NU520604.qxd

11/14/2003

2:18 PM

Page 361

Hispanic Chronic Disease Self-Management A Randomized Community-Based Outcome Trial Kate R. Lorig















Philip L. Ritter

Background: In light of health disparities and the growing prevalence of chronic disease, there is a need for community-based interventions that improve health behaviors and health status. These interventions should be based on existing theory. Objective: This study aimed to evaluate the health and utilization outcomes of a 6-week community-based program for Spanish speakers with heart disease, lung disease, or type 2 diabetes. Method: The treatment participants in this study (n  327) took a 6-week peer-led program. At 4 months, they were compared with randomized wait-list control subjects (n  224) using analyses of covariance. The outcomes for all the treatment participants were assessed at 1 year, as compared with baseline scores (n  271) using t-tests. Results: At 4 months, the participants, as compared with usual-care control subjects, demonstrated improved health status, health behavior, and self-efficacy, as well as fewer emergency room visits (p  .05). At 1 year, the improvements were maintained and remained significantly different from baseline condition. Conclusions: This community-based program has the potential to improve the lives of Hispanics with chronic illness while reducing emergency room use. Key Words: chronic disease 䡠 Hispanic 䡠 self-efficacy 䡠 selfmanagement

T

omando Control de su Salud (Taking Control of Your Health), a community-based program for Spanish-speaking Hispanics with chronic diseases, resulted from the convergence of three trends. The first trend is the growing health disparity between Hispanics and non-Hispanics. For example, studies indicate that the prevalence of diabetes is nearly two times greater among Hispanics than in non-Hispanic groups (National Diabetes Information Clearinghouse, 2002). Diabetes also is medically more severe among Hispanics (Harris, Klein, Cowie, Rowland, Nursing Research November/December 2003 Vol 52, No 6



Virginia M. González

& Byrd-Holt, 1998). In studies of respiratory diseases, data indicate that although the prevalence of these diseases is variable among different Hispanic subgroups, Hispanics tend to be hospitalized more frequently for asthma than either African Americans or Whites (Carr, Zeitel, & Weiss, 1992; De Palo, Mayo, Friedman, & Rosen, 1994; Schulman & Glaxo Wellcome Inc., 1998). In Los Angeles, Hispanics have the second highest mortality for cardiovascular disease of the four major ethnic minority groups in Los Angeles County (Centers for Disease Control, 1999; Haywood, 1990). Cardiovascular disease, the leading cause of death for Hispanics, is declining at a slower rate in this group than in the remaining populations (Council of Scientific Affairs, 1991; Furino & Muñoz, 1991; HayesBautista, Baezconde-Garbanati, Schink, & Hayes-Bautista, 1994). The second trend is the increasing prevalence of chronic diseases accompanied by comorbid conditions (Wu & Green, 2000). One in five Americans has a chronic condition (Wu & Green, 2000). This increases to 84% for people 65 years of age or older. Among older Americans, 62% have comorbid conditions (Wu & Green, 2000). It is not unreasonable to believe that the data are similar for the U.S. Hispanic populations. These factors become especially important considering the third trend, that one third of Hispanics are without any form of health insurance (Eberhardt et al., 2001). Approximately 32.8 million Hispanics live in the United States (Therrien & Ramirez, 2000). During the past decade, chronic disease self-management education programs have demonstrated their impact on health behaviors, health status, and healthcare utilization (Brown et al., 2000; Lorig, González, & Ritter, 1999; Lorig, Sobel et al., 1999; Clark et al., 1992; Glasgow et al.,

Kate R. Lorig, DrPH, is Professor (Research). Philip L. Ritter, PhD, is Programmer Analyst. Virginia M. González, MPH, is former Research Associate, Department of Medicine, Stanford University School of Medicine, Stanford, California.

