An Evaluation of Choose to Move 1999: An American Heart Association Physical Activity Program for Women

Background  Rates of physical inactivity and poor nutrition, which are 2 of the most important modifiable risk factors for cardiovascular disease in women, are substantial. Even so, studies of interventions designed to improve lifestyle behaviors in women have been limited and often confined to particular geographical areas.

Objective  To evaluate the effect of Choose to Move on increasing women's physical activity, improving their knowledge of heart disease and stroke, and improving their nutrition.

Participants and Methods  A prospective, nonrandomized, 12-week educational intervention designed by the American Heart Association for women across the United States. Participants received a welcome kit and manual with weekly information about how to manage cardiovascular disease risk factors and how to build a support system for lifestyle change. Women (N = 23 171) aged 25 years or older were recruited by direct mail, the media, health care providers, and other means. Follow-up evaluations were returned from 6389 women at 2 weeks, 5338 at 4 weeks, 4209 at 8 weeks, 3916 at 10 weeks, and 3775 at 12 weeks. Participants self-reported their physical activity, diet, and knowledge about heart disease, stroke, and related symptoms.

Results  Ninety percent of the participants were white and 56% were aged between 35 and 54 years. Among the participants who completed the week 12 follow-up evaluation, the percentage who reported being active (at least moderate exercise ≥5 times per week or >2½ hours per week for the past 1 to 6 months) increased from 32% at baseline to 67% at the program's end (P = .001). Participants currently limiting excess calories or fat increased from 72% to 91% at week 10 follow-up evaluation (P = .001). The proportion correctly identifying heart disease as the leading cause of death increased from 84% to 91% at week 10 follow-up evaluation (P<.001).

Conclusions  Women who completed the Choose to Move program evaluation reported that they significantly increased their levels of physical activity, reduced their consumption of high-fat foods, and increased their knowledge and awareness of cardiovascular disease risk and its symptoms. This program provides an important model for public health, voluntary, and other health organizations of population-based, targeted low-cost self-help programs that support the Healthy People 2010 objectives for physical activity, nutrition, and cardiovascular health.

IN THE UNITED STATES, heart disease remains the leading cause of death in women.¹ Physical inactivity and poor nutrition are major risk factors for heart disease and stroke in women and, thus, important to modify. Adherence to healthy lifestyles involving physical activity, diet, and smoking abstinence is associated with a very low risk of coronary heart disease in women.² In addition, physical activity, even moderate-intensity exercise such as walking, is associated with a lower risk of stroke in women.^3-5 Furthermore, research has demonstrated that diets low in saturated fat and high in fruits, vegetables, whole grains, and fiber are associated with a reduced risk of coronary heart disease.⁶

The Centers for Disease Control and Prevention,^7,8 the National Institutes of Health,⁹ and the American Heart Association (AHA)¹⁰ recommend that all Americans accumulate 30 minutes or more of moderate-intensity physical activity on most, preferably all, days of the week. However, a national study based on self-report found that 31% of American women had no leisure-time physical activity in the previous month¹¹; another national study found that only 3% participated in vigorous leisure-time physical activity 3 or more times per week.¹² As women age, activity levels decline further, and those with less than a high school education or lower income are the most physically inactive.⁸ Furthermore, only 25% of women aged 20 years or older are meeting the Healthy People 2010 recommendation for limiting dietary fat intake to 30% of daily calories or less.^13,14

To address the epidemic of physical inactivity, obesity, and poor dietary practices among American women, the AHA as part of its National Women's Heart Disease and Stroke Campaign, launched the Choose to Move program, a 12-week, self-help lifestyle intervention program. Previous work has shown that behavioral change programs with minimal contact that involve self-monitoring help women adhere to exercise programs.^15-19 The purpose of this study was to evaluate the effect of Choose to Move on improving their knowledge of heart disease and stroke, increasing women's physical activity, and improving their nutrition.

Piloted in 1998, the 1999 Choose to Move program, which was free, targeted women aged 25 years or older who wanted to become more physically active. Current innovative intervention programs focus on reaching large segments of the population using one of the following mediated approaches: mail and/or print, telephone, Internet, or a combination of these. Choose to Move is a 12-week, mail-mediated lifestyle intervention program designed by the AHA in collaboration with the Cooper Institute for Aerobics Research, Dallas, Tex, and the Centers for Disease Control and Prevention, Atlanta, Ga. Using a prospective, nonrandomized study design, we evaluated participants' knowledge of heart disease and stroke and their behaviors related to physical activity and diet.

Participants were recruited through local AHA offices, physicians, nurse practitioners, physical therapists, occupational therapists, exercise instructors, direct mail, national media, and other sources. Women could call the campaign's toll-free number or e-mail the AHA's Web site (http://www.women.americanheart.org) to request a registration packet.

