yrs), and 227 women (62.5%) in group 2 (46-65 yrs). Although this suggests women become more aware of the need for screening as they age, there is an argument for younger women to become more aware at an age when lifestyle changes might well provide a solution before the use of medication is indicated.
Both high cholesterol and high blood pressure are asymptomatic risk factors for CVD (Stangl et al., 2002). The annual campaigns of the British Blood Pressure Association (BBPA) draw attention to the need for blood pressure measurement every five years because of the potential to prevent stroke, myocardial infarction and heart failure (BBPA, 2010). The fact that blood pressure and cholesterol might be elevated for extended periods of time without the individual being aware of the increased risk they pose is concerning. Unfortunately, the results of the present study suggested that despite continuing efforts by the BHF and the BBPA to alert people to the need for screening there are women who do not avail
49
themselves of these important health checks. Furthermore, there is a demographic discrepancy in availability of the National Health Check Programme (NHS, 2010) whereby individuals aged between 40 and 74 years are invited to a consultation with their GP to assess their risk of stroke, CHD, kidney disease and diabetes. This important health check is currently not operational in the locality of the present study potentially leaving women at increased risk unless they take it upon themselves to request screening. The benefit of screening pertains to the possibility of timely intervention particularly because CVD risk tends to increase after menopause (Stangl et al., 2002). Moreover, according to Stangl et al. (2002), CVD risk factors tend to cluster in any one individual and it is therefore sound preventative behaviour for women to avail themselves of screening.
2.4.6 SMOKING STATUS (Q 12) One hundred and nine women reported as cigarette smokers (15.0%). There were 69 women (19.1%) in group 1 (25-45 yrs), and 40 women (11.0%) in group 2 (46-65 yrs).The Welsh Health Survey (WHS, 2007) stated that in women aged 16 years and above 24.0% reported smoking either daily, or occasionally. Future WHSs will be able to more accurately reflect the impact of smoking bans on prevalence rates in women and monitor any trend in age-related smoking behaviour.
Knowledge about smoking as a cause of CVD was sound with 94.0% of women aware of this risk. Amongst the 109 women reporting as smokers, 40.6% were overweight or obese. Furthermore, 55.0% stated that they had not had their blood pressure measured within the past five years compared with 47.3% of non-smokers. Of particular interest was that when asked to rate their risk of CVD compared to other women, 50.5% of women who reported as smokers thought their risk was either the same, or less than that of other women. This finding suggested a need for continued efforts to cascade information to women about the associated health risks of smoking cigarettes and for continuation of smoking cessation programmes.
2.4.7 PHYSICAL ACTIVITY STATUS (Q13)
The results of the seven-day self-report physical activity question revealed that 46.0% of women from the overall sample reported meeting guidelines for physical activity in the previous week. This was 46.5% of women in age group 1 (25-45 yrs) and 53.5% of women in age group 2 (46-65 yrs). It is concerning that over half of the sample reported failing to participate in sufficient moderately intense or vigorous physical activity to safeguard their 50
cardiovascular health. In 2003/2004, the percentage of Welsh women over the age of 16 years reporting inadequate physical activity was 78.0% (WHS, 2003/04) and in 2007, this was 77.0%. However, there is some ambiguity in relation to the results of previous WHSs because account was not taken of women reporting fewer sessions per week but at a vigorous intensity. Women participating in three vigorous sessions per week were categorised as insufficiently active because they did not meet the criteria for five sessions a week. In the present study women reporting five or more moderately intense sessions, a combination of five moderate and vigorous sessions, or three vigorous sessions per week were classed as fully active in keeping with the guidelines of the AHA and ACSM (Haskell et al., 2007). Physical activity status was examined against marital status to ascertain any relationship. Chi-square analysis revealed no significant association between any of the marital categories and participation in sufficient physical activity. It would appear that whilst it might be assumed that single, separated or divorced women might have less family responsibility and therefore more free time, for reasons not understood, a single status was not associated with any difference in reported physical activity. Similarly, physical activity status was examined against employment status and no relationship was established with any of the employment categories. It is naïve to assume that employment or marital status predict the availability of free time for women to engage in physical activity, or to assume that women will make a conscious choice to engage in physical activity if free time allows. The WSFF (2007) made the point that there are so many competing attractions for women‘s leisure-time, the challenge for providers, is to ensure physical activity pursuits are attractive to women, enjoyable and socially rewarding.
2.4.8 PERCEPTIONS OF PHYSICAL FITNESS (Q 26) Participants were asked to rate their physical fitness as ‗excellent‘, ‗very good‘, ‗reasonable‘, ‗poor‘ or ‗very poor‘. Six hundred and twenty-six women (86.5%) rated their physical fitness as ‗excellent‘, ‗very good‘ or ‗reasonable‘. There were 295 women (81.7%) in group 1 and 331 women (91.2%) in group 2. This is concerning because 54.0% of women were not meeting current guidelines for physical activity, suggesting a mismatch between women‘s perceptions of their physical fitness and the reality of their activity levels. For many women in the present study, it would be unlikely that their physical fitness was ‗excellent‘ or ‘good‘ if they reported as sedentary, or only partially active and misperceptions of their personal physical fitness might be leading these particular women to believe that the risks associated with a lack of physical activity do not apply to them personally.
51
2.4.9 PREFERRED SOURCES OF INFORMATION (Q 14) The participants were asked about their primary sources of cardiovascular health information and whether they had ever talked to their practice nurse or GP about their cardiovascular health. They were asked to indicate their preferred sources of information about cardiovascular health and could respond to more than one source, or state other sources. Figure 5 illustrates the sources of information accessed by women in the present study.
30
Percentage (%)
25 20 15 10 5 0 Watching TV
Talking to Health Women's Talking to Newspapers my doctor promotion magazines others leaflets Sources of information about CVD
Figure 5, Sources of information about cardiovascular disease (n = 724) Television and women‘s magazines featured heavily in informing women across both age groups about cardiovascular health issues. There might be reason to question the relevance behind some generalised media advice because CVD risk is unique to the individual and whilst general health messages are important to cascade, health care professionals might more appropriately address individual level of risk. In the present study only 10.0% of women reported accessing information from their GP and only 11.2% from reading health promotion leaflets prompting concern about the reliability of the health information women prefer to access. In response to an item asking participants to state their preferred source of cardiovascular health information if not on the given list of options, the majority of women stated that newspaper articles were the most popular, confirming the powerful influence of the media in cascading health information to women. Surprisingly, no women reported accessing information via the internet. This is interesting in light of the resources available from the BHF and which have sections specifically relating to women.
52
2.4.10 KNOWLEDGE AND PERCEPTIONS OF CARDIOVASCULAR DISEASE
The next section addresses the responses to the knowledge based questionnaire items. A breakdown of responses is presented in Appendix D. The questionnaire contained 50 knowledge items based on information available within the public domain. For the purpose of assessment a score of 40 or more correct responses indicated a ‗good‘ knowledge base, a score of 25-39 as ‗average‘, and a score of less than 25 as ‗poor‘. Only 15.0% of the sample scored 40 or more. Seventy-seven percent scored between 25 and 39, and 8.0% scored less than 25. In addition to assessing cardiovascular knowledge the participants were requested to state how well informed they thought they were about CVD (Q 15). The results are illustrated in Figure 6.
70
percentage (%)
60 50 40 30
g1 (25-45 yrs)
20
g2 (46-65 yrs)
10 0 I am very well informed
I am informed
I am not at all informed
Women's perceptions of how well informed they were about CVD
Figure 6, Participants‘ perceptions of how well they were informed about cardiovascular disease (n = 724) Only 17.5% of the overall sample perceived themselves as ‗very well informed‘ about CVD with more younger women than older women reporting they were ‗not at all informed‘ suggesting a need to target health messages specifically to address gaps in knowledge, and to consider channels of communication most appropriate for reaching women in the age groups of the study. In light of the diverse lifestyles women lead and the number of communication channels available, it would appear prudent to consider using multiple channels to disseminate information about CVD and how it can be prevented.
53
2.4.11 CARDIOVASCULAR DISEASE, BREAST CANCER AND CAUSE OF DEATH IN THE UK (Q17) Participants were asked to ‗agree‘, ‗disagree‘ or state they ‗did not know‘ if each of five general statements about CVD were true. A majority (70.0%) agreed that CVD is the main cause of death for men and women with slightly more women aged between 46 and 65 years responding correctly. Confusion was apparent regarding whether breast cancer causes more deaths annually in women than CVD. Fifty percent of women responded correctly affirming that it is CVD. The erroneous belief that breast cancer causes more deaths has implications for the prevention of CVD in that it could detract women‘s attention away from the need to address the cardiovascular risk factors that pre-dispose them towards development of the disease. Health information currently in the public domain emphasises that the same lifestyle behaviours are implicated in the development of CVD and some forms of cancer. It therefore would appear logical to continue cascading this health message together with strategies for prevention such as body-weight management and leading a physically active lifestyle.
2.4.12 KNOWLEDGE OF DIABETES (Q17) Two hundred and forty women (33.0%) were unaware that diabetes increases women‘s risk of CHD. Furthermore, 85.0% of women were unaware that diabetes is more likely to cause CHD for women compared with men. This indicated a need to heighten awareness of the relationship between overweight and obesity, a lack of physical activity, and subsequent development of type 2 diabetes particularly in light of the high prevalence of overweight and obesity within the sample. Current trends indicate that by 2050 the majority of the UK population will be overweight or obese with attendant health problems (Foresight, 2007). The results of the present study clearly indicate a need for targeted health messaging to women regarding the consequences of overweight and obesity and development of type 2 diabetes.
2.4.13 KNOWLEDGE OF THE CAUSES OF CARDIOVASCULAR DISEASE (Q 18) The participants were asked about what they perceived as the the causes of CVD. The results are illustrated in Figure 7.
54
Percentage (%)
120 100 80 60 40 20 0
Causes of CVD
Figure 7, Women‘s perceptions of the causes of CVD (n = 724) Responses to questions regarding causes of CVD were similar for both age groups and a breakdown is shown in Appendix D. Awareness of a fatty diet, and overweight or obesity as causes of CVD was high with 97.8% of the overall sample responding correctly. It is therefore intriguing that 46.3% of the sample reported as overweight or obese, and 8.5% of overweight or obese women perceived their weight to be in a healthy range. Ninety-eight percent of women aged between 25 and 65 years identified smoking as a cause of CHD although there was a gap in knowledge related to smoking and peripheral artery disease with 57.0% of women unaware of the risk factors. Awareness of inactivity as a risk factor for CVD was high with 682 women (94.3%) rating it as ‗important‘. Only 5.7% of women stated either that physical activity was ‗not important‘ or they ‗did not know‘. Paradoxically, 54.0% of women reported as insufficiently physically active. Participants were asked if they thought that in general, physically active women risk a heart attack as they get older and 83.0% correctly responded that they ‗did not‘. Moreover, 95.0% of women knew that in general it is not risky for women to start an exercise programme.
Colds and flu in older age do not constitute a major risk factor for the development of CVD. Older age is however associated with more chronic ill health (Spirduso et al., 2005). Forty two percent of women either did not know, or thought that colds and flu were risk factors for CVD suggesting a gap in knowledge. Just over half of the sample identified that air pollution is not a major cause of CVD. One hundred and thirty-eight women (19.1%) thought that not drinking enough water could cause CVD, 341 women rated it as ‗not important‘, and 245 women (33.8%) stated that they ‗did not know.‘
55
The theory that emotional stress is causal is controversial. The BHF public information internet site titled ‗Women and Heart Disease‘ (BHF, 2009) does not state that stress is a major cause of CVD, but includes information to explain how stress can potentially be injurious to the cardiovascular system, and how to avoid it. Women in this present study revealed diverse perceptions. Sixty-one percent affirmed emotional stress as an important cause of CVD and it would appear negligent to ignore its potential to exacerbate risk. Women who suffer chronic stress in the form of depression, anxiety, or chronic life stress, or who might exhibit psychological traits associated with coronary events will likely find it more challenging to adhere to a healthy lifestyle. It would appear that public information available from the BHF (2012) is relevant to the lives of women, who, once informed, are likely to be more aware of the potential impact of emotional stress on their health and well-being.
2.4.14 THE INITIATION OF CARDIOVASCULAR DISEASE (Q 19) The participants were required to indicate a stage of development during which they believed the process of atherosclerosis begins. Four hundred and three women (58.01%) correctly indicated childhood or adolescence. Four hundred and seventy-one women (65.0%) aged between 46 and 65 years responded correctly compared with three hundred and seventy-six women (52.0%) aged between 25 and 45 years.
2.4.15 KNOWLEDGE OF SYMPTOMS OF MYOCARDIAL INFARCTION (Q 20) The majority of the sample identified chest pain (97.4%) as a symptom of a myocardial infarction. This was 97.5% for women aged between 25 and 25 years, and 97.2% for women aged between 46 and 65 years. Similarly, 87.0% identified difficulty with breathing with 88.6% of women aged between 25 and 45 years and 85.4% of women aged between 46 and 65 years responding correctly. Ninety-two percent identified a pain in the arm with 91.4% of women aged between 25 and 45 years and 93.7% of women aged between 46 and 65 years responding correctly. Whilst this indicated a sound knowledge base, 565 women (70.0%) were unaware that pain can radiate to the jaw and 453 women (40.0%) were unaware that pain can radiate to the upper back. Over 40.0% of women either did not know, or mistakenly believed that a rash on the chest is a symptom of a myocardial infarction. Three hundred and fifty-seven women (49.3%) did not associate a feeling of nausea as a symptom. These results suggested a need to address identified gaps in women‘s knowledge base so that they are more likely to recognise symptoms in themselves and others.
56
With regard to physical symptoms that might be experienced prior to infarction, over 58.6% of the sample were unaware that recurring indigestion type discomfort, unusual tiredness (57.8%) or increased anxiety (54.7%) can be indicative of myocardial difficulty (Douglas and Ginsburg, 1996). There was little difference in responses between age groups. There may be some mitigation in this poor profile of women‘s knowledge because these atypical symptoms are unlikely to be experienced in isolation. Overall, however the results suggested that women in the present study lacked some necessary knowledge that would enable them to take urgent action should they experience these symptoms. Gaps in knowledge are important to address in light of the suggestion by Owen-Smith et al. (2002) that women who are experiencing symptoms might avoid visiting their GP and self-diagnose their symptoms or attribute chest pain to indigestion or other less malign causes than myocardial difficulty.
2.4.16 KNOWLEDGE OF STROKE INJURY (Q 21) Seventy-four women (10.2%) incorrectly stated that a stroke occurs in the heart. This was 45 women (6.2%) in group 1 (25-45 yrs) and 29 women (4.0%) in group 2 (46-65 yrs). Although they were low percentage incorrect responses, they suggest an opportunity to address this important gap in knowledge. A recent televised campaign to heighten awareness of the symptoms of stroke by the DoH (2010) used both male and female actors and was in contrast to the campaign of the BHF (2007) titled ‗Doubt Kills‘ where only a male actor demonstrated warning signs of a myocardial infarction. Use of both male and female actors is prudent in that women‘s knowledge about stroke is therefore not dependent on having to extrapolate information from a male-oriented health message.