361

NU520604.qxd

11/14/2003

2:18 PM

Page 362

362 Hispanic Chronic Disease Self-Management 1997; Lorig et al., 2001). For example, a 6-week arthritis self-management program in a 4-month randomized trial demonstrated that participants increased their practice of exercise while reducing their pain (Lorig, Lubeck, Kraines, Seleznick, & Holman, 1985). In a 4-year longitudinal study, this same intervention resulted in a 19% reduction in pain and a 42% reduction in outpatient visits to physicians (Lorig, Mazonson, & Holman, 1993). When the 6week arthritis program was culturally adapted and offered to Spanish speakers in a randomized trial, the 4-month results were similar to those for the English speakers. No significant deteriorations occurred between 4 months and 1 year (Lorig, Gonzales & Ritter, 1999b). To identify the mechanisms through which these programs achieved effects, the relations among gains in knowledge, adoption of self-management behaviors, and health outcomes were examined (Lorig, Seleznick et al., 1989). Correlations were either weak or absent (Pearson’s r, 0.0-0.15). Semistructured interviews with English-speaking program participants followed by theme analyses showed that those whose health status improved believed they had some control over their symptoms, whereas those who did poorly believed that they could exert very little control (Lenker, Lorig, & Gallagher, 1984). This suggestive finding led to an exploration of psychological determinants of health and the possibility that the participants’ perceived selfefficacy in coping with arthritis was an important mediator. An instrument was designed to measure perceived selfefficacy, and in two subsequent studies, a positive correlation with health status was found (Lorig, Chastain, Ung, Shoor, & Holman, 1989). On the basis of these findings, the program was revised to emphasize the efficacyenhancing strategies of skills mastery, modeling, reinterpretation of symptoms, and social persuasion (Bandura, 1997, 2000; Lorig & González, 1992). Participants in the revised program demonstrated greater improvements than participants in the original program. More recently, a chronic disease self-management program based on self-efficacy theory was evaluated for nearly 1,000 subjects with heart disease, lung disease, stroke, or arthritis (Lorig et al., 1999c). Outcomes of the 6-month randomized trial and the 2-year longitudinal follow-up evaluation demonstrated that participants had significant increases in health behaviors and health status as well as reduced healthcare use (Lorig et al., 2001; Lorig et al., 1999c). Self-efficacy also increased and was the one factor that best predicted health status outcomes. The qualitative work of Hunt, Valenzuela, and Pugh (1998) is of special interest. These researchers found that for Latino patients with non-insulin-dependent diabetes, the relation between behavior and illness is not so much determined by self-care attitudes as by the experience Latinos have trying to gain control over their disease. This finding supported theory grounded in control beliefs, which proposes that health behavior is strongly influenced by the belief that one can exercise some measure of control over one’s health condition. The findings of Hunt et al. (1998) suggested that behavior may be determined by a reciprocal interaction between behaviors and a person’s experience with these behaviors. The self-efficacy theory, which proposes such a reciprocal interaction, may offer a

Nursing Research November/December 2003 Vol 52, No 6

strong theoretical framework for building Latino diabetes self-management programs. This finding is supported by Abraido-Lanza (1997), who found that self-efficacy contributes to psychological well-being among Latinas with chronic illness. Based on the needs of the Spanish-speaking community, especially the growing prevalence of comorbid conditions, and on the aforementioned theoretical work, Tomando Control de su Salud was developed as one step toward bringing chronic disease self-management education to the Spanish-speaking community. Individuals with serious diseases of both high and growing prevalence in the Spanish-speaking community were chosen for the study. The aims of this study were • to develop a 6-week Spanish language self-management program for individuals with chronic disease (coronary artery disease, chronic obstructive pulmonary disease including asthma and chronic bronchitis, or type 2 diabetes) and to evaluate its effects on self-management behaviors, symptoms, health status, healthcare utilization, and self-efficacy. • to evaluate the extent to which beneficial outcomes are maintained in a 1-year cohort study.

The Intervention Similar to the successful English Chronic Disease SelfManagement Program (CDSMP), Tomando Control de su Salud (Tomando) is a 14-hour community-based program given in 21/2-hour sessions over 6 weeks (Lorig, González, & Laurent, 1999). It was developed in Spanish on the basis of several focus groups involving Spanish speakers who had heart disease, lung disease, or type 2 diabetes. In addition, focus groups were held with health professionals, most of whom were not Hispanic, to develop the key health messages to be taught. Two trained peer leaders taught Tomando in community settings such as churches, neighborhood centers, and clinics. The class sizes ranged from 10 to 15 people including participants’ family and friends. A typical class consisted of 11 people with chronic diseases and 2 to 4 significant others. Family and friends were not participants in the research. Most of the leaders had one or more chronic conditions. All aspects of the Tomando program including recruitment, data collection, leader training, and course activities were conducted in Spanish. The leaders taught from a standardized protocol that detailed both the course content and process (Marin, Gonzalez, & Lorig, 2000). Each leader received 4 days of training in the use of this protocol. Included in this training were two practice teaching sessions. Project staff conducting the training evaluated the final practice teaching as a means of determining whether those in training would be allowed to teach a course. Although the format of the Tomando program is similar to both the English CDSMP and the Spanish Arthritis Self-Management Program, it is not a direct translation of the former. Important cultural adaptations of concepts, content, and process were made on the basis of the information from the focus groups. For example, there is extensive material on healthy eating, selection of products, por-