Baseline information on current physical activity level and nutritional behaviors, knowledge of heart disease and stroke, and the presence of risk factors for cardiovascular disease (CVD) was collected at registration. The baseline survey also included the AHA's Health Risk Appraisal for CVD and stroke, available on their Web site, which helped women learn about their risk factors for these diseases.

Choose to Move incorporated principles of the "Transtheoretical Stages of Change Model," which postulates that people change their behavior in stages and that tailoring interventions to match a person's readiness to change is essential for success.^20,21 This model, which has been widely applied to health behaviors,²² has been used successfully to promote physical activity in the community,^23,24 the work site,²⁵ and clinic settings.^26,27 Choose to Move was developed to help women in the "contemplative" and "preparatory" stages of physical activity move to the next stage, but the program was open to any woman who registered for it (Table 1).

The program was designed to permit women to easily incorporate physical activity and healthy eating changes into their lifestyle. The AHA tried to adhere to the social marketing tenet that health promotion programs should reach a specific audience, satisfy consumer needs, and meet organizational objectives. Following these precepts was considered important for increasing women's acceptance of physical activity and low-fat eating as daily practice.^28,29

Registered participants were sent a welcome kit that included program materials and a bookmark; the program handbook included 12 weekly behavioral modification topics and information tailored to women (Table 2). The program was designed to teach women how to incorporate a daily routine of physical activity into their lives in creative and practical ways. Participants were asked to begin with 10 minutes per day of moderate-intensity physical activity. In week 5, the program began to address eating behaviors as well. By the eighth week, women were encouraged to do 30 minutes of physical activity daily, a goal consistent with the 1996 surgeon general's physical activity and health report⁸ and the AHA's guidelines for physical activity.³⁰ The program provided information to help participants increase their physical activity and manage their cholesterol level and weight by building a strong support system of friends and family. Special messages were also included to help women at high risk for heart disease and stroke.

The AHA also sent postcards, e-mails, and a newsletter to participants encouraging them to continue the program. In addition, women could log on to the AHA's Web site to get other health information. Women who completed the program and sent in a final summary card received a Choose to Move T-shirt and certificate. The postprogram newsletter was designed to reinforce program tactics for healthy living and to assist participants in maintaining the behavioral stage of change they had achieved.

The primary objective of the Choose to Move program was to increase the proportion of women who met national recommendations for physical activity and consumption of high-fat and cholesterol-laden foods. Secondary objectives were to increase the participants' knowledge about the risk of CVD and its symptoms and to increase use of the strategies to improve physical activity and diet. During the program, participants were asked to complete and send follow-up evaluation summary cards to the AHA every 2 weeks to measure their progress and these objectives.

To evaluate their knowledge about CVD risk and symptoms, participants were asked to identify the current leading cause of death for women and to check whether particular symptoms were warning signals of an oncoming heart attack or stroke. To evaluate physical activity behavior at both registration and week 12, participants completed the stages of change question, which included 8 graded statements about physical activity. We ultimately classified all participants in 1 of 2 groups: The first group consisted of levels 1 through 4 (ie, precontemplation, contemplation, and preparation) and was termed "inactive"; the second group consisted of levels 5 through 8 (ie, action and maintenance) and was termed "active." All those in the active group were considered to perform, at minimum, the equivalent of moderate exercise 5 or more times per week or more than 2½ hours per week for the past 1 to 6 months. Participants were also asked, "To lower your risk of developing heart disease or stroke, are you exercising more?"³¹ and to check the strategies they were using to increase their moderate-intensity physical activity.

To measure dietary behavior, participants were asked, "Are you currently limiting excess calories or fat in your diet?" and "To lower your risk of developing heart disease or stroke, are you eating fewer high-fat or high-cholesterol foods?" Both questions are also used in the Centers for Disease Control and Prevention's Behavioral Risk Factor Surveillance Survey,³¹ allowing comparison with a national sample of women. Participants were also asked to check the strategies they were using to reduce excess calories, fat, and dietary cholesterol.

To examine potential response bias, women who completed the final week 12 evaluation ("evaluation cohort") were compared based on demographic and other baseline characteristics with those who completed the registration only ("registration cohort"). Comparisons were based on knowledge of CVD, self-confidence for increasing one's physical activity level, current physical activity level, and the use of strategies to increase one's physical activity level and to improve one's nutrition. Differences in proportions between the 2 groups were examined using the χ² test for independence; continuous outcomes were examined using a t test for independent samples.