2.4.17 KNOWLEDGE OF STROKE SYMPTOMS (Q 22) Participants were asked to indicate the symptoms associated with experiencing a stroke. Ninety-three percent of women from the overall sample knew that poor co-ordination and a feeling of numbness were symptomatic. The majority of women in the present study also correctly recognised slurred speech (96.8%) and problems with vision (77.1%). The gaps in women‘s stroke knowledge related to a loss of bowel control, bad temperedness, and a rash on the chest that are not regarded as stroke symptoms although a minority of women perceived these as important. It is concerning that only 467 women (64.5%) correctly stated that a severe headache might be a symptom of stroke. As with a myocardial infarction, although the symptoms of stroke do not usually present in isolation it would benefit women to have pertinent information so that they avoid a delay in seeking medical treatment should they experience symptoms. 57
2.4.18 KNOWLEDGE ABOUT HIGH BLOOD PRESSURE AND PHYSICAL ACTIVITY (Q 9 and Q 24) Six hundred and fifty-seven women (90.7%) knew that high blood pressure is a risk factor for CVD. Four hundred and twenty women (58.0%) in group 1 (25-45 yrs) and 485 women (67.0%) in group 2 (46-65 yrs) identified a healthy measure of blood pressure as 130/80 mm Hg. Therefore, knowledge was slightly better in the older age group. Ninety percent of women knew that taking more exercise could lower blood pressure and again this was a slightly better result in older women with 87.3% of women in group 1 (25-45 yrs) and 93.1% of women in group 2 (46-65 yrs) responding correctly. It is interesting that based on these results although the knowledge base about high blood pressure could be considered sound, 46.0% of women had reported insufficient physical activity levels in the previous week and 47.3% had not undertaken any screening for high blood pressure although a majority of women (80.5%) knew that hypertension is asymptomatic. This was 78.4% of women in group 1 (25-45 yrs) and 82.6% of women in group 2 (46-65 yrs). Overall, the results suggested that although women were quite knowledgeable about high blood pressure and the benefit of exercise many were not relating this knowledge to their own health behaviour.
2.4.19 CHOLESTEROL AND HEALTHY FUNCTIONING OF THE BODY (Q 24) Five hundred and forty-nine women (75.9%) correctly identified a healthy cholesterol measure as under a recommended 5.0 mMol L-1. There was a noticeable difference in correct responses between the younger group (69.0%) and the older group (82.9%). Ninetyfive percent of women were aware that too much cholesterol could cause arterial injury. Thirty percent of women were unaware that cholesterol is necessary for healthy functioning of the body and 37.4% did not know that the body manufactures cholesterol. Some confusion arose in response to an item asking whether cholesterol could be lowered by exercise as well as diet, with 12.5% of women across the sample either incorrectly stating that it could not, or they did not know, suggesting that women might benefit from more detailed information about the role of cholesterol in health and disease. A sound level of knowledge was demonstrated regarding high cholesterol as a cause of CVD with 678 women (93.6%) responding correctly. It is however, disconcerting that only 264 women (36.5%) had sought to have cholesterol screening in the past five years.
58
2.4.20 OVERWEIGHT, OBESITY AND RISK OF DIABETES (Q 24) Ninety percent of women knew that overweight and obesity were risk factors for diabetes and CHD and this is concerning because 445 women (62.9%) reported as having abdominal adiposity, and 328 (51.9%) were overweight or obese as classed by BMI. It is even more concerning because 8.5% of overweight or obese women underestimated the risk associated with their BMI. Moreover, only 105 women (14.5%) knew that diabetes is more likely to cause CHD in women than men suggesting a need for gender oriented health messages to address this issue.
2.4.21 PERCEPTIONS OF RISK OF CARDIOVASCULAR DISEASE (Q 27) The participants were asked to rate their risk of CVD compared with other women of the same age. The results are illustrated in Figure 8.
60
Percentage (%)
50 40 30 20 10 0 Higher
Slightly more
Same as other women
Less than other women
Women's perceptions of their risk of CVD compared to peers
Figure 8, Women‘s perceptions of their cardiovascular risk compared to peers (n = 720) Twenty-seven percent of women perceived their risk of CVD to be ‗more‘, or ‗slightly higher‘ compared to peers. Whilst one method of evaluating perceptions of risk is to enquire about perceived susceptibility compared with peers, the present study highlighted a difficulty with this reasoning. It requires the individual to have a basic knowledge about susceptibility within the peer group in order to make a decision about personal susceptibility. If however, the individual has a lack of awareness of factors that would heighten susceptibility, or does not recognise these factors as relating to their own risk, they are unlikely to arrive at a meaningful judgement. 59
2.5.1 GENERAL CONCLUSIONS TO STUDY ONE The results of the present study highlighted five issues relating to concerning levels of physical activity, the prevalence of overweight and obesity, misperceptions of personal cardiovascular health risks, inadequate up-take of screening services for blood pressure and cholesterol, and gaps in women‘s knowledge base of CVD.
From a positive perspective, the results suggested that the majority of women had some sound knowledge of CVD although their behaviour in relation to physical activity, and accessing screening services did not always reflect this knowledge. Moreover, there was a low, but still concerning prevalence of smoking (15.1%). In terms of preventing CVD, these issues are important to address in the target audience of women aged between 25 and 65 years.
2.5.2 PHYSICAL ACTIVITY Only 46.0% of women reported as sufficiently physically active during the previous week according to physical activity guidelines (Haskell et al., 2007). Three distinct sectors of the sample could be identified consisting of sedentary women, those who were physical active but not sufficiently enough in terms of frequency, and those who were fully active. This in turn suggested the possibility of targeting health messages that specifically address the needs of women according to their identified stage of physical activity behaviour based on the transtheoretical model (Prochaska and DiClemente, 1983). For example, the targeting of sedentary women through messages that convey the benefits of physical activity and that suggest ways to identify and manage barriers to an active lifestyle. The present study did not address preferred modes of physical activity although some women were attending organised exercise session such as Zumba or step aerobics. In order to increase the frequency of physical activity the health message that brisk walking can also confer many health benefits might be particularly cogent in light of the relative ease with which it can be incorporated into daily life and because it does not incur added financial cost.
When asked to rate their fitness, many women probably underestimated, particularly women aged between 46 and 65 years with 91.0% reporting their fitness as ‗excellent‘, ‗very good‘ or ‗reasonable‘ which did not reflect their reported physical activity levels. Misperceptions of physical fitness might relate to a lack of knowledge about physical activity guidelines in terms of the advised frequency, duration, intensity and mode of physical activity necessary to improve and maintain physical fitness. The present study did not investigate this aspect of 60
women‘s knowledge. Overall, the findings suggested a need to target physical activity on multiple levels as encouraged through the on-going WAG Campaign ‗Health Challenge Wales‘ (2004) and through objectives of ‗Creating an Active Wales‘ (WAG, 2009) to make physical activity a natural part of people‘s lives.
2.5.3 OVERWEIGHT AND OBESITY The overall prevalence of overweight and obesity based on BMI and abdominal adiposity estimated by WC was 46.3% and 62.9% respectively. The women most in need of weight reduction were the least active; highlighting a need to target women who are both inactive and overweight or obese as a priority within intervention programmes. The results demonstrated that some women were unable to make correct judgements about their bodyweight in relation to associated health risks suggesting a need for heightened awareness of this issue. Risk of long-term illness including CVD was assessed by a combination of BMI and WC according to NICE guidelines (2006). Forty-one percent of women were categorised at various levels of increased risk and this is an obvious concern because of the increased risk of CVD.
Overweight and obesity not only exert an influence at an individual level but also on an already stretched NHS. It is timely therefore that the All Wales Obesity Pathway (DoH, 2009) has been established so that health boards, local authorities, and other agencies can work together to combat overweight and obesity at individual, community and population levels. The Pathway (DoH, 2009) adopts the stance that interventions should be long-term, based on a
societal approach, organised on multiple levels, and through multiple agencies. Furthermore, intervention should occur in various settings and address broad determinants of health. The results of the present study add weight to the relevance of this comprehensive multi-level, multi-agency approach to tackling overweight and obesity in women aged between 25 and 65 years.
2.5.4 SCREENING FOR HIGH BLOOD PRESSURE AND CHOLESTEROL Only 264 women had accessed cholesterol screening (36.5%), and 381 women had accessed screening for hypertension (52.7%). More of the older women than younger women had accessed these health checks. Cardiovascular risk can escalate because high cholesterol and hypertension silently cause injury to the vasculature (Gordon et al., 2006). Therefore, screening is a crucial preventative step that women can take. Health messages drawing attention to the enhanced risk of CVD caused by high blood pressure and high 61
cholesterol are relevant in efforts to prevent CVD, and continuing campaigns by the British Blood Pressure Association and the BHF appear salient in the drive to encourage women to be more pro-active in accessing these measures through their local GP surgery or pharmacy. The results of the present study indicated that women‘s knowledge base about the risks associated with high blood pressure and high cholesterol was sound but many women did not act on their knowledge and apply it to their own situation by accessing screening.
2.5.5 SOURCES OF HEALTH INFORMATION Improving women‘s knowledge and awareness of CVD requires communicating health messages to address the major modifiable risk factors of physical inactivity, smoking cigarettes, overweight and obesity, high cholesterol and high blood pressure (ESC, 2006). The task is made more complex by the nature of the preventative action required to lower the risk associated with each major risk factor, and because some risk factors act synergistically with others. For example, the overweight or obese person is more likely to develop high blood pressure and high cholesterol (National Audit Office, 2010). Therefore, a dilemma facing those who develop health messages about CVD prevention is whether to target individual risk factors and their link to the development of CVD, or to develop more holistic health messages to convey that risk factors act synergistically.
Some preventative health messages need to target long-term behaviours such as physical activity and healthy eating, and other messages relate to a need for urgent action as in the recognition of the symptoms of a myocardial infarction or stroke. The knowledge-based questions included in the survey in the present study were developed from information available in the public domain. Sources consisted of the British Hypertension Society and NHS printed information about the health risks of smoking cigarettes, the ‗Measure-Up‘ Campaign (Diabetes UK, 2006), printed information and internet resources available from the BHF, and printed information from ‗Health Challenge Wales‘ (WAG, 2004). When developing the knowledge items for the survey, it was acknowledged by the primary investigator that women might also source information from their GP, family and friends, and through the media. The results revealed that few women had sourced health information from talking to a GP (10.0%), or from reading health-promotion literature (11.2%). Dominant sources of health information included television (25.4%), and magazines and newspapers (41.6%). This suggests opportunity for health care providers to be more pro-active in efforts to raise women‘s knowledge of CVD and to signpost them to pertinent sources of information because there might be a gap in their awareness of these sources and in their motivation to access them. Communication between health care providers and individuals is an accepted 62
traditional method for sharing information about health issues, and can take place in a variety of settings. It is advantageous because it is personalised. For example, GPs are able to make use of computer-aided tools for evaluation of individual risk of CVD. The Third Joint European Societies‘ Task Force on CVD Prevention in Clinical Practice (2003) advised the use of a risk prediction system known as Systematic Coronary Risk Evaluation (SCORE) to assess individual level of risk, and to provide a platform for discussion between the GP and patient. However, a drawback is that risk evaluation within the context of the GP surgery might be time intensive and costly, particularly if there is a need to address several risky behaviours.
Prochaska and Sallis (2006) pointed out that traditional forms of health messaging adopted a ‗one size fits all‘ approach although more recent efforts have focused on the development of targeted communication directed at a specific segment of the population. Health messaging to an identified audience such as women in the present study would then take account of the source of the message, for example, a celebrity or other model of the behaviour, the delivery channel, for example, electronic or personal communication, and the context, for example, a worksite physical activity programme. A partial solution for disseminating targeted information about preventing CVD might consist of the use of interactive technology that can provide additional information accessible to the identified audience.
The DoH (2011)
recognised that in the face of competing interactive technological advances, the initial ‗Change4Life‘ web site was information driven and this was not attractive compared with other more user-friendly interactive internet sites, prompting the decision to explore the viability of changes to a more interactive format. However, on the basis that not everyone will make deliberate efforts to seek out credible sources of health information women in the target audience might be better informed if health professionals guided them to information that they could identify with, that they could access at convenient times, and that was personalised.
2.5.6 GAPS IN WOMEN’S KNOWLEDGE OF CARDIOVASCULAR DISEASE The strengths and weaknesses of women‘s knowledge base about CVD in the present study suggested a need for targeted gender-oriented health information. For example, the majority of women (70.0%) knew that CVD is the main cause of death in the UK but demonstrated confusion regarding whether CVD or breast cancer is women‘s greatest health threat. This is a concern as evidenced by mortality statistics. For example, 100,237 women died from CVD, and 11,995 from breast cancer in 2007 (BHF, 2007). Therefore, without undermining efforts
63
to raise awareness of breast cancer, there is also a need for women to have access to knowledge about CVD.
There was also a gap in knowledge about obesity-related medical conditions amongst women in the present study. A third of women did not know that diabetes increases a women‘s risk of CHD and 86.5% were unaware that diabetes was more likely to cause CHD in women compared to men. This gap in knowledge has connotations for a population of women at increased risk of CVD because of a concerning prevalence of overweight and obesity, and presents a further opportunity for gender-oriented health messages.
The results of the present study suggested a lack of awareness regarding the symptoms of a myocardial infarction in women. The majority of women did not know that pain could radiate to the jaw, or to the back, or that nausea, or recurring indigestion type pain or unusual fatigue can be symptomatic of myocardial difficulty. Women who do not recognise these symptoms in themselves, or other women, are at risk of delaying access to medical diagnosis and treatment. Health messages could be channelled through media sources to replicate the recent campaign of the BHF titled ‗Doubt Kills‘ (2007), thereby alerting women as well as men to the symptoms of a myocardial infarction and the need to seek an immediate medical diagnosis.
2.5.7 PERCEPTIONS OF RISK It is concerning if women incorrectly perceive they are physically active enough for their health, if they are unable to make correct judgements about their body-weight status in relation to health risks, and if women who smoke cigarettes do not perceive their heightened risk of CVD. The results of the present study highlighted these misperceptions which are of concern in light of the fact that only a minority of women reported that they were ‗very well informed‘ about CVD. Enhancing personal awareness of CVD risk is however a challenge because of a need to channel communication in such a manner that the information conveyed to the target audience promotes behaviour change where necessary, without a desensitising or intimidating effect.
2.5.8 SUMMARY The results of study one highlighted a need for action on several fronts. These pertained to gaps in women‘s knowledge of CVD, poor uptake of screening, smoking behaviour, inactivity,
64
and overweight and obesity. Atherogenic CVD usually results from exposure to multiple risk factors. Therefore, health messages about prevention require consideration of how to deliver multiple messages in a manner that captures women‘s attention and suggests the means by which behaviour change might be accomplished with confidence. This suggestion is cautiously made because prevention of CVD also necessitates environmental structures to support behaviour change and maintenance if the social marketing approach to health adopted by the WAG is to be successful.
In 2006, recommendations by the European Society of Cardiology included strong action to increase women‘s knowledge and awareness of how to prevent CVD (Stramba-Badiale et al., 2006). Furthermore recommending that research should address surveys of women‘s experience of CVD in areas where data is lacking. The results of the present exploratory study have served to indicate where attention might be focussed in order to contribute to the prevention of CVD in women aged between 25 and 65 years.