NU520604.qxd

11/14/2003

2:18 PM

Page 363

Hispanic Chronic Disease Self-Management 363

Nursing Research November/December 2003 Vol 52, No 6

tion size, and meal planning. Exercise is performed in class using a culturally appropriate audiotape, (Nacif de Brey, 1995a), and there is a section on family communications. As part of Tomando, participants receive a book entitled Tomando Control de su Salud written in broadcast Spanish (the Spanish used by national media in the United States) (González et al., 2002). This book contains all the Tomando content as well as chapters specific to each of the target conditions. The participants also receive an audio exercise tape, an illustrated booklet of the exercise routines, and an audio relaxation tape (Nacif de Brey, 1995b). The program is taught using activities to enhance self-efficacy. These activities involve leaders modeling for participants and participants modeling for each other, weekly action plans or “propósitos semanales” to enhance skills mastery, discussion of the meaning and causes of common symptoms to assist with the reinterpretation of these symptoms, and group work on menu planning and problem solving as forms of social persuasion. No disease-specific content is taught. According to a weekly action plan, sev-

Sessions

1

Difference between acute and chronic illness



Healthy eating Exercise

Methods The participants (all Spanish speakers who applied) were representative of Hispanics with chronic diseases in the northern California area. Specifically, the majority were born in Mexico. They were eligible for inclusion in the study if they had heart disease, lung disease, or type 2 diabetes. Other diagnoses were allowed as well. Participants were excluded if they had been treated for cancer during the preceding year. They were recruited via community outreach to churches, community centers, and clinics in 4-month cohorts over a period of 3.5 years. At entrance to the study and before randomization, the study participants completed informed consent and baseline questionnaires either by mail or telephone. Previous

2

3

4













Problem solving Action planning

eral topics that build on each other are presented: exercise (4 weeks), positive thinking (1 week), nutrition (4 weeks), relaxation techniques (2 weeks), and problem solving (5 weeks) (Figure 1).



5

























Breathing problems



Relaxation techniques





Depression



Self-talk (positive thinking)



Family relationships and making treatment decisions



Medications



Working with your healthcare provider deciding on your treatment plan

6





Evaluating alternative treatments



Planning for the future



Celebration of success



FIGURE 1. Overview of Tomando Control de su Salud content.

NU520604.qxd

11/14/2003

2:18 PM

Page 364

364 Hispanic Chronic Disease Self-Management studies have shown that the quality of the data collected by these two methods is similar (González, Stewart, Ritter, & Lorig, 1995; Marin, VanOss-Marin, & Pérez-Stable, 1990). In most cases, diagnoses were verified by the participant’s physician. When it was not possible to contact the physician, diagnoses were self-reported. The investigators were not involved in delivering the intervention, and those delivering the intervention were not involved in outcome assessment. The outcome assessors were blinded to participant status unless participants mentioned taking part in the intervention. To ensure that no fewer than 10 or more than 15 participants attended the program, as participants applied for Tomando at a specific site, randomization with a different ratio of intervention to control subjects was conducted at each site depending on the number of applicants to that site. Overall, intervention and usual care participants were assigned in a 3:2 ratio. Intervention participants took Tomando immediately, whereas the usual-care participants were wait-listed for 4 months. The control subjects received no intervention during this period. Altogether, 58 Tomando programs were offered. Data were collected again at 4 months, after which the usual-care subjects were offered Tomando. Data were collected again for all the participants at 4 months and 1 year after they began Tomando. Measures Outcome measures were chosen to evaluate the four main classes of outcomes: health behaviors, health status, healthcare utilization, and self-efficacy. The specific measures were based on the problems that had been identified in the participant focus groups as well as theoretical underpinnings of the study. Instruments were translated, back-translated, and standardized using a consensus meeting of all translators. The instruments then underwent standard psychometric testing including tests for internal consistency and test-retest reliability, item convergence, and discriminant validity. Health behaviors included a physical activities scale, which measured total minutes per week the participants spent doing flexibility, strengthening, and aerobic exercises, and a single question asking the frequency and type of relaxation techniques used. This scale had been developed for the authors’ earlier Arthritis Self-Management studies (González et al., 1995). Finally, a 4-item scale to measure communication with physicians was developed for the Chronic Disease Self-Management Study. The Spanish version had a coefficient alpha of .80 (n  147) and a test-retest validity of .86 (n  20). Health status measures included the self-rated health item from the medical outcomes studies and visual numeric scales for pain and fatigue. The health distress and role function scales also were developed for the CDSMP study (Lorig et al., 1996). In Spanish, they have coefficient alphas of .86 and .92, respectively (n  147). Healthcare utilization, visits to physicians, emergency department visits, and days in the hospital over the preceding 4 months were measured by self-report. These measures had test-retest validities ranging from .58 for emergency room visits to 1.00 for hospitalization (n  20).

