HOW BLACK AFRICAN AND WHITE BRITISH WOMEN PERCEIVE DEPRESSION AND HELP-SEEKING: A PILOT VIGNETTE STUDY

JUNE S.L. BROWN, SARAH J. CASEY, AMANDA J. BISHOP, MARTA PRYTYS, NAUREEN WHITTINGER & JOHN WEINMAN ABSTRACT Background: The detection of psychological problems of black African people has been found to be substantially lower, compared with white British and black Caribbean people. This may be due to differences in patients’ perceptions of illness. Little research has been carried out on factors that may influence the help-seeking behaviour of black Africans. Aims: To assess differences in the perceptions of depression of black African and white British women that may influence lower detection and to investigate whether there are ethnic group differences in reasons for not seeking formal help. Methods: A short quantitative illness perception measure, the Brief Illness Perception Questionnaire (BIPQ), was used in a community survey, using a standard text vignette methodology to control for variations in previous experiences of depression. Responses from women who indicated that they would not seek formal help for depressive symptoms were qualitatively analyzed. Results: Differences in perceptions of depression were found between black African (n = 73) and white British groups (n = 72) on five of the nine BIPQ dimensions. Black women were more likely to perceive depression to have less serious consequences; to be associated with fewer symptoms; to be less chronic; to be less amenable to treatment; and more frequently attributed depression to social factors. Over half the participants (n = 74) said they would not seek formal help for depressive symptoms. Six qualitative response categories emerged to explain non-consultation. The most common factor for both groups related to GP consultation difficulties. Significantly more white women cited preferring alternative help sources as a reason for nonconsultation. The greater number of black women citing anti-medication beliefs was marginally significant. There were no differences between the ethnic groups in their use of the remaining three categories: illness characteristics; service constraints; and stigma/shame. Conclusion: Differing perceptions of depression among black and white women could help explain GPs’ lower detection rates of depressive problems of black women. Differences in views about the formal help available may explain ethnic differences in help-seeking. Key words: ethnic, depression, perception of illness, detection, primary care, help-seeking

International Journal of Social Psychiatry. © The Author(s), 2010. Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav Vol 57(4): 362–374 DOI: 10.1177/0020764009357400



BROWN ET AL.: HOW BLACK AFRICAN AND WHITE BRITISH WOMEN PERCEIVE DEPRESSION 363

INTRODUCTION Given that most of the research on help-seeking by black people has investigated those with psychotic problems in the secondary services, help-seeking by black people for depressive problems in primary care is relatively under-researched. Some studies with black Caribbeans have shown higher rates of depression (Nazroo, 1997; Shaw et al., 1999) but a lower likelihood of receiving treatment compared with white people (Nazroo, 1997). While GP consultation differences have not always been found (Shaw et al., 1999), studies have shown that GPs are consistently less likely to detect psychological problems of black people (Odell et al., 1997; Shaw et al., 1999). Shortcomings in clinicians’ skills in detecting mental health problems have been reported (Goldberg & Huxley, 1992; Thompson et al., 2000). However, Odell et al. (1997) have suggested that detection problems may also reflect ‘different ways of expressing mental distress, different health beliefs and styles of help-seeking’. This fits in with the argument of Littlewood & Lipsedge (1982) that mental illness is culturally defined by western medicine. In a primary care study, Maginn et al. (2004) found that the psychological problems of black Africans were less likely to be detected and actively managed compared to those of black Caribbean and white British patients. They found that the strongest predictor of detection was the patient’s decision to talk to their GP and that black Africans were less likely to say they would discuss psychological problems with their GP. While no differences were found between black Caribbean and white British patients in prevalence, supporting findings by Weich et al. (2004), the probable prevalence rate was estimated to be significantly lower among African patients. To access illness beliefs, studies on other ethnic minorities, such as Indian women, have commonly used the semi-structured Short Explanatory Model Interview (Lloyd et al., 1998) or the full Explanatory Model Interview Catalogue (Weiss, 1997). A shorter quantitative measure widely used for physical health problems is the Illness Perception Questionnaire (IPQ) (Weinman et al., 1996). This is based on Leventhal’s Self-Regulation Model (SRM) (Leventhal et al., 1980) which postulates that patients attempt to act as problem-solvers and will develop their own personal model of their illness in order to find ways to cope with it. For physical health problems, consistent relationships have been found between illness perceptions, consultation and treatment adherence (Murphy et al., 1999; Petrie et al., 2002). More recently, illness perceptions have been studied in relation to mental health problems and have been found to relate to treatment adherence (Fortune et al., 2004; Holliday et al., 2005) as well as help-seeking for depression and anxiety (Edwards et al., 2007). It was decided to focus specifically on the illness perceptions of black African women for several reasons. One was the significantly lower rate of GP detection of psychological problems between black Africans and white British found by Maginn et al. (2004). The second was that Black Africans now comprise the fourth largest BME group, having doubled in size between 1991 and 2001 (Office of National Statistics, 2001). Third, black Africans have been found to have large health needs but under-utilize primary care services (Suckling, 2008). This study therefore aimed to assess differences in perceptions of depression of black African and white British women. Illness perceptions were measured using the Brief Illness Perception Questionnaire (BIPQ) (Broadbent et al., 2006) and a vignette methodology similar to that used by other community studies (Edwards et al., 2007; Jorm et al., 2000); To control for gender differences in distress and help-seeking (Moller-Leimkuhler, 2002) as well as illness beliefs (Edwards et al., 2007), only female participants were included.