65
ABSTRACT
The aims of this study were to: (i) determine the accuracy of self-perceptions of aerobic fitness in a sample of apparently healthy women aged between 25 and 65 years resident in one locality of the UK (n = 58) and (ii) to investigate aerobic fitness and physical activity levels, and overweight and obesity in light of thresholds for cardiovascular disease (CVD) risk. Following pre-exercise screening, participants performed a Chester Step test (CST) to assess aerobic capacity, completed a seven-day physical activity recall question, and physical measurements included stature and mass to calculate body mass index (BMI), and waist circumference (WC). The participants‘ physical activity levels were categorised according to current guidelines (Haskell et al., 2007). Waist circumference measures were categorized according to criteria of the International Diabetic Federation (2005). The risk of long-term health problems and CVD was evaluated by combining WC and BMI in accordance with guidelines of the National Institute for Health and Clinical Excellence (NICE, 2006). Aerobic fitness scores were recorded in ml·kg-1·min-1 and converted into a metabolic equivalent (MET) for comparison with thresholds for CVD risk established by Blair et al. (1989), Gulati et al. (2003), Kokkinos et al. (1995) and Kodama et al. (2009). An aim of the analysis was to establish how well the participants perceptions of their aerobic fitness reflected measured values obtained by use of the CST (Sykes, 2005) and based on the categories ‗excellent‘, ‗good‘, ‗average‘, ‗below average‘ or ‗poor‘. The statistical method employed to investigate the agreement between measured aerobic scores and perceived aerobic scores was the 95% limits of agreement (95%LoA) method recommended by Bland and Altman (1986). Outcomes were expressed as the mean difference plus and minus (±) 1.96 standard deviations. The results highlighted issues regarding women‘s perceptions of their fitness, physical activity and measured aerobic fitness, and overweight and obesity. Women were moderately accurate at estimating their aerobic fitness but it is unlikely that this would provide useful information to facilitate discussion between women and health-care providers about the role of aerobic fitness in CVD prevention. The majority of women partially met physical activity guidelines (75.8%). Fifty-seven percent were either overweight or obese. The criteria applied for combined BMI and WC to evaluate heightened risk of CVD and other long-standing illness (NICE, 2006) found 50.0% of women over advised thresholds indicating increased risk of long term illness including CVD. Nineteen percent of women had an increased risk of all-cause mortality and cardiovascular events based on an aerobic threshold of 7.9 METs suggested by Kodama et al. (2009). The study was limited by a small sample size and voluntary participation. Furthermore, participants‘ were asymptomatic for CVD, other long-standing disease, and were not pregnant. Therefore, generalisation of the results to a wider population of women is not advisable. Moreover, the aerobic fitness assessment consisted of a sub-maximal step test and as such, a degree of inaccuracy of measurement was to be expected. Further research could replicate the aims of this study in order to strengthen the knowledge base in relation to women‘s measured and perceived fitness, and to monitor physical activity, aerobic fitness, and overweight and obesity.
66
3.1.1 RECOMMENDATIONS FOR PREVENTING CARDIOVASCULAR DISEASE In 2007, the European Society of Cardiology (ESC) recognised the critical importance of cardiovascular disease (CVD) prevention through lifestyle and environmental changes and endorsed initiatives by major organisations and governments to take preventative action. Five measures were considered of paramount importance. These consisted of the avoidance of tobacco, adequate physical activity of at least 30 minutes a day, healthy food choices, avoiding being overweight, blood pressure below 140/90 mm Hg and, total cholesterol below 5 mMolL-1.
In more detail, the advice for asymptomatic individuals to ‗be active for 30
minutes daily‘ necessitated the mode of physical activity most likely to improve aerobic fitness; for example, brisk walking, swimming, and aerobic dance (ESC, 2007). The present study addressed issues relating to two of the measures for risk reduction consisting of women‘s physical activity and aerobic fitness and the prevalence of overweight and obesity in a sample of Welsh women. Furthermore, it addressed women‘s perceptions of their aerobic fitness compared with measured aerobic fitness.
3.1.2 WOMEN’S MEASURED AND PERCEIVED AEROBIC FITNESS Although the advice for 30 minutes of daily exercise appears to be a simple strategy for reducing risk of CVD, there continues to be a high prevalence of physical inactivity in the Welsh population of women (Welsh Health Survey, 2009). Attempts by government and major organisations to reverse the current situation are comprehensive and operate on multiple levels. For example, consciousness raising through mass media campaigns, individualised exercise prescription as in general practitioner (GP) referral schemes, and subsidised access to some physical activities for older people. If however, women do not perceive that they would benefit from increased levels of physical activity because they think they are already physically fit enough for their health, they are unlikely to consider acting on the above advice.
In study one of this thesis 86.5% of women aged between 25 and 65 years had subjectively rated their fitness as ‗excellent‘, ‗good‘ or ‗reasonable‘. However, in light of their self-reported physical activity levels these were unlikely to represent accurate self-perceptions because 54.0% reported insufficient physical activity to meet current physical activity guidelines (Haskell et al., 2007). Evidence of misperceptions of fitness was reported in a survey by the Women‘s Sport and Fitness Foundation (WSFF, 2007, n = 1000) where 60.0% of women believed they were already sufficiently physically active although 80.0% had reported low activity levels. However, there is a paucity of information about Welsh women‘s perceptions 67
of their fitness and there is value therefore, in investigating their perceptions in order to better understand the reasoning underpinning their physical activity behaviour.
3.1.3 THE VALUE OF ASSESSING AEROBIC FITNESS AND PHYSICAL ACTIVITY There is a relationship between physical activity and aerobic fitness in that the latter is mainly, although not solely, determined by physical activity performed in previous weeks or months (Blair et al., 2001). Data from the Aerobics Centre Longitudinal Study (n = 7080) concluded that low fitness is an independent predictor of all-cause mortality in men and women of similar strength to cigarette smoking and may be a stronger predictor than elevated blood pressure or cholesterol. Furthermore, Blair et al. (1996) concluded that whilst physical inactivity and low fitness increase the risk of CVD, habitual physical activity and higher fitness are both protective, and impact favourably on other major risk factors.
Whilst measurement of blood pressure and cholesterol are health assessments for adults easily accessed via a GP surgery or pharmacy, it is less so for the assessment of physical activity or aerobic fitness. This is despite accumulated evidence that they are indicators of cardiovascular health or disease. In a clinical context, the assessment, counselling and support of patients in the maintenance of physical activity are regarded as core tasks for physicians and other health workers engaged in the prevention of CVD by the ESC (Graham et al., 2007). Furthermore, this recommendation was recognized in ‗Creating an Active Wales‘ (Welsh Assembly Government [WAG], 2009) so that GPs and other health professionals were to be supported in their efforts to advise and encourage patients to increase their physical activity levels. This positive recommendation was important because the assessment of an individual‘s physical activity in a clinical context can be valuable in determining whether they are sufficiently active to accrue health benefits (Graham, et al., 2007). However, the use of self-reported physical activity is accepted to have limitations, and it is therefore pertinent to consider the clinical usefulness of measuring aerobic fitness in asymptomatic individuals in light of evidence for its prognostic value in determining the risk of CVD (LaMonte et al., 2000; Gulati et al., 2003; Sui et al., 2007; Lyerly et al., 2009). The issue is not without contention because from the clinical perspective of risk evaluation, there is not a consensus of agreement about aerobic fitness thresholds to indicate risk of CVD (Kodama et al., 2009).
Fitness assessment can however establish a baseline of fitness and the results compared to existing norms. For example, norms have been developed in the United States by the 68
Cooper Aerobics Centre (2009) whilst in the UK the Chester Step test ([CST], Sykes, 2005) is based on norms developed from large data-sets obtained in the UK, matched to international data-sets. In asymptomatic women who do not have a lifestyle conducive to cardiovascular health but who are not clinically classed as at high risk of CVD, and when it is difficult to establish physical activity levels with certainty, the results of sub-maximal aerobic fitness assessment can form a basis for discussion with the health care provider. Additionally, the results of fitness assessment can form a basis of appropriate exercise prescription.
Moreover, measurement of subsequent incremental gains in fitness can
provide the necessary feedback to contribute towards a motivational strategy for encouraging the individual to continue with an exercise programme (Sykes, 2005).
Although it is known that many women in Wales have low levels of physical activity little is known about their aerobic fitness and it would be unsafe to assume that by association Welsh women also have poor aerobic fitness. On this basis, an intention of the present study was to assess aerobic fitness in a sample of asymptomatic women aged between 25 and 65 years. Furthermore, the intention was to compare the results with various aerobic fitness thresholds relating to the risk of CVD and all-cause and CVD mortality.
3.1.4 THE RISK OF OVERWEIGHT AND OBESITY A further concern resulting from the findings of study one was the prevalence of overweight and obesity in the sample. This was 46.0% based on body mass index (BMI) estimated from self-reported body-weight and height, and 63.0% based on self-reported waist circumference (WC). This was even more concerning because there was a decreasing involvement in physical activity associated with increasing BMI. Fifty-four percent of ‗under‘ or ‗normal‘ weight women (< 18.5 and 18.5 to 24.9 kg m-2), 33.0% of ‗overweight‘ women (25 to 29.9 kg m-2), and only 13.0% of ‗obese‘ women (30.0 kg m-2 or more) reported meeting physical activity guidelines (Haskell et al., 2007). Successive Welsh Health Surveys (WAG, 2007/2009) have estimated the prevalence of overweight and obesity in women based on self-reported height and body-weight to calculate BMI. However, it is known that women tend to over-report height and underestimate body-weight leading to a potential for error (Danubio et al., 2008). Moreover, BMI as a measure of obesity is not without its limitations in that it needs to be interpreted with caution because it is not a direct measure of adiposity and fails to take account of proportional composition or body fat distribution. However, BMI has a moderate association with body fat and disease risk so that as BMI increases so does the health risk (McArdle et al., 2006). Within a clinical setting, it is easy to calculate and the result can be compared against categorisations of risk (World Health Organisation [WHO], 2004). 69
The pros and cons of using BMI to indicate risk of ill health serves to highlight an issue for consideration in that the choice of measurement should reflect the purpose of that measurement. For example, weight scales are commonly used as a measure of ‗overweight‘ in commercial slimming clubs. However, a measure of body-weight provides little in the way of information about the level of health risk associated with over fatness (McArdle et al., 2006). Measurement of body-weight might even be considered misleading in that it provides limited and unreliable information about body composition, does not measure adiposity, does not assess components of body composition, and furthermore, weight-height tables are based on data primarily from white populations and, unless taking into account frame size, the measure is unreliable (McArdle et al., 2006). Within a health-oriented setting such as a GP surgery or obesity clinic more accurate measures of excess adiposity might better serve the aim of assessing any associated health risk. Other measures exist for example, hydrostatic weighing but would be impractical. Therefore, measures of adiposity must be chosen in light of the evidence base for their clinical usefulness in assessing risk, and the practicality of their use with a clinical setting. In light of this, three measures were used in the present study consisting of BMI, WC and combined BMI and WC. The latter combined measure was in accord with guidelines of the National Institute for Health and Clinical Excellence (NICE, 2006). Objective confirmation of the overweight and obesity status of women aged between 25 and 65 years facilitated discussion and contributed to what is currently known through self-reported information published in Welsh Health Surveys.
3.1.5 AIMS OF STUDY TWO On the basis of the issues highlighted above the aims of the present study were as follows:
i.
To determine the accuracy of self-perceptions of aerobic fitness in a sample of apparently healthy women aged between 25 and 65 years.
ii.
To investigate aerobic fitness and levels of physical activity and to examine these in light of the risk of CVD.
iii.
To investigate the prevalence of adiposity and to examine this in light of the risk of CVD.
For clarity of understanding the term ‗cardiorespirarory fitness‘ is sometimes used interchangeably in the present study with the term ‗aerobic fitness‘. The latter term is more commonly used by women to describe their fitness and is the term predominantly used in the present study.
70
3.2.1 INTRODUCTION TO THE REVIEW OF LITERATURE The review begins with a discussion about the potential accuracy of women‘s perceptions of their aerobic fitness and why this might be an important issue related to the prevention of CVD. This is followed by a review of studies pertaining to the risk of CVD associated with aerobic fitness and optimum thresholds found to be necessary for preventing CVD. The final section reviews the literature with regard to the risk associated with overweight and obesity, and measures used in the clinical assessment of CVD risk.
3.2.2 WOMEN’S PERCEPTIONS OF THEIR AEROBIC FITNESS Many women in Wales are insufficiently active and it is acknowledged by the WAG (2009) that low levels of physical activity and unhealthy eating patterns are leading to increases in the prevalence of overweight and obesity. Knowledge of physical activity determinants is useful in understanding women‘s sedentary behaviour for the purpose of developing strategies to promote an active lifestyle, increase fitness levels and facilitate personal responsibility for health. There is however a paucity of information about women‘s selfperceptions of their aerobic fitness and whether these perceptions are influential in determining activity levels. Information regarding women‘s ability to estimate their fitness accurately is of interest within a health prevention paradigm. This is because unfit women might incorrectly perceive their fitness level as adequate for their health and fail to meet recommendations for physical activity of sufficient frequency, intensity and duration. The Women‘s Sport and Fitness Foundation (WSFF, 2007) exposed a serious gap between women‘s perceptions and reality with three out of five in a survey of 1,000 women believing they were already active enough for health although many were not meeting physical activity guidelines. This situation was more pronounced as women aged with three-quarters of women aged over 65 years believing they were active enough for their health although only seven percent were achieving three sessions of 30 minutes of physical activity a week. On this basis, it would be unlikely that most women‘s perception of their fitness would be congruent with their measured fitness. This is an important issue because misperceptions regarding personal fitness could influence women‘s decision-making about the need to remain physically active particularly in older age when participation levels generally decrease (Donaldson, 2009). A discrepancy between a perception and reality raises issues relating to the basis for women‘s personal perceptions of their fitness. It is possible that women are not familiar with current guidelines for physical activity and therefore underestimate the
71
frequency, intensity, duration and mode of exercise necessary to maintain beneficial levels of health-related fitness. Moreover, women might be unaware of the need to maintain physical activity throughout the lifespan, or they might assume that their busy lifestyle confers sufficient physical fitness. Alternatively, they might give no thought to the matter.
In a meta-analysis of 28 studies having a perceived and objective measure of physical fitness, Germain and Hausenblas (2006) examined the relationship between perceived and actual fitness, hypothesising that the magnitude of the association would be small; furthermore, that neither gender nor age would significantly moderate this relationship. Germain and Hausenblas (2006) referred to the human tendency to hold positive illusions about life in order to make the world appear predictable and controllable, arguing that selfperceptions of fitness might be exaggerated in a positive direction, and that this illusion is hazardous to health. The authors concluded that whilst in general people see themselves as better than others in a variety of situations they have moderately accurate perceptions of their physical fitness, with little evidence of illusory self-perceptions. Whilst no difference was found with regard to gender Germain and Hausenblas (2006) found that age moderated the perceived-actual fitness relationship and perceptions were found to be less accurate in older people.
This has important implications for cardiovascular health because women more so than men tend to become less active as they age (Donaldson, 2009). Older women might be unreliable at estimating their fitness, with overestimation possibly having a detrimental impact on their uptake and adherence to a physically active lifestyle. From a health perspective, it would be reassuring if women were found to be accurate at estimating their level of aerobic fitness in addition to knowing about the benefits of maintaining health-related fitness throughout the lifespan.