Nursing Research November/December 2003 Vol 52, No 6

The 4-item self-efficacy scale asked participants concerning their certainty of controlling the fatigue, pain, emotional distress, and other symptoms caused by their disease in order to perform daily activities. The Spanish scale had a coefficient alpha of .85 (n  147) and a test-retest validity of .80 (n  20). In a separate study, we found that self-reported utilization and chart audit each had weaknesses (Ritter et al., 2001). Whereas self-reports resulted in some systematic underreporting, chart audits sometimes misrepresented visits. For example, the chart audit recorded 22 physician visits for a patient who reported 2 physician visits, but who also had weekly allergy injections. At baseline, in addition to the standard demographic information of age, gender, years of education, country of birth, and marital status, the Marin Short Acculturation Scale was administered to assess acculturation as determined by language use and preference (Marin & VanOssMarin, 1991). The Acculturation Scale is scored 1 (monolingual Spanish) to 5 (monolingual English). This scale was used because it was hypothesized that outcomes might be affected by level of acculturation. It should be noted that although bilingual individuals were not excluded from the study, all subjects had to speak and understand Spanish because it was the only language of instruction. Data Analysis In the 4-month randomized study, analysis of covariance was used to compare intervention and usual care groups for each outcome variable at 4 months after the same variable was used at baseline and control was used for age, gender, education, acculturation, and number of chronic conditions. A 1-year longitudinal study also was conducted. Paired t-tests were used to determine whether changes between baseline and 1 year differed from zero. All analyses were repeated as intent-to-treat analyses. In these analyses, all the participants were included, and where data were missing, the last known data were substituted.

Results Altogether, 551 individuals agreed to participate in the program. After completing an informed consent form and a baseline questionnaire, 327 of these participants were randomized to intervention and 224 to serve as usual-care control subjects. A 4-month questionnaire was completed by 265 treatment participants (81%) and 178 usual-care control subjects (79%) (Figure 2). The participants’ demographics (Table 1) showed that 55% were married, 94% were foreign born, and 79% were female. Their mean education was 7.6  4.9 years, with 14% reporting fewer than 3 years and 17% reporting more than 12 years of education. The mean age of the participants was 57.0  13.9 years. They had an average of 1.9 diseases. Among the participants, 19% had heart disease, 52% had hypertension, 45% had diabetes, and 19% had lung disease including asthma, chronic bronchitis, and emphysema. In addition, 49% reported various other conditions including hypolipidemias (28%) and arthritis (15%). The participants had a mean of 1.37  0.65 on the 5-point Marin Acculturation Scale, with 1

NU520604.qxd

11/14/2003

2:19 PM

Page 365

Nursing Research November/December 2003 Vol 52, No 6

Hispanic Chronic Disease Self-Management 365

FIGURE 2. Subjects.

indicating monolingual Spanish and 5 indicating monolingual English. According to their self-reports, 38% of the participants had no form of medical insurance. Of the six sessions, intervention participants completed a mean of 4.3  1.8. When the baseline data for the dropouts and nondropouts were compared, there were few differences. The only significant difference between those who completed the 4-month questionnaire and those who did not was that the completers were slightly less likely to have smoked at baseline than the dropouts (24% versus 30%; p.05). When those who completed the 1-year questionnaire were compared with those who did not, the findings showed that the completers had higher self-efficacy at baseline (6.4

versus 5.9 on a 10-point scale; p .05). The treatment and control groups also were compared at baseline for differences in health behavior, health status, and utilization variables. There were no other significant differences between the groups at baseline. At baseline, there were no significant differences between the treatment and control groups for any of the demographic variables. However, the treatment groups had a statistically higher number of diseases than the control group at baseline (1.9 versus 1.7; p  .01). Thus, the number of diseases was included as a covariate, along with the demographic variables, in the analyses of covariance that compared treatment and control groups on outcome variables at 4 months.

NU520604.qxd

11/14/2003

2:19 PM

Page 366

366 Hispanic Chronic Disease Self-Management

Nursing Research November/December 2003 Vol 52, No 6

TABLE 1. Baseline Demographic Variable

Mean years of age (SD) Number female Mean years of education (SD) Number married Number employed Number retired Number with any insurance (including Medical, Medicare) Total number of disease conditions Number born U.S. Number born Mexico Number born Central America Number born South America Mean Marin Acculturation Scale