364

INTERNATIONAL JOURNAL

OF

SOCIAL PSYCHIATRY 57(4)

As far as we know, no studies have investigated reasons why black Africans do not seek formal help. Help-seeking may be influenced by several factors, including feeling the GP did not understand (Pill et al., 2001) as well as differences in symptom presentation, and health beliefs held about depression and mental illness generally (Odell et al., 1997). Karlsen et al. (2005) and Bhui et al. (2005) have found evidence of discrimination and that fear of discrimination may lead to reluctance to seek help for psychotic problems. Stigma and negative views about medication (Weich et al., 2007) may also deter help-seeking. This study therefore also sought to qualitatively examine reasons why women would decide not to consult their GP if they experienced depressive symptoms, and whether there might be ethnocultural differences in this decision.

METHOD Design This study used a cross-sectional survey design with a convenience sample, similar to that successfully used by Edwards et al. (2007).

Vignette methodology A standard text was used for all participants to control for variations in previous experiences of depression. The vignette was taken from two studies (Jorm et al., 1997; Jorm et al., 2000) and represented a person with DSM-IV criteria of major depression. The original vignette was written in the third person but for the purposes of the study, was adapted and written in the second person to increase the likelihood of participants’ engagement. Participants were asked to imagine themselves in the position of the vignette character. ‘You have been feeling unusually sad and miserable for the last few weeks. Even though you are tired all the time, you have trouble sleeping nearly every night. You don’t feel like eating and have lost weight. You can’t keep your mind on your work and put off making decisions. Even day-to-day tasks seem too much for you. This has come to the attention of your boss who is concerned about your lowered productivity.’

Study setting and sample This was one of several studies conducted in public libraries in south London. It was decided to recruit in libraries for two main reasons. One was the need to sample from non-political or healthrelated settings, to maximize the likelihood of sampling individuals who might choose not to consult their GP even though they may be experiencing symptoms of depression or other mental health disorders. The second was that libraries are used by large numbers of the general public, who are likely to be reasonably representative of the population in terms of age. Ethical approval to conduct the study was obtained from the local Ethics Committee. Inclusion criteria specified women aged between 18 and 45 years, who were black African (to be referred to as black) and white British (to be referred to as white) women. It was decided to only study women who indicated previous experience of depression to reduce the variation in experience.



BROWN ET AL.: HOW BLACK AFRICAN AND WHITE BRITISH WOMEN PERCEIVE DEPRESSION 365

As the mean age of black people in the general population has been found to be much lower than that of white people in the UK (Office of National Statistics, 2001), age was controlled for by only including participants aged between 18 and 45 years old.

Measures Demographics and help-seeking behaviour questionnaire Background data collected included age, ethnicity and educational background. Ethnicity was selfdefined by participants. In the screening questionnaire, participants were asked to define themselves as: White – UK; White – Irish; White – other; Black – African; Black – Caribbean; Black – other; or Other. Information about previous depressive experiences was sought using the question: ‘Have you ever in your life been markedly depressed, that is for several weeks or more, you felt sad, lost interest in things and felt lacking in energy?’. Information about previous formal help-seeking was obtained using the question: ‘If you have, did you seek help/advice from a GP?’