3.2.3 AEROBIC FITNESS AND RISK OF CARDIOVASCULAR DISEASE Within the context of assessing CVD risk status, it is important to assess physical activity levels because inactivity is a major risk factor for CVD (Blair et al., 1989; Graham et al., 2007). Conversely, habitual physical activity of sufficient intensity, duration and frequency is protective;
conferring
beneficial
changes
in
hemodynamic,
hormonal,
metabolic,
neurological, and respiratory function resulting from increased exercise capacity (Fletcher et al., 1992). However, habitual physical activity consists of a variety of complex behaviours that can be challenging to measure with accuracy particularly if it involves subjective selfreporting. For example, most women engage in numerous modes of activity throughout an 72
average day including housework, occupational, child-care and leisure time pursuits and it might be challenging to arrive at overall accurate judgements about physical activity status. An alternative to the assessment of physical activity is to measure aerobic fitness which might provide more objective information pertinent to the assessment of the risk of long-term illness. Prospective epidemiological studies in healthy women as well as in women with hypertension, obesity, diabetes or CVD, have found that low aerobic fitness is an indicator for all-cause and cardiovascular mortality, and non-fatal cardiovascular events (LaMonte et al., 2000; Gulati et al., 2003; Sui et al., 2007; Lyerly et al., 2009).
A contentious issue concerns the level of aerobic fitness at which the risk of CVD significantly increases. In light of this issue, the following section focuses on studies that have investigated the relationship between aerobic fitness and the risk of CVD in women. The criteria for inclusion of these studies were that their stated exposure variable was aerobic fitness and that this was measured using established valid and reliable protocols. Furthermore, that the outcome variables consisted of reduced cardiovascular risk and events, or death from CVD or all-causes, also that there was a substantial follow-up period following a base-line assessment of aerobic fitness. Table 6 shows a synopsis of studies meeting these criteria.
73
Table 6, Studies of aerobic fitness and risk of cardiovascular disease Author
Blair et al. (1989)
Title
Physical fitness & allcause mortality
Population/Design
Longitudinal cohort of men & women. Free of known disease (n = 13 344): (3 120 women)
Blair et al. (1996)
Gulati et al., (2003)
Influences of aerobic fitness & other precursors on CVD & allcause mortality in men & women
Observational cohort study (n = 7080)
Exercise capacity & the risk of death in women.
Longitudinal study
Women, aged 20 – 88 yrs
(n = 5721) Aged 52.4 ±10.8 yrs Asymptomatic women.
Assessment
Outcome
Summary
Age, Smoking, Cholesterol, Systolic BP Fasting blood glucose, Parental history of CHD
All-cause mortality during 8 yrs follow up
Strong graded inverse relationship between fitness & mortality suggestive of moderate levels of physical fitness protects against early mortality. Asymptote established at 9 METs (32.5 ml.kg 1 -1 .min ) at all ages. Major reduction in all-cause death rates between the st nd 1 & 2 quintile.
Aerobic fitness Assessment with maximal exercise test. Modified BalkeWare protocol -1 -1 ml kg min converted to METs. Low fitness defined as lowest 20.0%. Moderate & high fitness defined as 40.0% & 40.0% respectively
Smoking, Systolic BP, Cholesterol
CVD & allcause mortality
Low fitness an important precursor of mortality. Protective effect of fitness held for smokers & nonsmokers, those with & without elevated cholesterol levels or elevated BP, & unhealthy or healthy persons. Moderate fitness seems to protect against influence of other predictors on mortality.
Symptomlimited treadmill test using Bruce protocol, Exercise capacity in METs Estimated by speed & grade of treadmill.
Major risk factors
Risk of death
Exercise capacity an independent predictor of death in asymptomatic women. Risk of death doubled in women in 5 to 8 MET exercise capacity category. Highest fitness category was more than 8 METs.
Maximal exercise treadmill test categorized in fitness quintiles
74
Adjustment
43 deaths
Kokkinos et al., (1995)
CRF & CHD risk factor association in women.
Observational cohort study (n = 522) Women, 22 - 79 yrs (mean age 45 ± 5 yrs) free of chronic disease.
Medical history Physical examination, BP Fasting blood chemistry & lipoprotein, BMI. Maximal Bruce graded exercise test
BMI Blood glucose BP Lipoprotein.
Association between CRF & coronary risk factors in healthy nonsmoking adult women.
Women in lowest fitness category had less favorable lipid profiles, blood glucose levels, BP & BMI than women in moderate & high fitness categories. CRF equivalent to 10 METs required for improving coronary risk profile.
La Monte et al., (2000)
Aerobic fitness & CHD risk factors
Observational study between 1975 & 1997 n = 1128 women (mean age 43.8 ± 12.8 yrs)
Health examination, maximal treadmill exercise test
BMI, BP Fasting blood glucose Lipids Pulmonary function.
Aerobic fitness & CHD risk factors in men & women with & without existing CHD.
Aerobic fitness inversely associated with all CHD risk factors in women regardless of CHD status in low risk population. Only HDL demonstrated significant linear association with fitness.
Sui et al., (2007)
Aerobic fitness as a predictor of nonfatal cardiovascular events in asymptomatic women & men.
ACLS (1971 – 2004) longitudinal cohort study with follow up over 10 yrs. (n = 5909) Women free of CVD
Maximal treadmill exercise test, health examination
Smoking Alcohol intake, Family history of CVD
Association between aerobic fitness & nonfatal CVD events
Aerobic fitness is a significant determinant of nonfatal primary CVD events. Assessment provides prognostic information independent of ECG responses & traditional risk factors. Exercise testing of functional capacity may enhance CVD risk stratification in asymptomatic adults.
Kodama et al. (2009)
Aerobic fitness as a quantitative predictor of allcause mortality & cardiovascular events in healthy men & women.
Systematic literature search for observational cohort studies (n = 102 980 participants).
Meta-analysis (33 studies)
To define quantitative relationship between CRF & CHD or CVD events, or all-cause mortality in healthy men & women.
Higher aerobic fitness associated with lower risk of allcause mortality. Participants with maximum aerobic capacity of ≥ 7.9 METs had substantially lower rates of all-cause mortality & CHD/CVD events compared with those with maximum aerobic capacity < 7.9 METs.
_________________________________________________________________________________________ Key: ACLS = Aerobic centre longitudinal study; BP = blood pressure; BMI = body mass index; CHD = coronary heart disease; CRF = cardio-respiratory fitness; CVD = cardiovascular disease; ECG = electrocardiogram; METs -1 -1 = metabolic equivalent where one MET is equivalent to 3.5 ml·kg ·min
75
Blair et al. (1989) investigated physical fitness and all-cause, and specific-cause mortality in men and women asymptomatic for CVD at an initial assessment of aerobic fitness. The results demonstrated ―a strong, graded and consistent inverse relationship between physical fitness and mortality in men and women‖ (p 2401). Less fit women had a higher risk of death compared with more fit women and there was a higher risk for all-cause mortality in the least fit women. Blair et al. (1989) cautiously proposed an optimal aerobic fitness level occurring in women at approximately 9 METs. The results supported the hypothesis that low aerobic fitness is a risk factor for all-cause mortality. The representativeness of the population studied however was limited by a bias towards participants who were well educated and from middle and upper socioeconomic strata. In mitigation, Blair et al. (1989) pointed out that on key variables such as serum cholesterol and blood pressure, the study population was similar to that of other large epidemiological studies. In a later study, Blair et al. (1996) quantified the relationship between aerobic fitness and CVD, and all-cause mortality within a framework of personal characteristics predisposing to early mortality. For the risk factors of cigarette smoking, systolic hypertension, elevated total cholesterol, and a BMI of 27 kg m-2 or more, high-fit women had lower death rates than women with low fitness. Aerobic fitness estimated in METs was categorised as low fitness (least fit 20.0%), moderate fitness (next 40.0%) and high fitness (highest 40.0%). Table 7 shows aerobic fitness in METs according to age categories in the Aerobics Centre Longitudinal Study (1971-2001). Table 7, Aerobic fitness estimated in METs according to age category in the Aerobics Centre Longitudinal Study (ACLS, 1971-2001).
Aerobic fitness estimated in METs and classed according to age categories in the ACLS population (Blair et al., 1971-2001).
20-39 yrs 20-39 yrs 20-39 yrs
Low fitness Moderate fitness High fitness
10.4
40-49 yrs 40-49 yrs 40-49 yrs
Low fitness Moderate fitness High fitness
9.4
50-59 yrs 50-59 yrs 50-59 yrs
Low fitness Moderate fitness High fitness
8.5
≥60 yrs ≥60 yrs ≥60 yrs
Low fitness Moderate fitness High fitness
7.6
Key: ACLS = Aerobics Centre Longitudinal Study (1971-2001); METs = metabolic equivalent where one MET is -1 -1 equivalent to 3.5 ml·kg ·min
76
Even with elevated cholesterol, high-fit women had a 23.0% lower risk of death than did lowfit women and for systolic blood pressure of 140 mm Hg or higher, women with higher fitness had an 81.0% lower risk of all-cause mortality compared with women of low fitness (Blair et al., 1996). The results demonstrated a graded inverse trend of death rates from the low to high-fitness groups, within strata of the number of other risk factors leading Blair et al. (1996) to conclude that there was strong evidence for low aerobic fitness as a precursor for CVD and all-cause mortality. Moreover, a review of the literature by Blair et al. (2001) regarding the dose-response relation between physical activity and health, aerobic fitness and health outcomes, and the outcome gradient across categories for the two exposures, found that most studies demonstrated an inverse dose-response gradient across physical activity categories for most health outcomes. For aerobic fitness, Blair et al. (2001) found consistency across studies with all reporting a strong inverse gradient of mortality across fitness groups.
Gulati et al. (2003) similarly confirmed an association between low aerobic fitness and increased risk of death from CVD in a study assessing the prognostic value of aerobic fitness as an independent predictor of all-cause death in asymptomatic women. The risk of death was doubled for women whose aerobic fitness was in the range of 5 to 8 METs, and tripled for those in the least fit when compared with the highest category of aerobic fitness of more than 8 METs. Gulati et al. (2003) concluded that aerobic fitness determined by exercise capacity is an independent risk factor for death additional to other CVD risk factors in asymptomatic women, and that exercise capacity is even more predictive in women than it is in men. A limitation of the study however, was the voluntary nature of participation that would have influenced the demographic makeup of the cohort.
The impact of risk factors on blood lipids, blood pressure, and blood glucose in healthy women was investigated by Kokkinos et al. (1995) who found a strong association of aerobic fitness with cardiovascular risk factors. Women with low fitness had less favourable lipid profiles, blood glucose levels, blood pressure and anthropometric indices than women with moderate or high fitness; furthermore, a fitness level of ~6 to 11 METs was associated with significant improvement in all coronary risk factors examined. The relationship between aerobic fitness and CVD risk factors was similarly investigated by LaMonte et al. (2000). The authors found that aerobic fitness was inversely associated with CHD risk factors of blood pressure, blood glucose, and blood lipids independent of age, percent body fat, smoking status, and family history of CHD and this was the case even in low risk populations. Of interest within a health prevention paradigm was the observation by LaMonte et al. (2000) that although fitness is affected by detraining, training gains and genetic influence, health77
related levels of aerobic fitness can be achieved through regular moderate-intensity activity such as brisk walking. This is significant in that it supports the rationale underpinning public health recommendations for habitual physical activity to promote health benefits (Donaldson, 2009).
3.2.4 CLINICAL ASSESSMENT OF AEROBIC FITNESS Sui et al. (2007) reported on the prospective association of aerobic fitness and non-fatal CVD events as a part of the Aerobic Centre Longitudinal Study (ACLS, 1971-2001). Women who developed CVD tended to be older and sedentary, with other major risk factors, and poor aerobic fitness. Aerobic fitness was categorised for women as the lowest 20.0% of the sample (low fitness), the next 40.0% (moderate fitness) and the remaining 40.0% as high fitness (See Table 7). After adjustment for co-variables, women with moderate and high fitness had a 26.0% and 37.0% lower risk of CVD events than did women with low fitness (p = 0.05). Sui et al. (2007) concluded that aerobic fitness is a significant determinant of nonfatal primary CVD events in women and that assessment can provide important prognostic information independent of exercise electrocardiogram responses and traditional risk factors. The authors recommended that clinicians should consider the benefits of assessing aerobic fitness and should counsel sedentary patients to become more physically active to improve their aerobic fitness as a cornerstone of primary CVD prevention.
The American Heart Association ([AHA], Lauer et al., 2005) had previously considered the value of exercise testing in risk stratification of asymptomatic individuals acknowledging that population-based studies had shown that impaired functional capacity could predict increased CVD risk. Moreover, that exercise capacity could predict risk over and above established risk algorithms. The AHA (2005) did not however recommend routine exercise testing until further evidence was available to substantiate its efficacy within a clinical setting. Furthermore, the AHA (2005) identified a need for large-scale randomised trials to evaluate whether exercise screening leads to improvement in outcomes. Cost-effectiveness and clinical value were regarded as particularly important in proposed clinical trials. The AHA (2005) did not elaborate on what was meant by clinical value. However, it is interesting to consider that data from exercise testing of aerobic fitness can indicate risk of CVD but serves no purpose unless communicated successfully to the individual, together with a management plan for physical activity to improve fitness. Unlike treatment for hypertension or high cholesterol, there is no medication for low fitness. A part of the efficacy of clinical value therefore depends on a successful partnership between the health-care provider and the individual who might benefit from improved aerobic fitness. 78
According to Kodama et al. (2009), a major reason for an absence of fitness testing in clinical assessment is that the quantitative association of aerobic fitness for cardiovascular risk is not well established. Furthermore, Kodama et al. (2009) pointed out that the degree of risk reduction associated with each incremental higher level of fitness, the criteria for low fitness, and the magnitude of risk associated with low fitness has not been consistent among studies. This ambiguity formed the basis for a meta-analysis to define the quantitative relationship between aerobic fitness and CHD events, CVD events, or all-cause mortality in healthy men and women (Kodama et al., 2009). Aerobic fitness was estimated as maximal aerobic capacity expressed as a metabolic equivalent (METs). A 1 MET higher level of maximum aerobic fitness was associated with a 15.0% decrement in risk of all-cause mortality, and CHD and CVD respectively; individuals with low aerobic fitness of less than 7.9 METs had a substantially higher risk of all-cause mortality and CVD compared with individuals with an intermediate level of aerobic fitness between 7.9 and 10.8 METs, and high aerobic fitness of 10.9 METs or more. Kodama et al. (2009) suggested that consideration of low aerobic fitness as a major risk factor could be useful in a clinical setting through identification of low exercise tolerance. This viewpoint endorsed that of Franklin (2009) who stated that measurement of aerobic fitness could provide additive and independent information to risk estimation. Franklin (2009) added the suggestion that health care professionals should expand medical evaluations and risk factor profiling to include objective data regarding aerobic fitness expressed relative to age and sex norms.