Randomized Comparison at 4 Months Table 2 shows the comparison between the treatment and usual-care control groups 4 months after baseline. There were improvements in health behaviors, health status, and self-efficacy as well as reductions in healthcare utilization for the participants, as compared with the usual-care control subjects (p  .05). There were no differences in tobacco use or days of hospitalization. Intent-to-treat analyses also were conducted with the assumption that those who did not complete 4-month surveys would have had no change in their baseline values. The results were nearly identical to those of the original analyses with one exception. Instead of the treatment group showing significantly fewer visits at 4 months (p  .021), the difference in changes in the number of physician visits showed only a trend (p  .057). Retention of Benefits at 1 Year To determine the long-term retention of the effects found at 4 months, the outcomes were examined 1 year after baseline for all who had taken the program (Table 3). Because the control participants had now crossed over, this was no longer a controlled study. This analysis used t-tests to determine the probability that the 1-year change scores were significantly different from zero (no change). All health behaviors and health status outcomes showed continued significant improvement, as compared with baseline (p  .001). Self-reported physician visits and hospitalizations remained unchanged from baseline. Emergency room visits in the last 4 months of the year after the program were significantly fewer than in the 4 months before the beginning of the course (p  .01). The analyses were run, with inclusion of all those eligible for the 1-year study, using last data for missing data.

Treatment n327

Control n224

Total N551

56.6 (13.4) 260 (80%) 7.8 (5.0) 186 (57%) 96 (29%) 67 (20%) 208 (64%)

56.1 (14.6) 178 (79%) 7.5 (4.8) 118 (53%) 68 (30%) 42 (19%) 132 (59%)

57.0 (13.9) 438 (79%) 7.6 (4.9) 304 (55%) 164 (30%) 109 (20%) 340 (62%)

1.9 (0.85) 20 (6%) 208 (65%) 71 (22%) 19 (8%) 1.39 (0.67)

1.7 (0.79) 11 (5%) 145 (64%) 50 (22%) 17 (6%) 1.34 (0.63)

1.9 (.83) 31 (6%) 353 (64%) 121 (22%) 36 (7%) 1.37 (0.65)

Although the mean values for changes from baseline differed from those in the 4-month analyses, the probabilities were unchanged using three decimal places (Table 3).

Discussion After program completion, there were consistent positive results in health behaviors and health status both at 4 months and 1 year. Thus, it appears that participants increased their healthful behaviors and also improved in health status. In addition, their self-efficacy in managing their conditions was enhanced. Nevertheless, the possibility cannot be ruled out that those who did not benefit from the course were more likely to have dropped out. Thus, the results apply most strongly to those willing to participate in a program and complete questionnaires after participation. Tomando affected healthcare utilization. Although there were no significant changes in hospitalizations or physician visits, emergency room visits were significantly lower at 4 months and remained significantly lower at 1 year than at baseline. Populations with less access to standard healthcare often use the emergency room as a source of primary care (Beland, Lemay, & Boucher, 1998). This result suggests that Tomando participants made better decisions about where and when to seek healthcare. The reduction in emergency room visits was small, with only a .2 difference in visits, as compared with control subjects in the first 4 months after the program, and .12 fewer visits in months 8 to 12, as compared with the 4 months immediately before entrance to the program. The high cost of emergency room visits suggests that these differences are important. Few public health self-management interventions are available for Spanish-speaking populations with single or

NU520604.qxd

11/14/2003

2:19 PM

Page 367

Hispanic Chronic Disease Self-Management 367

Nursing Research November/December 2003 Vol 52, No 6

TABLE 2. Mean Health Status, Behavior, Healthcare Utilization, and Self-Efficacy Baseline Values and 4-Month Changes Baseline

4-Month Change Treatment N  265

Control N  178

Probability (Effect Size of Difference)

Intent to Treat Probability

Treatment N  265

Control N  178

3.80 (0.753)

3.80 (0.754)

0.392 (0.999)

0.028 (0.833) .0001 (0.48)

.0001

2.33 (1.44)

2.27 (1.43)

0.743 (1.62)

0.069 (1.57)

.0001 (0.47)

.0001

4.95 (3.13)

5.07 (3.40)

1.24 (3.66)

0.376 (3.63)

0.001 (0.27)

0.002

4.82 (3.46)

4.68 (3.55)

1.26 (4.11)

0.463 (3.95)

0.014 (0.23)

0.016

1.08 (1.11)

1.11 (1.12)

0.392 (1.17)

0.108 (1.28)

0.0003 (0.26)

0.002

106 (123) 1.69 (143)

96.2 (103) 1.59 (1.41)

63.7 (172) 0.700 (1.65)

31.0 (132) 0.223 (1.51)

0.001 (0.28) .0001 (0.34)

0.001 .0001

0.447 (1.49)

0.371 (1.46)

0.833 (2.35)

0.213 (1.78)