Program objectives were evaluated based on responses from women who completed the baseline and final follow-up evaluation (evaluation cohort). Comparisons were made based on program outcomes using the longitudinal cohort of respondents at registration with matched responses at various stages of follow-up. Follow-up data for each objective were collected on summary feedback cards at different points in time; therefore, each cohort is composed of a slightly different group of participants. Specific comparisons included the proportions of women who (1) met the AHA guidelines for physical activity, (2) used Choose to Move strategies to increase their moderate-intensity physical activity, (3) currently limited excess calories or fat in their diet, (4) ate fewer high-fat or high-cholesterol foods to lower their risk of developing heart disease or stroke, and (5) used Choose to Move strategies to reduce excess calories, fat, and cholesterol in their diet. Differences in proportions were examined using the McNemar test for matched pairs. A paired t test was used to examine changes over time in the number of signals of an oncoming heart attack or stroke correctly identified as well as the number of Choose to Move physical activity and nutrition strategies used. Analysis of each outcome excluded women with missing data for the evaluation item(s) corresponding to that outcome. The SAS Version 6.12³² was used for statistical analysis. All values are given as means±SDs.

Of the 23 171 women who registered for the program, 6389 responded at the end of week 2; 5338, week 4; 4209, week 8; 3916, week 10; and 3775, week 12. The study population was primarily white women with more than half aged between 35 and 54 years (Table 3). Almost half had 2 to 3 risk factors for heart disease or stroke. The registration cohort and the evaluation cohort were similar for these characteristics.

The registration cohort differed significantly from the evaluation cohort for numerous variables related to program objectives. The evaluation cohort were more knowledgeable that heart disease is the leading cause of death for women (84% vs 80%, P = .001). In addition, they reported more physical activity (Table 1) on an 8-point scale (4.3 ± 1.6 SD vs 3.8 ± 1.5, P<.001). They also were more confident about increasing their physical activity level (on a scale of 1 [not at all confident]-5 [extemely confident], mean confidence score, 3.8 ± 0.93 vs 3.6 ± 0.98, P<.001). They reported using more strategies (possible, 12) to improve their physical activity (mean score, 3.5 ± 2.2 vs 3.1 ± 2.2, P<.001), and they reported using more strategies (possible, 15) to improve their nutrition (mean score, 6.3 ± 3.1 vs 6.0 ± 3.2, P<.001).

Evaluation of each program objective excluded women with missing data for the corresponding question(s). Among the 3313 women who responded at both registration and week 9 to 10, more correctly identified heart disease as the leading cause of death at follow-up than at baseline (91.3% vs 83.8%; P = .001). Among the 5160 who responded at both registration and week 3 to 4 follow-up, the mean number of warning signals of an oncoming heart attack or stroke correctly identified was significantly greater at follow-up than at baseline (9.4 ± 1.6 vs 8.3 ± 1.9; P<.001).

There was an increase in the use of each of the 12 recommended strategies to increase physical activity (Table 4). Among the 3835 women who responded to both registration and week 9 to 10 evaluation, the mean number of strategies almost doubled (6.8 ± 2.6 vs 3.5 ± 2.2; P<.001). More important, among the 3254 who responded to both registration and week 11 to 12 evaluations, a significantly greater proportion of women scored active (level ≥5) at the program completion than at the beginning (67% vs 32%; P = .001) (Figure 1).

At week 7 to 8 follow-up, 91% of the women (n = 3369) reported that they were limiting excess calories or fat in their diet, compared with 72% of that group at baseline (P = .001). Of 2423 women who reported at registration that they were limiting excess calories or fat, 96% continued to report this behavior at week 7 to 8. Of the 946 women who were not limiting fat or calories at baseline, 78% were doing so at week 7 to 8. At week 11 to 12, women (n = 3627) were more likely to report that they were eating fewer high-fat or high-cholesterol foods to lower their risk of heart disease or stroke than they were at registration (94% vs 81%; P = .001). Of 2935 women who reported at registration that they were eating fewer high-fat or high-cholesterol foods, 97% continued to report this behavior at week 11 to 12. In addition, 82% of the 692 women who at baseline did not report this behavior had changed this behavior by their week 11 to 12 evaluation.

For each of the 15 dietary strategies, a higher percentage of the 4159 participants who responded at both registration and week 7 to 8 evaluation indicated they were using each strategy at follow-up (Table 5). For this group, the mean number of strategies being used increased from 6.4 ± 3.1 to 9.5 ± 2.8 vs; P<.001).

The main finding of this study is that Choose to Move, a mail-mediated lifestyle intervention program, significantly improved self-reported physical activity levels, heart healthy food choices, and increased knowledge about heart disease among a cohort of women. Overall results show that a social marketing approach—promoting a targeted, self-help lifestyle intervention program designed to satisfy women's needs and reduce risk of heart disease and stroke—can reach a large number of women and help them to positively change their behavior within 12 weeks.