3.2.5
OVERWEIGHT
AND
OBESITY:
DETERMINING
THE
RISK
OF
CARDIOVASCULAR DISEASE The importance of addressing the issue of overweight and obesity in the UK was highlighted by the Foresight report (UK Government, 2007). This suggested that the prevalence in the British population was unacceptably high with trends indicating that by 2050, 9 out of 10 UK adults will be overweight or obese, and will be facing deteriorating health and a lower quality of life. In study one of the present theses it was argued that women could assume some personal responsibility for reducing their risk of CVD through maintaining their BMI and WC within advised parameters. This was because obesity is an independent risk factor for CVD and has adverse effects on other risk factors (National Heart Lung and Blood Institute Obesity Task Force [NHLBI OTF], 1998; Manson et al., 1995; Calle et al., 1999; Hu et al., 2004; Dyer et al., 2005). However, for the health professional or fitness instructor the literature regarding the assessment of CVD risk associated with overweight and obesity could be considered confusing. This is because in efforts to establish the relationship
79
between adiposity and CVD risk or mortality, a variety of measures have been adopted including BMI, percent body fat, waist to hip ratio and WC (Manson et al., 1995; Rexrode et al., 1998; Ashton et al., 2000; Hu et al., 2004; Zhang et al., 2008). Litwin (2008) pointed out a lack of consensus regarding which measures of adiposity best indicate CVD risk with research focussing on measures of either generalised or central obesity. A further factor to consider was highlighted by Farrell et al. (2002) who found that in studies examining the relationship of BMI with mortality, a failure to measure cardio-respiratory fitness might have confounded the relationship if there was a false assumption that all overweight or obese individuals were sedentary. Farrell et al. (2002) found that low aerobic fitness was a more important predictor of all-cause mortality than BMI, leading the authors to suggest that health professionals would spend their time wisely if they allocated time to encouraging sedentary women to be more physically active in addition to counselling about overweight and obesity. This is pertinent because health professionals and fitness instructors need confidence in the advice they give to patients or clients. In relation to overweight and obesity, this should be based on kinanthropometric measures that are underpinned by strong evidence of an increased risk of CVD and mortality. It would however, appear logical that as both inactivity and obesity are major risk factors, women should receive advice that acknowledges the role of both in the development of CVD. A further point for consideration is that within the context of the GP surgery or fitness facility, time is often constrained and there is a need for measures of adiposity that are relatively easy to implement and that facilitate the communication of risk easily to patients, together with strategies for prevention.
Balkau et al. (2007) in a study of European male and female patients aged between 18 and 80 years attending their GP surgery recorded BMI, WC, smoking status, and the presence of diabetes or CVD (n = 177 345). The frequency of overweight patients with a BMI between 25 and 30 kg m2 was similar across geographical regions in males and females. Thirty percent of women (n = 98 750) were classified as overweight and 27.0% as obese providing clear evidence of a problem with overweight and obesity in this particular European population. Furthermore, 71.0% of women presented with abdominal adiposity based on criteria of the International Diabetic Federation (2006) of a WC measuring at, or more than 80 cm. Overall, the standardized ORs for CVD were higher for WC than for BMI in both men and women. The ORs (95.0% CI) for women of WC versus BMI were 1.21 (1.17 to 1.25) versus 1.20 (1.16 to 1.24). Both WC and BMI were independently associated with CVD and CVD was significantly associated with WC even in lean individuals with a BMI at, or less than 25 kg m2. Balkau et al. (2007) highlighted the importance of assessing body fat distribution even in normal-weight individuals and advised the measurement of WC or WHR in addition to BMI for assessing the risk of cardiovascular mortality. In this large cross-sectional study, 97.0% of 80
patients approached with information about the study agreed to participate. Although implementing measures of adiposity within GP surgeries in the UK is not routine unless the ‗Health Check‘ programme (NHS, 2010) is operating in that locality, the high voluntary participation rate in the study by Balkau et al. (2007) suggests that patients might not be averse to measures of adiposity taken during their consultation.
In the UK, the National Obesity Forum (2010) recommended a healthy WC of less than 80 cm for women, with more than this measure indicating ‗increased cardiovascular risk‘, and more than 88 cm indicating ‗significantly increased risk‘. These same measures are in agreement with advice of Diabetes UK (2006) in their campaign titled ‗Measure Up‘ that aims to raise public awareness of the link between Diabetes and abdominal adiposity. It is pertinent to note that with regard to assessment of long-term risk of health problems including CVD, NICE (2006) recommended taking into account the combination of BMI and WC. For women, a BMI in the normal range (18.5 to 24.9 kg m-2), and a WC of less than 80 cm (31.5 inches) was given as the criteria to indicate no increased risk of cardiovascular, or other long term health problems. Combined BMI and WC was also advised by the World Health Organisation (WHO, 2011) on the basis that combining BMI with a measure of abdominal adiposity might improve discriminatory capability with regard to estimation of CVD risk, and furthermore, that account should be taken of ethnic and age differences in risk thresholds. Because these measures are not overly time consuming to implement they might form the basis of useful discussion with patients or clients with regard to preventing CVD and other long-term health problems.
3.2.6
BODY
MASS
INDEX,
PHYSICAL
ACTIVITY
AND
RISK
OF
CARDIOVASCULAR DISEASE It has been established that women‘s risk of CVD is heightened when they are insufficiently active combined with being overweight or obese. In an exploratory study, Mora et al. (2006) determined separate and joint effects of physical activity and body-weight on inflammatory and lipid biomarkers for CVD in healthy women (n = 27 158). The mean age was 54.7 years. Physical activity was categorised into two groups consisting of women who participated in two and a half hours or more a week, and those who participated in less than this threshold. Body mass index was categorised into three groups of normal BMI (18.5-24.9 kg m-2), overweight (BMI, 25.0-29.9 kg m-2), and obese (BMI, ≥ 30.0 kg m-2). Mora et al. (2006) examined the odds ratios for clinical biomarker thresholds within categories of ‗normal weight-active women‘, ‗inactive-overweight women‘, ‗active-overweight women‘, ‗inactiveoverweight women‘, ‗active-obese women‘, and ‗inactive-obese women‘. Lower levels of 81
physical activity and higher BMI were independently associated with adverse levels of nearly all lipid and inflammatory biomarkers (p = < 0.001). Mora et al. (2006) used the reference group of physically active normal-weight women with a BMI between 18.5 and 24.9 kg m-2. and after adjusting for age, race, smoking, blood pressure, diabetes, menopausal status, and hormone use found that the odds ratios (95.0% confidence intervals [CIs]) for having Creactive protein of more than 3 mg L were: for ‗inactive-normal-weight women‘ 1.26 (1.151.37); ‗active-overweight‘ 2.68 (2.41-2.98); ‗inactive-overweight‘ 3.11 (2.84-3.41); ‗activeobese‘ 8.25 (7.15-9.51); and ‗inactive-obese 9.86 (8.84-10.99). The odds ratios (95.0% CIs) for having HDL cholesterol of less than 50 mg dL were 1.20 (1.11-1.30); 2.25 (2.04-2.49); 2.62 (2.41-2.85); 4.21 (3.68-4.81); and 5.27 (4.77-5.84), respectively. For apolipoprotein B100 less than 120 mg dL they were 1.21 (1.11-1.33); 1.86 (1.66-2.08); 2.06 (1.88-2.67); 2.35 (2.04-2.70); and 2.33 (2.09-2.59). Fibrinogen, ICAM-1, apolipoprotein A1, total cholesterol, and LDL cholesterol showed similar associations although homocysteine, lipoprotein (a), and creatinine showed weak or non-significant associations. Therefore, lower levels of physical activity and a higher BMI were independently associated with nearly all of the lipid and inflammatory biomarkers examined (p = 0.001). Although high BMI was strongly related to adverse cardiovascular biomarkers, within categories for BMI, physical activity was also associated with more favourable levels of biomarkers than inactivity and the relationship between physical activity or BMI and biomarkers did not show a threshold effect but a linear relationship. Mora et al. (2006) suggested that both physical activity and BMI have significant independent associations with inflammatory and lipid factors possibly related to the development of CVD. The most favourable biomarker values were observed in women with a BMI in the range 18.5 to 24.9 kg m2 and who were meeting physical activity guidelines leading Mora et al. (2006) to conclude that a BMI in this range and adherence to guidelines for physical activity are necessary components of a healthy lifestyle for women. The results of this study are relevant in that they highlight the need for women to receive advice about healthy body-weight management and physical activity as cornerstones of CVD prevention.
3.2.7 SUMMARY There is a paucity of information regarding women‘s self-perceptions of their aerobic fitness although inaccurate perceptions might act as a barrier to physical activity. This is particularly important in an era where the individual is encouraged to assume more responsibility for their health. Over-estimation of aerobic fitness might contribute to deterring women from physical activity and therefore to acquiring sufficient levels of aerobic fitness for CVD prevention. It 82
was argued that whilst physical activity and low aerobic fitness are recognised as important for CVD prevention, they remain tangential in risk evaluation systems and despite evidence that exercise testing has prognostic value it is not included in risk evaluation systems due to a lack of agreement in relation to aerobic fitness thresholds. A recent large-scale metaanalysis however, suggested that assessment of aerobic fitness should be a part of routine screening for CVD risk, with low fitness defined as aerobic fitness of less than 7.9 METs (Kodama et al., 2009).
Finally, this review considered the clinical efficacy of measures of adiposity used for risk evaluation. The relationship between adiposity and cardiovascular risk is complicated by the various measures used for assessment and by the influence of gender, age, and ethnicity (WHO, 2011). Whilst researchers have evaluated the reliability and validity of various measures of adiposity, Farrell et al. (2002) cautioned that many studies had ignored the effect of aerobic fitness when investigating the adiposity-mortality relationship. With regard to physical activity and adiposity, Mora et al. (2006) found that the most favourable cardiovascular profile for women consisted of a BMI in the normal range together with adherence to physical activity guidelines.
3.3.1 STUDY PROCEDURES Participants completed the following procedures:
i.
A sub-maximal test of aerobic fitness.
ii.
Measures of height, mass and waist circumference.
iii.
Completion of a seven-day physical activity question.
iv.
Personal assessment of perceived aerobic fitness.
3.3.2 ETHICAL CONSENT AND UNDERPINNING RATIONALE The proposal for the thesis was reviewed and approved by the ethics sub-committee of the Cardiff School of Sport Ethical Committee. The concept of respectful research was implicit in the rationale of the study design. According to McNamee (2007), this is research that respects the autonomy of the participant so that external pressure is not applied, and the participant is fully aware of what is required of them. Thus, the researcher is committed to telling the truth about the research and respecting the privacy of those who do, and those who do not, want to participate. Additionally, respectful research preserves the anonymity of the participant, takes robust steps to ensure confidentiality, and obtains informed consent. 83
Moreover, respectful research is mindful of the necessity to protect the participant from psychological or physical harm (McNamee, 2007). Therefore, during each step of the study design, consideration was given to preserving the safety, privacy and well-being of the participant.
3.3.3 ELIGIBILITY CRITERIA The participants were apparently healthy women, aged between 25 and 65 years residing in the Vale of Glamorgan. The term ‗apparently healthy‘, described women with no reported known disease. Pregnant women or those with pre-existing medical conditions such as hypertension, diabetes, or arthritis were ineligible.
3.3.4 RECRUITMENT OF PARTICIPANTS The researcher approached nine organisations in the study locality and asked for permission to talk to women about the study. Women were not approached individually but were given initial verbal and written information within a group setting (Appendix E). The written information explained the purpose of the study and rights of the participant, health and safety implications, the procedures taken to protect confidentiality, and what would be expected of a participant if deciding to volunteer for the study. According to McNamee (2007), it is important that the written information enables potential participants to make a free choice based on sufficient knowledge. Therefore, the written information was tested and amended for readability by four women attending an exercise cardiac rehabilitation class. Potential participants were requested to read the information, and asked to carefully consider their decision to take part in the study. As a part of this initial process, a pre-exercise screening questionnaire was enclosed with information about the study. This was included so that potential participants would be fully informed about questions they would be asked relating to their health and well-being if they decided to participate. The approach to women in a variety of locations was intended to increase the probability of a representative sample of apparently healthy women in the study age group. The researcher visited each venue one week after the initial visit to enquire if anyone had decided to participate, and to address any concerns. The organisations included in the study were as follows:
i.
Two Local Authority community fitness classes (15 participants).
ii.
Employees at one local leisure centre (3 participants).
iii.
One privately owned fitness club (6 participants).
iv.
A locally based Women‘s Institute (4 participants). 84
v.
A local church ladies group (4 participants).
vi.
A line dancing group (6 participants).
vii.
Young mothers at a play-group (9 participants).
viii.
A knitting circle (5 participants).
ix.
Teachers from a local junior school (6 participants).
Sixty-three women expressed an interest in participating in the present study. Two women were ineligible due to high blood pressure measures during the initial pre-exercise health screening. These women were advised to visit their GP and subsequently informed the primary researcher that they were prescribed medication. A further two women were ineligible on the basis of musculoskeletal pathology which would have impacted on their safety to perform the CST (Sykes, 2005). One woman had a pregnancy confirmed and was therefore ineligible. Consequently, 58 women aged between 25 and 65 years were eligible to participate in the present study following pre-exercise screening.
3.3.5 INFORMED CONSENT Arrangements to meet women who had agreed to participate in the study were made on an individual basis. During this meeting there was opportunity for the participant to ask further questions. It was explained verbally to the participant that they were free to change their mind at any time. The concept of participant autonomy underpinned communication between the researcher and participant throughout the process of obtaining informed consent, preexercise screening and exercise testing. Those women who volunteered to participate signed a consent form and were given a copy (Appendix C). Consent was obtained before completion of a pre-exercise health-screening questionnaire, and prior to exercise testing.
3.3.6 ANONYMITY AND CONFIDENTIALITY Participants were informed of their right to remain anonymous. The informed consent and pre-exercise screening forms were initialled by those participants choosing not to give their full name. Confidentiality was assured verbally, and in writing. This took the form of honouring the promise to keep information securely, and not to divulge information to a third party. Forms were identifiable through assignment of a study number.
85
3.3.7 THE TESTING VENUE
The venue for completion of the pre-exercise screening questionnaire and the fitness test was a studio in two community leisure centres. Permission was obtained from the manager of the leisure centres to use the facility. The studio was not accessible to other users during pre-exercise screening or exercise testing, thereby protecting the privacy of the participant. This venue was appropriate for ease of access for participants, and because of availability of changing rooms, access to water, toilets and air-conditioning. There was also access to firstaid equipment and a telephone. All participants completed the screening and subsequent exercise test on a Sunday afternoon between 3.30 pm and 6.30 pm. The room temperature of the studio was set at 190 Celsius. The researcher sought to establish a rapport with the participant on arrival at the venue by meeting them at reception and initiating conversation to put them at ease.
3.3.8 PRE-EXERCISE SCREENING A pre-exercise health-screening questionnaire provided information relevant to the safety of fitness testing (ACSM, 2010). It also provided an additional filter to meet the eligibility criteria of the study for ‗apparently healthy‘ women (Appendix F). The decision not to proceed with exercise testing was taken if cause for concern became obvious as a result of pre-exercise screening. For example, where blood pressure measurement indicated hypertension, the participant was informed of the reading and advised to visit their GP. The researcher was aware that in communicating this information there was a potential to cause anxiety, and was therefore mindful to communicate information sensitively. A section on the pre-exercise screening form inquired about serum cholesterol. It was not possible to ascertain missing values for serum cholesterol therefore missing information was included as one cardiac risk factor according to guidelines of the ACSM (2010). Where the participant was unaware of their serum cholesterol, they were asked if they would be interested in receiving published information about cholesterol by the BHF. This was made available to participants who expressed an interest. Participants considered at risk for other reasons, for example, with signs and symptoms of coronary heart disease were advised to visit their GP before embarking on an exercise programme and were excluded from the study. Procedures for measurement of waist circumference, stature, mass, and blood pressure are presented in Appendix G.