.0001 (0.71)

.0001

Health statuses Self-reported health (0–23,   better) Health distress (0–5, better) Fatigue (1–10,  better) Pain/physical discomfort (0–10,  better) Role function (0–7,   better)

Behaviors Exercise (# minutes week) Communication with physician (1–3,  better) Mental stress management (times in last week) Currently use tobacco

6% (.239)

4.5% (0.208)

0.015 (0.151)

0.011 (0.106)

0.964 (0.02)

0.997

2.78 (2.71)

2.69 (2.25)

0.475 (2.78)

0.034 (2.44)

0.021 (0.18)

0.057

0.275 (0.682)

0.242 (0.649)

0.083 (0.622)

0.002 (0.29)

0.005

0.151 (0.681)

0.393 (3.13)

0.011 (0.836)

0.079 (3.54)

0.126 (0.04)

0.481

6.28 (2.65)

5.89 (2.67)

1.157 (3.08)

0.719 (3.09)

0.0006 (0.16)

0.0006

(percent) Health care utilization Physician visits (number within past 4 months) ER visits (number within past 4 months) Hospital days (number within past 4 months)

Self-efficacy (1–10,   better)

0.101 (0.722)

Note. Standard deviations are shown in parentheses. Probabilities are that the outcomes are the same for the control versus treatment participants. They are calculated using ANCOVA models estimating 4-month variables from baseline variables and intervention group, with age, sex, education, total number of diseases, and whether married as covariates. Intent-to-treat probabilities assume no change at 6 months from baseline for those with no 4-month data (N327 for treatment, N224 for controls). The effect size reported is the difference between the change scores of the treatments and usual care controls divided by the pooled standard deviation at baseline.

combined chronic conditions. This study is possibly the first to provide an intervention for Spanish speakers that combined different chronic conditions into a single intervention. Although the medical interventions for these conditions may differ, the self-management skills are similar. Combining people with different chronic conditions into one educational intervention causes problems for participants, who often do not understand the term “chronic condition” (i.e., “I do not have a chronic condition; I have diabetes and hypertension”). Other characteristics of this intervention did not fit easily into more traditional healthcare. Tomando is taught by peer leaders, which requires a detailed protocol as well as training. In addition, health

professionals are concerned that participants will not be given correct information. It is for this reason that leaders always teach in pairs. Thus, they can act as a check and balance for each other. In reality, there are few problems with leaders not following the protocol or voiced concerns from participants about being taught by peers. The use of peer instructors has several advantages. The instructors are enhancing self-efficacy while serving as excellent models, allowing more programs at times and places outside of normal working hours, allowing more courses than could be offered if the program were more dependent on the small supply of Spanish-speaking health professionals, and adding more community members who have knowledge about chronic conditions.

NU520604.qxd

11/14/2003

2:19 PM

Page 368

368 Hispanic Chronic Disease Self-Management

Nursing Research November/December 2003 Vol 52, No 6

TABLE 3. Mean Health Status, Behavior, Health Care Utilization and Self-Efficacy Baseline Values and One-Year Changes

Health statuses Self-reported health (0–23,   better) Health distress (0–5,   better) Fatigue (1–10,  better) Pain/physical discomfort (0–10,  better) Role function (0–7,   better) Behaviors Exercise (numbers of minutes/week Communication with physician (1–3,  better) Mental stress management (times in last week) Currently use tobacco (percent) Health care utilization Physician visits (number within past 4 months) ER visits (number within past 4 months) Hospital days (number within past 4 months) Self-efficacy (1–10,  better)

Baseline N271

One-Year Change N271

3.73 (0.767) 2.24 (1.41) 4.77 (3.14) 4.49 (3.48) 1.01 (1.08)

0.277*** (0.935) 0.788*** (1.52) 1.34*** (3.42) 1.31*** (4.42) 0.394*** (1.10)

103 (104) 1.73 (1.42) 0.378 (1.42) 0.063 (0.244)

59.0*** (148) 0.732*** (1.68) 0.621*** (2.48) 0.011* (0.104)

2.75 (2.57) 0.269 (0.676) 0.295 (2.54) 6.44 (2.55)

0.146 n.s. (2.53) 0.119** (0.752) 0.045 n.s. (1.71) 1.17*** (3.10)

Note. Standard Deviations are shown in parentheses. n.s.  not significant. Probabilities are calculated using t-tests and are the likelihood that the outcomes were actually unchanged from baseline. All probabilities also calculated as intent-to-treat probabilities, which assume no change at 12 months from baseline for those eligible for 12-months but with no 12-month data (N541). The results from the intent-to-treat probabilities were identical. *p  .05 **p  .01 ***p .0001