Brief Illness Perception Questionnaire (BIPQ) The BIPQ is a short version of the IPQ-R (Moss-Morris et al., 2002) and comprises nine items. The first eight items assess beliefs about: Consequences of the illness (How much does your illness affect your life?); Timeline (How long do you think your illness will continue?); Personal control (How much control do you feel you have over your illness?); Treatment control (How much do you think your treatment can help your illness?); Illness symptom Identity (How much do you experience symptoms from your illness?); Emotional concern (How concerned are you about your illness?); Emotional upset (How much does your illness affect you emotionally?); and Illness coherence (How well do you understand your illness?). Responses are assessed on a 10-point Likert scale. The ninth item elicits Causal beliefs, and participants’ three most important causal factors are rank-ordered. In this study, only the main perceived cause was analyzed, and classified by two authors (JB and MP) into ‘medical’, ‘social’ and ‘psychological’. The original BIPQ required self-ratings, but in this study participants were asked to react to the experience of the depressed vignette character. Items were adapted to enable participants to use their own illness perceptions for the vignette character. Thus, ‘How much do you think your treatment can help your illness?’ was replaced by ‘How much do you think treatment can help these difficulties?’. One additional question was also asked: ‘Would you seek help/ advice from a GP for these difficulties?’. For those women who indicated that they would not seek help/advice from a GP, an open-ended question was asked: ‘Why would you NOT seek help/advice from a GP for these difficulties?’.

General Health Questionnaire (GHQ-12) The General Health Questionnaire (GHQ-12) (Goldberg, 1972) contains 12 items that assess current levels of distress. The range of scores is 0–12 and a cut-off score of 3 or more is used to indicate a ‘probable case’.

Procedure With the permission of the librarians, women were approached and invited to take part in a short research study. If they agreed, they were provided with a brief verbal and written explanation of the

366

INTERNATIONAL JOURNAL

OF

SOCIAL PSYCHIATRY 57(4)

study. A screening questionnaire was used to exclude women who were aged over 45, who were not black African or white British and who had not previously experienced depression. Informed consent was obtained from study participants before they completed the assessment questionnaires. Participants were able to complete questionnaires with sufficient privacy in the library area, usually taking 15–20 minutes.

Statistical analysis A power calculation indicated that a sample size of 69 in each group would have 80% power to detect group differences with a probability of 0.638 using a Mann-Whitney test with a 0.05 twotailed significance level. χ2 tests were used to analyze categorical data, and t-tests were used for normally distributed data. As BIPQ data were not normally distributed, Mann-Whitney U tests were used to compare groups on each sub-scale. For the qualitative analysis, a general inductive approach (Thomas, 2003), which combines elements of grounded theory (Glaser & Strauss, 1967) with the utilization of a priori categories based on previous research, was adopted to categorize the answers to the open-ended question ‘Why would you NOT seek help/advice from a GP for these difficulties?’. After the qualitative answers were read to identify emerging themes, an initial coding frame was developed and emergent themes were checked by independent raters to refine the basic categories for subsequent analysis. An inter-rater reliability analysis using the κ statistic was performed to determine consistency among raters. Following this, a multiple-response analysis was conducted using the χ2 statistic to analyze the categorical data. All analyses were conducted using SPSS 15.0 (SPSS Inc. Chicago, Illinois).

RESULTS Data were collected from 73 black and 72 white women aged between 18 and 45, who had previously experienced depression. In total, 933 participants were approached and 512 refused, giving a participation rate of 45.1%. The most common reason for refusal was that they were just popping into the library and did not have time. Although 421 women agreed, 276 (65.6%) did not meet the inclusion criteria, leaving 145 in the study sample.

Illness perceptions (n = 145) The black and white women did not significantly differ on demographic details (p > 0.05), but black women were significantly more distressed on the GHQ. The ethnic groups did differ in whether they had actually consulted their GPs when they were depressed, with black women being less likely to have previously consulted their GPs (Table 1). On the BIPQ, ethnic differences were found on five dimensions (Table 2). Black women held stronger beliefs that depression would not seriously affect their lives (Consequences), be less chronic (Timeline), and less amenable to treatment (Treatment control). They also associated fewer symptoms with depression (Identity) and were more likely to perceive it as socially rather than medically caused (Causes). No ethnic differences were obtained in response to the question ‘Would you seek help from a GP for these difficulties?’, which was interpreted as a willingness to seek help.



BROWN ET AL.: HOW BLACK AFRICAN AND WHITE BRITISH WOMEN PERCEIVE DEPRESSION 367 Table 1 Characteristics of participants in total sample and of participants who had been previously depressed, by ethnic group Previously depressed participants (n = 145) Black (n = 73)

GHQ Age (mean) Educational qualifications   No formal qualifications   GCSEs/A levels/similar   Further education   Other Previous help-seeking   Depression and had sought help   Depression and did not seek help



White (n = 72)

5.63 31.32



Analyses

3.96 33.33

t = 1.59, df = 143** t = –1.65, df = 143 χ2 = 2.18, df = 3

  5 (6.9%) 17 (23.6%) 45 (62.53%)   5 (6.9%)

  4 (5.6%) 25 (34.7%) 39 (54.2%)   4 (5.6%)





χ2 = 4.7, df = 1* 30 (41.7%) 42 (58.3%)

43 (59.7%) 29 (40.3%)

*** p