86
3.3.9 MEASURING AEROBIC FITNESS Maximal oxygen consumption (VO2max ) is a recognised measure of aerobic fitness and is defined by Franklin (2007) as ―the maximum rate of oxygen transport and utilisation, as verified by a levelling off of oxygen consumption with increasing workloads‖ (p42). The term
VO2peak is often used as a substitute due to volitional fatigue occurring below the level of a physiological maximum effort. Work capacity is expressed in litres of oxygen per minute (L min-1), or because larger individuals with a greater muscle mass may have larger absolute oxygen consumption, the value is divided by body mass in kilograms to enable comparison between individuals of varying size. Therefore, capacity is expressed as millilitres of oxygen per kilogram of body mass per minute of exercise (ml·kg-1·min-1) or as a metabolic equivalent (MET). VO2max can be estimated from the speed of the treadmill and percent grade, and the Bruce treadmill protocol (Bruce et al. 1959) is widely used in research studies and in clinical settings for this purpose Where it is not feasible to assess cardiorespiratory fitness in a clinical or laboratory setting, VO2max can be extrapolated from performance using a cycle ergometer, step test protocol or Multistage Shuttle Run (Sykes, 2005). Maximal test protocols for measurement of VO2max represent the gold standard for measurement of aerobic fitness (Heywood, 2002). According to the ACSM (2010), however, maximal testing incurs a higher risk of injury to the participant, particularly when the participant is older, or less fit. Maximal test protocols also require specific monitoring equipment, and usually take place in a laboratory setting. This can incur financial costs for the participant in terms of travelling expense and is time consuming. For these reasons, maximal testing can limit the level of voluntary participation.
3.3.10 SUBMAXIMAL EXERCISE TESTING The decision to use a sub-maximal test of aerobic fitness in the present study was taken for reasons of participant safety and convenience. For general use within a field setting a submaximal test can more easily be widely administered. A sub-maximal fitness test requires the participant to perform a fixed amount of work, per unit of time. This limits the level of exertion so that the test is terminated at a pre-determined heart rate (HR) intensity allowing the prediction, or estimation of VO2max (ACSM, 2010). Sub-maximal tests assume a steady state HR at each exercise intensity as well as a linear relationship between HR, oxygen uptake, and work intensity (Heyward, 2002).
87
According to Sykes (2005), there are limitations to sub-maximal tests in terms of accuracy, with most having a standard error of estimate of around 12.0% to 15.0%. The predictive score therefore, may overestimate or underestimate VO2max . According to Heyward (2002; p.67), ―VO2max predicted by a sub-maximal test tends to be overestimated for highly trained individuals, and underestimated for untrained sedentary individuals‖. A further disadvantage of sub-maximal testing is that they assume a degree of mechanical efficiency that is constant for all individuals. Poor efficiency can result in a higher sub-maximal HR at a given work load resulting in an underestimation of VO2max (McArdle et al., 2007). A further limitation is that sub-maximal tests assume that the maximal HR for clients at a given age is similar. Maximal HR however, can vary with a 10 to 12 beat per minute standard deviation in maximal HR in normal subjects (Fleg et al., 2000). In mitigation, according to Mackinnon et al. (2003) although an error in estimation exists, sub-maximal exercise testing can be less stressful for the participant, and is less time consuming.
3.3.11 RATIONALE FOR THE USE OF THE CHESTER STEP TEST There are several sub-maximal test protocols available. These include the Harvard step test, (Brouha et al., 1943), the Astrand cycle test (Astrand, 1986) and the Chester Step test (CST), (Sykes, 2005). The CST is a sub-maximal, multistage, aerobic capacity test originally designed for use in the workplace and for medical/fitness screening (Sykes, 2005). This test protocol was adopted for the present study because there is evidence to support its reliability and validity, because it is easily transportable, and the step height is flexible to accommodate a range of ages and abilities. Many women are now familiar with exercise protocols that use a step because of their experience of attending step-aerobic fitness classes. Furthermore, according to Sykes (2005), the rationale behind the use of a step is that the action of stepping is reasonably familiar to most people and uses the large muscles of the legs. It allows the researcher to observe the participant working at a moderate rate of exertion in a controlled environment. It is designed for use with a variety of step heights (15, 20, 25 and 30 cm). The CST is progressive, starting with a slow step rate of 15 steps per minute. The tempo increases gradually every two minutes. The test ends when the participant‘s HR reaches approximately 80.0% of maximal HR which is calculated prior to the assessment. This is estimated by 220 minus age (yrs), or the participant reports a rating of perceived exertion (RPE) of ‗moderately hard‘ using a Borg Scale 6 to 20 (Borg, 1994). Choice of step height was based on guidelines given in the test manual. For example, the 15 cm step was generally suitable for those aged over 40 years who took little or no physical exercise, and for those aged over 40 years who were moderately overweight (Sykes, 2005).
88
3.3.12 VALIDITY AND RELIABILITY OF THE CHESTER STEP TEST Stevens and Sykes (1996) compared the CST with the 6-minute cycle test (Astrand 1986) and the 20-metre multi stage shuttle run test ([MSSR], Leger et al., 1988) in a sample of 20 students of sports science. The results demonstrated high correlations between the CST and the cycle test (r = 0.94; p < 0.01), CST and the MSSR (r = 0.81; p < 0.05), with mean scores of the CST (54 ± 9 ml·kg-1·min-1), Astrand cycle test (52 ± 9 ml·kg-1·min-1) and MSSR (51 ± 10 ml·kg-1·min-1) with non-significant differences observed between the means (p > 0.05). Moreover, Sykes and Roberts (2004) compared results from a CST with a VO2max treadmill test in a wide range of 68 males and females aged between 18 and 52 years and found a high correlation (r = 0.92; p < 0.001) between VO2max and the CST with a standard error of the predicted estimate being ± 3.9 ml·kg-1·min-1. The CST was also shown to be highly reliable in test-retest situations (r = 0.95) and suitable for monitoring improvements over time.
Norms provided in the CST (Sykes , 2005) are presented in Appendix H.
3.3.13 EQUIPMENT FOR THE CHESTER STEP TEST, PRE-TEST PROTOCOLS AND PROCEDURE FOR COMPLETION OF THE GRAPHICAL DATA SHEET. The equipment consisted of the following items: I.
TA Reebok standard exercise step with modifiable heights (Reebok UK Lancashire).
II.
A Polar FT 40 heart rate monitor (Polar Instruments Europe Ltd).
III.
A Rating of Perceived Exertion Chart (Borg, 1982).
IV.
A Chester Step Test audio CD (ASSIST Physiological Measurement Resource Manual: ASSIST Creative Resources).
V.
A graphical data sheet (Chester Step test manual), and cassette player.
With regard to pre-test conditions the participant was requested on the initial information sheet to abstain from vigorous activity on the day of the exercise test. There were procedures in place on the day of the test in case the participant became unwell, or wanted to stop. Access to a telephone and first aid was within the immediate vicinity. In addition to written consent, verbal consent was obtained from the participant on the day of the test and the participant was encouraged to ask any questions regarding the exercise test. They were asked if they had adhered to the study protocols regarding not eating, smoking, drinking tea, coffee or alcohol for at least two hours beforehand, and if they felt in good health and not suffering from an infection. The participant wore loose-fitting comfortable clothes. The room 89
was well ventilated and the air-conditioning was set at 19 degrees Celsius. There was access to a defibrillator on the premises. Data for each participant was entered on the CST data sheet. The participant‘s name and age, their maximum heart rate (HR) and HR at 80.0% of maximum were recorded and two horizontal lines were drawn representing maximum HR and 80.0% of HR max. Vertical lines were drawn to indicate the five stages, each lasting 2 minutes. Using a ruler, a line was drawn through the heart rate points on the graphical data sheet and this was continued up to the HR max horizontal line in accordance with CST (Sykes, 2005) protocols. A perpendicular line was drawn from where the HR line crossed the Max HR line. This was read to give the aerobic capacity score in mls O2 kg-1 min-1, and the score was entered on the data sheet. Before commencing the CST (Sykes, 2005) the concept of rating perceived exertion was explained to the participant and a chart was made clearly visible, adhering to the test protocols (Borg, 1998). A heart rate monitor was fitted to the participant and after gentle warm up exercises; the participant was positioned ready to start the test. If their pre-test HR was more than 100 bpm the participant was re-assured in order to alleviate any anxiety. The participant was requested to listen to the instructions on the CD and to commence stepping at the appropriate rhythm as determined by the cassette. After the first two minutes, HR and rating of perceived exertion (RPE) were recorded. Providing the HR was below 80.0% max, and the RPE below 14, the participant maintained stepping according to the rhythm indicated on the cassette. Heart rate and RPE were recorded after each two-minute level. This procedure was followed for the following four levels until termination of the test. If 80.0% of HR max was achieved mid level, providing the participant reported a RPE of 14 or below, the test was continued to the end of the level, when HR and RPE were recorded and the test terminated. Participants‘ were instructed to inform the researcher if they felt unwell or wanted to terminate the test for any reason drinking water was provided at the end of the test. Following completion of the test the participant was advised to complete gentle stretching exercises for the lower body and the researcher demonstrated stretches for the gastrocnemius, soleus, quadriceps and hamstring muscle groups. The heart rate monitor was used to indicate that the participant had returned to a pre-test heart rate and before leaving the building, the participant was asked if they were feeling well.
3.3.14 PERCEPTIONS OF AEROBIC FITNESS The pre-screening questionnaire contained an item asking the participant to estimate their level of aerobic fitness. This was in order to evaluate the accuracy of their perceptions. 90
According to Germain and Hausenblas (2006), a limitation of past research examining selfperceptions of fitness has been a failure to take account of the multidimensionality of fitness. Therefore, when enquiring of a participant how fit they think they are there is a need to stipulate whether this refers to overall physical fitness, or an aspect such as strength or endurance. Furthermore, Germain and Hausenblas (2006) referred to the subjectivity of author-developed questions to ascertain perceptions. To minimise these potential limitations in the present study the participant was informed that the study was about women‘s aerobic fitness levels referring to moderate, rhythmic activity which should make them feel warm and breathing more heavily compared to sitting, and which lasted for a period of 30 minutes or more. This specific explanation of aerobic exercise was based on advice given by the BHF (2009) in an internet publication for the general public entitled ‗Keeping Your Heart Healthy‘. Participants were additionally reminded of example activities given in the seven-day recall physical activity item found in the pre-exercise screening questionnaire. The participant was asked to rate their aerobic fitness as ‗excellent‘, ‗good‘, ‗average‘, ‗below average‘ or ‗poor‘ for their age. This facilitated comparison for accuracy with the participant‘s objective measure for aerobic capacity as estimated by the CST (Sykes, 2005) using the same categories.
3.3.15 ANALYSIS OF MEASURED AND PERCEIVED SCORES OF AEROBIC FITNESS The scores to be analysed included a measured categorisation of the participants‘ aerobic fitness obtained from the CST results (measured scores), and the participants‘ own perceptions of their aerobic fitness category (perceived scores) based on the same categories as those used for the CST – indicating whether the participant‘s aerobic fitness was ‗excellent‘, ‗good‘, ‗average‘, ‗below average‘ or ‗poor‘. The aim of the analysis was to establish how well the participants‘ perceptions of their aerobic fitness reflected the measured values – the extent to which perceived scores agreed with the measured scores.
Atkinson and Nevill (1998) identified that by far the most common statistical methods used to analyse scores collected in this kind of study involve the use of correlation coefficients. Indeed, it is probable that the most commonly used statistical method applied to the resolution of such problems is Pearson‘s zero-order correlation coefficient (r). In applying Pearson‘s correlation to the scores collected in the present study the correlation would be used to test the null hypothesis (H0) that r = 0. Confirming whether the perceived scores are not related (in a linear fashion) to the participants‘ measured scores. In this scenario, the alternative hypothesis (H1) might be that r > 0, where large perceived scores are paired with
91
large measured scores and low perceived scores are paired with low measured scores. If, after testing the H0 using Pearson‘s r it was rejected in favour of the H1 it could be concluded that perceived scores are related to the criterion measured scores in a linear fashion. The major drawback, however, with the application of r as an appropriate analysis statistic here becomes immediately obvious if attention is drawn to the actual statistical question that r is being used to answer. That is the H0 that the participant‘s scores from the measured and perceived methods were not related in a linear fashion. Clearly, r is actually a test of the linear relationship between two sets of scores. When applied to the comparison being made here what is required is information about whether perceived scores agree with measured scores and not whether they are related in a linear fashion. Indeed, it would be highly surprising if scores from two test methods, designed to measure the same thing, applied to the same sample of participants at about the same time did not produce a high statistically significant outcome (i.e. reject the H0 in favour of the H1 (p ≤ 0.05)) – a high correlation coefficient is almost guaranteed. Other issues identified by Bland and Altman (1986) in their seminal paper on the appropriateness of using r as a statistical index for assessing agreement includes: i.
Changes in measurement scale will not affect r but it will certainly affect the agreement between scores.
ii.
Correlation depends to a large degree upon the range of values in a measured sample. If this range is wide, r will be greater than if the range is narrow. This is known as sample heterogeneity. This might happen if a study was conducted on scores gathered from particular tests where the participants were women, some of whom were young and old, fat and thin, and well-conditioned and sedentary etc. In other words, exactly the type of sample being considered in the present study. This weakness in the Pearson correlation method is perhaps the most important because, clearly, r is highly sensitive to the amount of random variation in scores from different methods of measurement, but on its own, it is unable to identify the source of this variation. Indeed, Atkinson (1995) has identified that the relationship between such variations in sample heterogeneity and their net effect upon r can adversely affect the conclusions made about scores when compared to criterion scores.
iii.
The usefulness of r depends upon the degree to which it is statistically significant. The test of significance is both spurious and irrelevant to the issue of whether scores agree with one another.
iv.
Scores that seem to be in poor agreement can often produce quite high statistically significant values of r.
92
v.
On its own anyway, r cannot provide an estimate of the amount of systematic bias between scores from a method comparison.