Tomando Control de su Salud has demonstrated improvements in health behaviors, health status, and healthcare utilization in both short-term (4-month) and long-term (1-year) trials. The training and leader manuals as well as the participant materials are standardized to promote intervention fidelity. This study suggests that a community-based intervention requiring little technology and focused on the growing problem of chronic disease and comorbidity in a Spanish-speaking population can have positive effects, improving health behaviors, self-efficacy, and health status while decreasing healthcare utilization. ▼ The authors thank Maria Marin, formerly of the Stanford University School of Medicine, and Mirna Sanchez, Stanford University School of Medicine, for their invaluable assistance at many stages of this study; and our many volunteer group leaders. This program was supported by NINR grant 5R01NOR4438 and by University of California Berkeley Tobacco Related Disease Research Program grant 6LT0107. The Web site for the Stanford Patient Education Research Center is http://patienteducation. stanford.edu. The instruments used in this study are available there. Corresponding author: Kate R. Lorig, DrPH, 1000 Welch Road, Suite 204, Palo Alto, CA 94304 (e-mail: [email protected]).

References Abraido-Lanza, A. F. (1997). Latinas with arthritis: Effects of illness, role identity, and competence on psychological well being. American Journal of Community Psychology, 25(5), 601-627. Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman. Bandura, A. (2000). Health promotion from the perspective of social cognitive theory. In P. Norman, C. Abraham, & M. Conner (Eds.), Understanding and changing health behaviour (pp. 299-339). Reading, UK: Harwood. Beland, F., Lemay, A., & Boucher, M. (1998). Patterns of visits to hospital-based emergency rooms. Social Science and Medicine, 47(2), 165-179. Brown, S. A., Harrist, R. B., Villagomez, E. T., Segura, M., Barton, S. A., & Hanis, C. L. (2000). Gender and treatment differences in knowledge, health beliefs, and metabolic control in Mexican Americans with type 2 diabetes. Diabetes Education, 26(3), 425-438. Carr, W., Zeitel, L., & Weiss, K. (1992). Variations in asthma hospitalizations and deaths in New York City. American Journal of Public Health, 82(1), 59-65. Centers for Disease Control. (1999). Statistics: Diabetes surveillance, 1999. Atlanta, GA: National Center for Chronic Disease Prevention and Health Promotion. Clark, N. M., Janz, N. K., Becker, M. H., Schork, M. A., Wheller,

NU520604.qxd

11/14/2003

2:19 PM

Page 369

Nursing Research November/December 2003 Vol 52, No 6

J., Liang, J., et al. (1992). Impact of self-management education on functional health status of older adults with heart disease. Gerontologist, 32, 438-443. Council of Scientific Affairs, Hispanic Health in the United States. (1991). Journal of the American Medical Association, 265(2), 248-252. De Palo, V. A., Mayo, P. H., Friedman, P., & Rosen, M.U. (1994). Demographic influences on asthma hospital admission rates in New York City. Chest, 106, 447-451. Eberhardt, M., Ingram, D., Makuc, D., et al. (2001). Urban and rural health chartbook. Health United States 2001. Hyattsville, MD: National Center for Health Statistics. Furino, A., & Muñoz, E. (1991). Health status among Hispanics: Major themes and new priorities. Journal of the American Medical Association, 265(2), 255-257. Glasgow, R. E., LaChance, P. A., Toobert, D. J., Brown, J., Hampson, S. E., & Riddle, M. C. (1997). Long-term effects and costs of brief behavioural dietary intervention for patients with diabetes delivered from the medical office. Patient Education Counseling, 32, 175-184. González, V., Hernandez-Marin, M., Lorig, K., Holman, H., Sobel, D., Laurent, D., et al. (2002). Tomando control de su salud: Una guia para el manejo de las enfermedades del corazon, diabetes, asma, bronquitis, enfisema y otros problemas cronicos. Boulder, CO: Bull Publishing Company. González, V. M., Stewart, A., Ritter, P. L., & Lorig, K. (1995). Translation and validation of arthritis outcome measures into Spanish. Arthritis and Rheumatism, 38(10), 1429-1446. Harris, M. I., Klein, R., Cowie, C. C., Rowland, M., & ByrdHolt, D. D. (1998). Is the risk of diabetic retinopathy greater in non-Hispanic blacks and Mexican Americans than in nonHispanic whites with type 2 diabetes? A U.S. population study. Diabetes Care, 21, 1230-1235. Hayes-Bautista, D. E., Baezconde-Garbanati, L., Schink, W.O., & Hayes-Bautista, M. (1994). Latino health in California, 19851990: Implications for family practice. Family Medicine, 26(9), 556-562. Haywood, L. J. (1990). Hypertension in minority populations: Access to care. American Journal of Medicine, 88(3B), 17S-20S. Hunt, L. M., Valenzuela, M. A., & Pugh, J. A. (1998). Porque me toco a mi? Mexican American diabetes patients’ causal stories and their relationship to treatment behaviors. Social Science and Medicine, 46(8), 959-969. Lenker, S. L., Lorig, K., & Gallagher, D. (1984). Reasons for the lack of association between changes in health behavior and improved health status: An explanatory study. Patient Education and Counseling, 6(2), 69-72. Lorig, K., Chastain, R. L., Ung, E., Shoor, S., & Holman, H. R. (1989b). Development and evaluation of a scale to measure perceived self-efficacy in people with arthritis. Arthritis and Rheumatism, 32(1), 37-44. Lorig, K., & González, V. (1992). The integration of theory with practice: A 12-year case study. Health Education Quarterly, 19(3), 355-368. Lorig, K., González, V., & Laurent, D. (1999a). The chronic disease self-management leader’s manual (revised). Palo Alto, CA: Stanford Patient Education Research Center.