These results compare favorably with other self-monitoring interventions^15-19 as well as with results from interventions that are tailored to women's needs, experiences, and readiness to change.^25,33 The results are also consistent with more intensive intervention programs designed to modify physical activity and nutritional behavior.^{16,18,27,34-46} In addition, the outcomes support the current recommendation for programs to focus more on developing behavioral skills that integrate moderate-intensity physical activity, such as walking, into one's daily life.^47-52 For women, the home-based Choose to Move program may have advantages over more structured formats, such as group sessions, that require greater time, as well as transportation, staff involvement, and considerable costs.

At baseline, the registrants demonstrated a high level of knowledge about heart disease, stroke, and their symptoms, a finding that differs sharply from that of a recent national study of women conducted by the AHA.⁵³ In that study, only 33% identified heart disease as the leading cause of death, and women aged 25 to 44 years indicated they were not well informed about heart disease and stroke. The success of the Choose to Move program may be attributed in part to the knowledge levels of the participant. Perhaps women who are more aware of CVD risk participate in programs and change their health behavior. Because 92% of the women who completed the program were white, future efforts should explore opportunities to reach a more heterogeneous population of women, including women of color.

This program evaluation has several limitations. First, owing to the nature of the program, no control group was used in assessing the efficacy of the program. Self-selection may have biased our results by overestimating the benefits of the program. Although the program registrants were similar in their exercise and dietary behaviors compared with a national sample of American women,³¹ the results may not generalize to less-motivated women. However, the purpose of the program was to provide skills and support to women who were ready to change their health behavior.

Second, typical of correspondence or mail-mediated programs with limited direct contact, a large number of women registered for the program, but a much smaller number completed the follow-up evaluations.^54-56 Because women who completed the final follow-up evaluation were slightly more health conscious and confident than those who did not, the results may reflect some response bias. Alternatively, one would expect the evaluation cohort to be more successful at changing their behavior and, thus, more likely to return their evaluations. Future efforts include plans to follow-up with a sample of dropouts, one method of gauging cohort bias. The AHA is working with teams of behavioral research scientists who are participating in the Innovative Approaches to Behavior Change Consortium (cosponsored by the National Institutes of Health and the Robert Wood Johnson Foundation) to develop tools to enhance recruitment and retention efforts in programs like Choose to Move.

For feasibility, our study used self-reported data, a limitation for validating results. A prospective study by Blair et al,⁵⁷ however, found that self-reported physical activity was the predominant predictive factor of cardiorespiratory fitness among adults in all the age and sex subgroups they analyzed. In addition, 3 of the questions that measured physical activity and fat consumption have been field tested and are used in the national Behavioral Risk Factor Surveillance Survey.³¹ Furthermore, the stages of change question for physical activity level has been used widely in other studies²² and has been found to be an acceptably reliable and valid measure of physical activity.^58,59

With the shift in emphasis toward population-level interventions to change health risk behaviors,^60-63 mail-mediated intervention programs such as Choose to Move have great appeal. This program was provided at a low cost, enrolled large numbers of women, and had a high rate of success for those who completed it. The results will provide insight into what strategies help to make health promotion programs for women effective. Participant feedback is being used to better tailor messages and to improve program content in Choose to Move 2000. Future efforts may include different methods to reach those at greater risk for CVD and a longer follow-up period with more personalized feedback, something that participants requested. We know that regular physical activity and improvements in diet need to be continued on a long-term basis for health benefits to be attained; the challenge is to increase the total number and diversity of women who complete the program and make healthy lifestyle changes. Widespread diffusion of programs like Choose to Move may help to further reduce the CVD burden in US women.

Choose to Move is one of the largest health promotion interventions reported to date that has examined changes in physical activity and dietary behaviors. It is the only program conducted by a voluntary organization that has used a social marketing approach to design a physical activity and nutrition program for women. Although the results for those who completed the program are promising, the limited completion rate underscores the need to continue to develop long-term health behavior maintenance programs for diverse populations.

Accepted for publication March 13, 2001.

We thank Penelope Logan, BS, Katrina McGhee, MBA, and Susan Earhart-Brower, BA, and the staff at the Cooper Institute for Aerobics Research for their superb coordination and contribution to Choose to Move.

Corresponding author and reprints: Dyann Matson Koffman, DrPH, MPH, CHES, Centers for Disease Control and Prevention, 4770 Buford Hwy NE, Mailstop K-47, Atlanta, GA 30341-3717