For the reasons cited above the statistical method chosen in this study to investigate the agreement between measured aerobic fitness scores and perceived aerobic fitness scores was the 95% limits of agreement (95% LoA) method recommended by Bland and Altman (1986). The 95% LoA method requires the computation of the differences between measurements by the two methods for each participant (measured scores minus perceived scores) and from which is calculated the mean ( x
diff)
and standard deviation (SDdiff) of these differences. For
the limits of agreement analysis the outcomes are expressed as the mean difference plus and minus (±) 1.96 standard deviations – that is a 95% probability – 95% of the differences between the methods of measurement should lie between these limits. If it is assumed that the measured scores are the most objective categorization of participants‘ aerobic fitness then these differences could also be called errors (or residuals) between the two methods of measurement. Like all statistical methods the 95% LoA depend upon some assumptions about the original scores: i) that the x
diff
and SDdiff are constant throughout the range of measurements – this
is the condition of equal residual variance (homoscedasticity), and, ii) that these differences are from an approximately Normal distribution – that is that the differences, when plotted as a histogram, would form approximately a classic Gaussian (bell-shaped) curve. To check these assumptions Bland and Altman (1986) proposed two plots: i) a scatter diagram the difference (Y-variable) against the mean of the two measurements (X-variable) for each participant, and, ii) a histogram of the differences. Bland and Altman (1983) also suggested, once they have been computed, adding the upper and lower 95% limits of agreement and the mean difference (bias) to the scatter plot – this has become known as a Bland and Altman plot: Another feature which Bland and Altman were at pains to stress in their 1986 paper was that agreement is a question of estimation, and that it is not about hypothesis testing. Estimates are usually made with some sampling error, and limits of agreement are no exception. In which case, it would be useful to estimate confidence intervals for the upper and lower limits of agreement. The standard error (SE) of the limits is approximately = √(3SDdiff2/n). By constructing the Bland and Altman plot, and performing the computations identified above, it is possible to comment upon each of the components relevant to an assessment of
93
agreement: i) any systematic bias between the scores, ii) the random variations, and, iii) homoscedasticity.
3.3.16 PHYSICAL ACTIVITY LEVELS A seven-day physical activity recall item was included on the pre-exercise screening form. This item has been used in successive Welsh Health Surveys to estimate population levels of physical activity and was used in study one of this thesis. Participants were categorised as ‗sedentary‘ if they did not report undertaking any habitual moderate or vigorous intensity activity. Participants were categorised as ‗partially active‘ if they reported undertaking some moderately intense or vigorous activity, but not enough to meet current guidelines (Haskell et al., 2007). Those participants reporting adherence to current guidelines were categorised as ‗fully active‘. The data was analysed to ascertain the percentage of participants who were sedentary, partially active, or fully active and this categorisation formed the basis for discussion regarding women‘s risk of CVD. The seven-day physical activity question is presented in Appendix F.
3.3.17 AEROBIC FITNESS SCORES The participants aerobic fitness score was recorded from the CST in ml·kg-1·min-1 and the score for each participant was converted into a metabolic equivalent. This was calculated on the basis that one MET is considered equivalent to 3.5 ml·kg-1·min-1 (ACSM, 2010). According to Sykes, (2005) international databases compiled over the last two decades have enabled researchers to relate aerobic fitness scores to age and sex. A table of norms for the CST enabled the tester to predict aerobic capacity of the participant and categorise accordingly as ‗excellent‘, ‗good‘, ‗average‘, ‗below average‘ or ‗poor‘.
The data was
analysed to ascertain the percentage of participants‘ with MET values indicating increased CVD risk according to thresholds established by Blair et al. (1989), Gulati et al. (2003), Kokkinos et al. (1995), and Kodama et al. (2009) and facilitated discussion regarding the efficacy of aerobic fitness assessment of asymptomatic women.
3.3.18 OVERWEIGHT AND OBESITY STATUS The data was analysed to ascertain the percentage of participants‘ presenting with increased risk of long-term illness including CVD, based on risk criteria given by NICE (2006). The resulting data contributed to the overall risk profile of the study sample. The guidelines stated that the risk of health problems should be identified using both BMI and WC for individuals‘ 94
with a BMI of less than 35 kg m-2. For those with a BMI of 35 kg m-2 or more, it is assumed that risk is high with any WC. Women with a BMI of 25 to less than 30 kg m-2 and with a WC of less than 80 cm are not at increased risk. If however, the WC is between 80 and 88 cm they are at increased risk. Furthermore, women with a WC of more than 88 cm are at high risk of developing chronic health problems, particularly when BMI is more than 30 kg m-2.
3.3.19 BODY MASS INDEX AND PHYSICL ACTIVITY STATUS The participants were profiled according to combined physical activity status and BMI. Selfreported physical activity enabled categorisation based on adherence to physical activity guidelines (Haskell et al., 2007). Body mass index data was categorised according to criteria of the World Health Organisation (WHO, 2004) as normal BMI (18.5-24.9 kg m-2), overweight BMI, (25.0-29.9 kg m-2), and obese BMI (≥ 30.0 kg m-2).
This categorisation facilitated
description of the study sample in terms of combined BMI and physical activity status as follows:
i.
Normal BMI and physically active.
ii.
Normal BMI and physically inactive.
iii.
Overweight and physically active.
iv.
Overweight and physically inactive.
v.
Obese and physically active.
vi.
Obese and physically inactive.
3.3.20 PRESENTATION OF THE RESULTS AND DISCUSSION The results and discussion are presented as follows:
i.
Participants‘ perceptions of their aerobic fitness.
ii.
Aerobic fitness, physical activity and risk of CVD.
iii.
Overweight and obesity, and risk of CVD.
3.4.1 CHARACTERISTICS OF THE PARTICIPANTS Characteristics of the participants are shown in Table 8.
95
Table 8. Characteristics of the participants (n = 58)
Variable
Mean and SD
Age (yrs)
46.0 (SD ±11.8) -2
BMI (kg m )
27.7 (SD ± 4.1)
WC (cm)
81.6 (SD ± 9.4) -1
Aerobic fitness (ml kg min)
32.3(SD ± 6.0)
MET value
9.1 (SD ± 1.7)
Key: BMI = body mass index; WC = waist circumference; MET = metabolic equivalent where one MET is -1 equivalent to 3.5 ml kg min
3.4.2 WOMEN’S PERCEPTIONS OF THEIR AEROBIC FITNESS Participants were asked to estimate their aerobic fitness as ‗excellent‘, ‗good‘, ‗average‘, ‗below average‘ or ‗poor‘. Their perception was compared against their measured score category obtained on the CST (Sykes, 2005). Twenty-two women (38.0%) made an accurate estimate of their fitness in that their perceived score matched their measured score. Twenty women (34.4%) underestimated, and 16 women (27.5%) overestimated their fitness. Fourteen of the 16 women who overestimated their fitness self-reported as partially meeting physical activity guidelines (Haskell et al., 2007), and only two were fully active. Of these 16 participants‘ (27.5%) it is troubling that six had a measured aerobic capacity of 7.9 METs or less which is the fitness threshold suggested by Kodama et al. (2009) as necessary to avoid a substantially higher risk of all-cause and CVD mortality.
3.4.3 COMPARING MEASURED AND PERCEIVED AEROBIC FITNESS SCORES The measured and perceived aerobic fitness scores for the participants (n = 58) were analysed using the statistical software package Minitab version 15 (Minitab Inc.) In these data, the differences (measured – perceived) had a mean ( x
diff)
of 0.3 units and standard
deviation (SDdiff) of ± 1.1 units. Hence, the 95% limits of agreement given as x
diff
± (1.96
SDdiff) were 0.3 – (1.96 x 1.1) = -1.9 units, and 0.3 + (1.96 x 1.1) = 2.5 units. Hence a perceived measurement would be between 1.9 units less than a measured score and 2.5 greater. The width of the 95% limits of agreement, -1.9 to 2.5, was 3.4 units of measurement. Raw scores are presented in Appendix I.
96
For the measured fitness category scores versus perceived fitness category scores the bias and the upper and lower limits of agreement have been superimposed on the Bland and Altman plot and are presented in Figure 9.
Scatterplot of Differences vs Means (Bland & Altman Plot)
Differences (Measured - Perceived)
4 3 2
upper limit = 2.5
1 bias = 0.3 0 -1 -2
lower limit = -1.9
-3 1.5
2.0
2.5 3.0 3.5 Means ((Measured + Perceived)/2)
4.0
4.5
Figure 9. Bland and Altman plot showing the components of agreement between the n = 58 participants‘ measured and perceived aerobic fitness scores In Figure 9, those data points that lie on the zero line (Y-axis) indicate that there is perfect agreement between the measured and perceived scores for these participants (difference = 0). Because perceived scores were subtracted from measured scores, data points above the zero line indicate that in these cases the differences between measured and perceived scores is positive and therefore the participants‘ measured scores were larger their perceived scores. For those points below the zero line, the opposite is true – the difference is negative and therefore the participants‘ perceived scores are greater than their measured scores. As the mean difference (±SD) was positive (0.3 ± 1.1 units) this indicates that, on average, there was a slightly positive bias – that is, on average, the measured scores for the n = 58 participants was slightly larger than the perceived scores by the group. In other words, these participants, on average, ever so slightly underestimated their category of fitness. Even so, from observation of Figure 1 there seems to be a large amount of random variation in these scores indicated by the spread of the plots about the zero line extended from the Y-axis. This is supported by the large SD for the differences compared to the relatively smaller mean 97
difference. The other important issue to observe in this plot is that there seems to be little evidence for heteroscedasticity – that is, it is likely that the mean and SD of the difference scores is consistent throughout the range of the scores – there is equal residual variance (homoscedasticity). About 95% of data points (errors) should lie within the limits -1.9 through 2.5 units. It should be highlighted that in this analysis there were an original n = 58 data points but only 13 appear in the Bland and Altman plot which means that there were many overlapping data points considered in this analysis. As far as the Normality of the errors is concerned, the histogram presented as Figure 10 includes a superimposed ‗best-fit‘ Gaussian curve over the bars of the graph. This bellshaped curve seems to be acceptable. In which case, it was concluded that parametric 95% limits of agreement, as described above, were an acceptable statistical method to employ in analysing these scores.
Histogram of Residuals Normal
25
Mean StDev N
0.3103 1.063 58
Frequency
20
15
10
5
0
-2
-1
0 1 Residuals
2
3
Figure 10. Normality of errors or residuals To put these limits into some practical context by taking the example of a hypothetical participant from the population of interest who, say, rated her own aerobic fitness as a score of 3, in which case, if her aerobic fitness were measured using the CST there would be a 95.0% probability that her aerobic fitness rating could be as low as 3.0 – 1.9 = 1.1 units of measurement (a rating that corresponds to ‗poor‘ aerobic fitness) or as high as 3.0 + 2.5 = 98
5.5 units of measurement (a rating above ‗excellent‘ aerobic fitness). It would probably be safe to conclude therefore that her assessment of her aerobic fitness was unacceptable. As the standard error (SE) of the limits is approximately √(3SDdiff2/n) for the upper and lower limits computed in this study: SE = √((3 1.12)/58) = √((3 1.21)/58) = ± 0.3 units of measurement. The 95% confidence interval for the limits of agreement is given by plus or minus (±) 1.96 SE = ±0.5 units of measurement. So for the lower limit (-1.9 units) the 95.0% confidence interval is -2.4 to -1.4 units and for the upper limit (2.5 units) the 95% confidence interval is +2.0 to +3,0. This means that whilst the lower and upper limits for the n = 58 participants was estimated, respectively as -1.9 and 2.5 units there is a 95% confidence that the limits in the population from which the 58 participants were drawn would lie between: lower = -2.4 and -1.4 units and upper = + 2.0 and + 3.0 units. The results suggested that women‘s perceptions of their fitness are moderately accurate and this is in agreement with the findings of Germain and Hausenblas (2006). For a health and fitness practitioner however this would provide unreliable information for decision making regarding lifestyle advice for CVD prevention.
The present study did not investigate the cognitive processes involved in forming selfperceptions although Germain and Hausenblas (2006) speculated that physical activity in itself leads to higher fitness, which in turn gives rise to higher self-perceptions of fitness. On this basis, it would appear reasonable to suggest that inactive women might similarly be selfaware and rate their fitness as below average or low, but this is speculative. Where it is important to account for aerobic fitness status, for example, where a health practitioner would want to estimate overall risk of CVD, the results of the present study cautiously suggest that objective measures of aerobic fitness would be useful. For example, a sub-maximal exercise test might provide cogent information for CVD risk estimation. This is particularly important with older women who are known to reduce their physical activity levels with age but who might hold erroneous perceptions of their fitness.
3.4.4 AEROBIC FITNESS AND CARDIOVASCULAR DISEASE RISK Measures of aerobic fitness were categorised according to norms of the CST (Sykes, 2005) as ‗excellent‘, ‗good‘, ‗average‘, ‗below average‘ and ‗poor‘, and adjusted for age. Table 9 illustrates categorisation according to criteria of the CST (Sykes, 2005).
99
Table 9, Categorisation according to norms of the Chester Step Test (n = 58).
Aerobic Fitness Category
Frequency
Percentage
Excellent
2
3.4%
Good
19
33.0%
Average
31
53.3%
Below average
6
10.3%
Poor
0
0.0%
The ‗average‘ and ‗below average‘ aerobic fitness for women in each age category of the CST (Sykes, 2005) are below a value of 9.0 METs with the exception of the 20 to 29 years age category where the ‗average‘ capacity is given as 9.1-11.1 METs (Sykes, 2005). However, Blair et al. (1989), found an asymptote occurring at 9 METs in women aged 20 years or more, below which, risk of all-cause and CVD mortality was significantly increased. Thirty-three women in the present study (57.0%) were assessed as having an aerobic fitness less than 9.0 METs (mean age (± SD) 50.2 ± 10.5 years). Moreover, women assessed as having aerobic fitness below the asymptote of 9.0 METs were not exclusively older. Six women (10.0%) 40 years of age or younger scored less than 9.0 METs and based on a 9.0 MET threshold this does not bode well in terms of the primary prevention of CVD. A lack of adherence to physical activity guidelines epitomised the activity status of those women found to be in the ‗below average‘ category according to norms of the CST (Sykes, 2005) with all but one reporting as either sedentary, or only partially active.
Using the thresholds suggested by Gulati et al. (2003) the results were similarly concerning. Gulati et al. (2003) found that exercise fitness independently predicted death from CVD in women aged over 35 years, with risk doubling in women whose aerobic fitness was between five and eight METs. Fourteen women (24.0%) in the present study aged over 35 years were assessed as having an aerobic fitness of 8.0 METs or less. None of these women reported full adherence to physical activity guidelines (Haskell et al., 2007). Furthermore, in relation to women‘s cardiovascular risk factor profile, Kokkinos (1995) found that moderate aerobic fitness equivalent to 10 METs or more is required for favourable lipids, blood glucose levels, blood pressure, and anthropometric measures. In the present study, 37 women (63.7%) were assessed as having an aerobic fitness below a 10 MET threshold indicating a level of aerobic fitness below minimum thresholds for a healthy cardiovascular risk factor profile. Although it is accepted that aerobic fitness is influenced by other variables than physical activity, for
100
example, smoking and genetic inheritance, poor aerobic fitness can indicate an inadequate level of physical activity and this epitomised women‘s physical activity status in the present study.
Table 10, shows the percentage of women who did not meet the thresholds for aerobic fitness as suggested in past research, and who therefore, were potentially at increased risk of CVD.
Table 10, Percentage of women at risk according to aerobic fitness thresholds Researcher End point Threshold Percentage at risk _________________________________________________________________________________ Blair et al. (1989)
All cause & specific cause mortality (n = 3120)
9.0 METs
57.0%
Kokkinos et al. (1995)
Favourable impact of aerobic fitness on CVD risk factors (n = 522)
10.METs
64.0%
Gulati et al. (2003)
Low aerobic fitness & increased risk from death from CVD in women > 35 yrs (n = 5721)
8 METs
24.0%
Kodama et al. (2009)
Quantitative relationship of aerobic fitness with CVD events & all-cause mortality in men and women (n = 102 980)
7.9 METs
19.0%
Comparison of MET values with thresholds for low, intermediate, and high aerobic fitness suggested by Kodama et al. (2009) confirmed a prevalence of concerning low of levels of fitness. Table 11. shows the percentages of participants‘ within categories.