Hispanic Chronic Disease Self-Management 369 Lorig, K., González, V. M., & Ritter, P. (1999b). Communitybased Spanish language arthritis education program: A randomized trial. Medical Care, 37(9), 957-963. Lorig, K., Lubeck, D., Kraines, R. G., Seleznick, M., & Holman, H. R. (1985). Outcomes of self-help education for patients with arthritis. Arthritis and Rheumatism, 28(6), 680-685. Lorig, K., Seleznick, M., Lubeck, D., Ung, E., Chastain, R. L., & Holman, H. R. (1989a). The beneficial outcomes of the arthritis self-management course are not adequately explained by behavior change. Arthritis and Rheumatism, 32(1), 91-95. Lorig, K., Stewart, A., Ritter, P., González, V., Laurent, D., & Lynch, J. (1996). Outcome measures for health education and other healthcare interventions. Thousand Oaks, CA: Sage. Lorig, K. R., Mazonson, P. D., & Holman, H. R. (1993). Evidence suggesting that health education for self-management in patients with chronic arthritis has sustained health benefits while reducing healthcare costs. Arthritis and Rheumatism, 36(4), 439-446. Lorig, K. R., Ritter, P., Stewart, A. L., Sobel, D. S., Brown, B. W., Jr., Bandura, A., et al. (2001). Chronic disease self-management program: 2-year health status and healthcare utilization outcomes. Medical Care, 39(11), 1217-1223. Lorig, K. R., Sobel, D., Stewart, A. L., Brown, B. W., Jr., Bandura, A., Ritter, P., et al. (1999c). Evidence suggesting that a chronic disease self-management program can improve health status while reducing hospitalization: A randomized trial. Medical Care, 37(1), 5-14. Marin, G., VanOss-Marin, B., & Pérez-Stable, E. J. (1990). Feasibility of a telephone survey to study a minority community: Hispanics in San Francisco. American Journal of Public Health, 80(3), 323-326. Marin, G., & VanOss-Marin, B. V. (1991). Research with Hispanic populations. Newbury Park, CA: Sage. Marin, M., Gonzalez, V., & Lorig, K. (2000). Curso tomando control de su salud: Manual de los instructores. Palo Alto, CA: Stanford Patient Education Research Center. Nacif de Brey, V. (1995a). ¡Hagamos ejercicio! [let’s exercise!] (audiocassette). Recorded by M. L. S.-B. Palo Alto, CA: Stanford Patient Education Research Center. Nacif de Brey, V. (1995b). Relajacion muscular progresiva (audiocassette). Recorded by M. D. Stork. Palo Alto, CA: Stanford Patient Education Research Center. National Diabetes Information Clearinghouse. (2002). National diabetes statistics. Retrieved April 4, 2002, from www.niddk. nih.gov/health/diabetes/pubs/dmstats/dmstats.htm Ritter, P. L., Stewart, A. L., Kaymaz, H., Sobel, D. S., Bloch, D. A., & Lorig, K. R. (2001). Self-reports of healthcare utilization compared to provider records. Journal of Clinical Epidemiology, 54(2), 136-141. Schulman, R. & B., Inc., & Glaxo Wellcome Inc. (1998). Asthma in America. Retrived April 4, 2003 from www.AsthmaInAmerica.com Therrien, M., & Ramirez, R. (2000). The Hspanic population in the United States (No. P20-535). Washington, D.C.: U.S. Department of Commerce. Wu, S., & Green, A. (2000). Projection of chronic illness prevalence and cost inflation. Santa Monica: Rand Corporation.