Table 11, MET values for low, intermediate and high aerobic fitness (n = 58) __________________________________________________________________________ MET Values
Frequency
Percent
0-7.8
(Low fitness)
11
19.0%
7.9-10.8
(Intermediate fitness)
36
62.1%
10.8 or higher
(High fitness)
11
19.0%
Based on the lower threshold of less than 7.9 METs (Kodama et al., 2009), 19.0% of women were categorised as at increased risk for all-cause mortality and CVD. This suggests a potential value of aerobic fitness assessment in asymptomatic women as suggested by 101
Gulati et al. (2003), Sui et al. (2007) and Kodama et al. (2009) in order that preventative action can be taken if thought necessary. However, this is unlikely to occur unless women are highly motivated to seek out assessment of their fitness, or are referred for fitness assessment which is currently unrealistic for asymptomatic women within the setting of primary practice. Assessment of aerobic fitness can however be undertaken when an individual is referred by a GP to an exercise referral scheme. These schemes presently operate throughout Wales, the aim being to support people with medical conditions to improve their fitness with the intention of improving their health. This scheme however is not inclusive of unfit asymptomatic individuals. In the first instance the individual must come to the attention of his or her GP to receive a referral, although the results of study one of this thesis demonstrated that more than 90.0% of women attending their surgery had never discussed CVD prevention with their GP or practice nurse. A shift in focus with emphasis placed on identifying asymptomatic women with a low level of fitness might favourably alter the present situation where women come to the attention of fitness professionals in exercise referral programmes only after a diagnosis of CVD, and not before. Fitness assessment might be clinically informative in terms of CVD primary prevention in much the same way as screening for cholesterol abnormalities or hypertension and make a substantial contribution to individual CVD risk profiling. In the absence of this possibility, the objective of current campaigns such as ‗Health Challenge Wales‘ (WAG, 2004) that aim to increase physical activity levels within the general population continues to be a necessary health promotion strategy.
3.4.5 PHYSICAL ACTIVITY LEVELS There was an expected variability in the frequency of physical activity reported by women. The most expedient method of categorisation therefore was not in rigid terms of whether or not they adhered to physical activity guidelines (Haskell et al., 2007), but whether they were ‗fully active‘, ‗partially active‘, or ‗sedentary‘, as shown in Table 12. Table 12, Women‘s physical activity status categorised as fully active, partially active, or sedentary (n = 58) __________________________________________________________________________ Activity status
Frequency
Percent
Fully active
5 moderate, or 3 vigorous sessions a week, or combination
11
19.0%
Partially active
Not fully meet physical activity guidelines
44
75.8%
Sedentary
No moderate or vigorous physical activity
3
5.2%
102
If surveys of women‘s physical activity levels categorise women as either adhering to guidelines, or not, this fails to account for women who are habitually active, but who do not quite meet physical activity guideline expectations. This is salient because the protective effects of physical activity occur on a continuum with increasing levels having a strong, graded, inverse association with the risk of CVD events (Blair et al., 2001; Williams, 2001; Manson et al., 2002; Franco et al., 2005; Mora et al., 2006). In the present study, the three women (5.0%) who were sedentary were an obvious concern, however the 44 women (75.8%) reporting a partial adherence to physical activity guidelines were a further consideration, with some women on the cusp of meeting recommended activity levels. The results of study one of this thesis found that knowledge about the importance of an active lifestyle was excellent with 94.3% of women reporting that physical activity can help prevent CVD. Therefore, although the majority of women appear to know that an active lifestyle is health protective, and are partially physically active, they might not know the specific recommendations regarding the frequency, duration and intensity of physical activity necessary to benefit health. This was evident in the Health Survey for England (2009) where lack of knowledge about current guidelines for physical activity was reported with only nine percent of women able to specify a level equivalent to the CMOs minimum recommendations (2004). Twenty-three percent of women specified a level of physical activity greater than the minimum recommendations, and 68.0% of women either under-estimated how much physical activity adults should do, or did not know.
3.4.6 BODY MASS INDEX AND RISK OF CARDIOVASCULAR DISEASE The distribution of body mass index (BMI) based on categorisation of the WHO (2004), is shown in Table 13. Table 13, Body Mass Index Categories (n = 58) _______________________________________________________________________ Category
BMI
Frequency
Prevalence
-2
Underweight < 18.5 kg m 1 1.7% -2 Normal weight 18.5-24.9 kg m 24 41.3% -2 Pre-obese (overweight) 25-29.9 kg m 23 39.7% -2 Obese ≥ 30.0 kg m 10 17.2% ______________________________________________________________________________
Based on the criteria of BMI, 33 women (56.8%) were either overweight or obese. The results of the present study of women aged between 25 and 65 years therefore lend weight to alarming results of the Welsh Health Survey (WAG, 2009) which found a prevalence of 52.0% in women aged more than 16 years.
103
3.4.7 ABDOMINAL ADIPOSITY AND RISK OF CARDIOVASCULAR DISEASE Eighteen women (31.0%) met the target of a waist circumference (WC) less than 80 cm (31.5 inches) recommended by the International Diabetic Federation (IDF, 2005) and the Diabetic Association (2006). Twenty-five women had a WC between 80 and 88 cm (43.0%), and were categorised as ‗at high risk‘ of developing diabetes and CVD, and 15 women (26.0%) had a raised WC over 88 cm categorised as ‗at substantially higher risk‘. Data from the Health Survey for England (2007) found the mean WC in adult women of all ages was 86.5 cm. In the present study the mean WC for women aged between 25 and 65 years was 81.6 cm suggesting some concern because abdominal adiposity is strongly associated with insulin resistance, dyslipidemia and systemic inflammation (Calle et al., 2004; Berg et al., 2005). Furthermore, abdominal adiposity acts independently of overall adiposity (Snijder et al., 2006; Balkau et al., 2007). Although precision of measurement is required within a clinical or research setting, the measurement of WC can easily be accomplished by most women. In study one of this thesis however, the pilot study revealed that women were unaware of their WC highlighting the pertinence of the campaign by Diabetes UK, titled ‗Measure Up - are you at risk of diabetes‘ (Diabetes UK, 2011). This campaign aims to heighten awareness of the risks of abdominal adiposity, and explains to the general public how to measure their WC.
When combined BMI and WC measures were analysed according to advice contained in NICE guidelines (2006), 50.0% of the study sample were categorised as ‗at risk of CVD and other long term illnesses‘. Table 14. shows the risk distribution. Table 14, Risk of cardiovascular disease or other long term health problems based on combined body mass index and waist circumference (n = 58; NICE, 2006). Risk level:
No risk
Increased risk
High risk
Very high risk
50.0% (n = 29)
19.0% (n = 11)
19.0% (n = 11)
12.0% (n = 7)
Fifty percent of women were at risk of CVD and other long-term illness based on the criteria of combined BMI and WC. According to NICE guidelines (2006), women rated as ‗at risk‘ would benefit from advice on diet and physical activity from a health professional and this should be available to women particularly at times when weight gain is more likely, such as during pregnancy, when giving up smoking cigarettes, or at menopause. In light of the 104
prevalence of overweight and obesity in Welsh women of all ages this advice appears to skim the surface of an escalating problem and reflects the considerable challenge faced by the WAG and health care providers to reduce overweight and obesity in the Welsh population of women.
3.4.8 COMBINED BODY MASS INDEX AND LEVEL OF PHYSICAL ACTIVITY Table 15 illustrates the combined BMI and physical activity status of the participants. For this analysis the participants were categorised as active if they met current guidelines for physical activity (Haskell et al., 2007) and inactive if they only partial met them, or were sedentary. Table 15, BMI and level of physical activity (n = 58)
BMI and inactivity status
Under weight / inactive Normal BMI / active Normal BMI / inactive Overweight BMI / active Overweight BMI / inactive Obese BMI / active Obese BMI / inactive
Percentage
1.7% 12.0% 31.0% 3.4% 34.4% 1.7% 15.5%
(n = 1) (n = 7) (n = 18) (n = 2) (n = 20) (n = 1) (n = 9)
In each category of BMI more women were inactive than active and it is particularly concerning that only three of the 32 overweight or obese women reported full adherence to current physical activity guidelines (Haskell et al., 2007). A relationship between physical activity levels and the prevalence of obesity was found in the Health Survey for England (2009), with women who reported higher levels of physical activity less likely to be obese than less active women. The cardiovascular risk associated with low levels of physical activity and higher BMI were found by Mora et al. (2006) to be independently and jointly associated with inflammatory and lipid biomarkers for CVD. The most favourable profile consisted of a BMI within the normal range combined with adherence to physical activity guidelines (Mora et al., 2006). Only 12.0% of women in the present study met these criteria highlighting the necessity for a dual approach to body-weight management focussing on healthy eating and habitual physical activity.
105
3.5.1 CONCLUSION TO STUDY TWO The present study was exploratory with a small sample size and age restriction. Furthermore, the participants were asymptomatic white British women who had volunteered to participate and the majority reported as partially meeting guidelines for physical activity in the previous week. Generalisation of the results would therefore be naïve in light of these restrictions. However, the results generated several issues regarding women‘s perceptions of their fitness, their physical activity and measured aerobic fitness, and their overweight and obesity that merits discussion.
3.5.2 WOMEN’S ABILITY TO ESTIMATE THEIR AEROBIC FITNESSS Women were only moderately accurate at estimating their level of aerobic fitness. This is concerning for two reasons: it is unlikely that this would provide useful information to facilitate discussion between women and health-care providers about CVD prevention, and women who overestimate their aerobic fitness are unlikely to perceive any need to increase their physical activity levels for health reasons. Sixteen women (27.5%) overestimated their aerobic fitness with six having a measured aerobic fitness of 7.9 METs or less, indicating a level of fitness well below the threshold of 7.9 METs suggested by Kodama et al. (2009) to avoid a higher risk of all-cause and CVD mortality. This suggests that measured aerobic fitness within the clinical setting, and where not possible, objective measures of physical activity such as the use of pedometers, would be more useful as a basis for discussion between women and health care providers about CVD prevention, than relying solely on women‘s perceptions of how fit they think they are.
3.5.3 WOMEN’S AEROBIC FITNESS AND PHYSICAL ACTIVITY In accepting that women were moderately accurate at estimating their aerobic fitness the results also revealed that some women were not very aerobically fit. Nineteen percent failed to meet a threshold of 7.9 METs and on this basis, a concerning percentage of women presented with a potentially increased risk of CVD. Whilst it is accepted that CVD is associated with an array of modifiable and non-modifiable risk factors of which aerobic fitness is one, the result is concerning because atherosclerotic CVD causes approximately 20.0% of deaths in women before the age of 75 years, and is an important health issue for women (BHF, 2009). Measured against the norms of the CST (Sykes, 2005) only 3.4% of women were assessed as having ‗excellent‘ aerobic fitness and 33.0% were found to have
106
‗good‘ aerobic fitness. Sixty-four percent of women however were assessed as having ‗average‘ or ‗below average‘ aerobic fitness suggesting considerable opportunity for improvement.
The frequency of participation in physical activity varied considerably and the majority of women (75.8%) only partially met physical activity guidelines (Haskell et al., 2007). When considered in light of a dose-response relationship where higher fitness results from more rigorous and sustained engagement in habitual physical activity, partially active women are likely to accrue some health benefits compared to sedentary women. However, it would appear that women in the sample would benefit from more moderately intense or vigorously intense physical activity on a more frequent basis.
3.5.4 MEASURED OVERWEIGHT AND OBESITY Three criteria were adopted to investigate overweight and obesity consisting of BMI, WC, and combined BMI and WC. The results demonstrated cause for concern related to each criterion. Thirty-three women (57.0%) had a BMI classed as either overweight (25-29.9 kg m2
), or obese (≥ 30.0 kg m-2). Based on criteria of the IDF (2005) for a healthy WC, 25 women
(43.0%) were classified as ‗at high risk‘ of developing diabetes and CVD, and 16 women (26.0%) classed as ‗at substantially higher risk‘. The criteria of NICE (2006) for combined BMI and WC to evaluate heightened risk of CVD and other long-standing illness was no more optimistic with 29 women (50.0%) measuring over advised thresholds.
On an individual level, the health implications related to overweight and obesity are significant in that it is accepted they have adverse effects on other CVD risk factors and are independent risk factors for CVD (National Heart Lung and Blood Institute Obesity Task Force (NHLBI OTF), 1998; Manson et al., 1995; Calle et al., 1999; Hu et al., 2004; Dyer et al., 2005). The results demonstrated a worrying prevalence of overweight and obesity confirming the need for interventions outlined in the All-Wales Obesity Pathway (WAG, 2010) on multiple levels.
3.5.5 COMMUNICATING RISK The results of this study indicated cause for concern regarding modifiable risk factors for CVD of overweight and obesity, and low aerobic fitness or physical activity. The results indicated that women‘s judgments about their personal level of aerobic fitness should not be
107
relied on by health professionals. In discussion with patients, they might need to consider other avenues of assessment. The view of the WAG is that health is a shared responsibility with people actively involved in taking steps to avoid preventable disease (Wanless, 2004). This is good reason for robust strategies to be in place to support asymptomatic women minimize their risk of CVD; initially and preferably, through informed discussion with a health care provider regarding their overall cardiovascular risk factor profile. This particularly applies to women who lack awareness of their aerobic fitness and the relationship with the risk of CVD, or who are insufficiently active, who underestimate the health risk associated with overweight or obesity, and who lack knowledge of the association with development of longterm illness.
108
ABSTRACT The purpose of this study was to investigate women‘s perceptions of exercise benefits and barriers. Participants were drawn from one demographic locality within the UK and were informed of the study through venues where they met for employment or leisure. Specifically, the study investigated the perceptions of women aged between 25 and 65 years (n = 97, x ± SD: 48.5 ± 12.7 years), and women aged between 66 and 79 years (n = 31, x ± SD = 70.1 ± 3.9 years). The mean body mass index (BMI) for women aged between 25 and 65 years was 25.8 ± 4.2 kg m-2 and 50.0% were overweight or obese. Nineteen women (19.5%) aged between 25 and 65 years were classed as fully active, 73 women were partially active and five were sedentary according to current guidelines (Haskell et al., 2007). The mean BMI for women aged between 66 and 79 years was 25.3 ± SD 3.8 kg m-2 and 52.0% of women were overweight or obese. Six women (19.3%) were fully active, and 25 women (80.6%) were partially active. Participants completed an Exercise Benefits and Barriers Scale consisting of 43 items using a Likert format (Sechrist et al., 1985), a seven-day physical activity re-call question, and stature and body weight were measured in order to calculate body mass index (BMI). Partially active women perceived more exercise benefits than barriers. Physical performance and psychological outcome benefits were the most agreed with exercise benefits in women aged between 25 and 65 years but in women aged between 66 and 79 years social interaction was the most important. Preventative health benefits were not ranked highly across the spectrum of age. The most agreed with barrier across the age span was that of exercise as tiring or fatiguing, and restrictions caused by time and family responsibilities were also ranked highly. Significant differences between age groups were found for the subscales of ‗Life enhancement‘ (t = 3.35, p
