Self-efficacy and mood in cardiac rehabilitation: should gender be considered? – Statistical Data Included
Chris M. Blanchard
Coronary heart disease (CHD) remains the leading cause of death for men and women in the United States. Furthermore, 1.1 million new or recurrent coronary events were expected in the United States in the year 2000. (1) On the basis of a national survey of phase II cardiac rehabilitation (CR) programs in the United States, 14.8% of the 1.1 million people hospitalized as a result of CHD were expected to participate in structured phase II CR. Of this 14.8%, 75% were expected to be men and 25% to be women. (2) Moreover, once men and women enroll in a phase II CR program, women have consistently had lower adherence rates and have been more likely than men to drop out of the program. (3-6)
As a result of women’s low enrollment and adherence rates and high dropout rates from phase II CR, research examining gender differences in various psychosocial outcomes has been rare. Identifying any potential psychosocial differences (eg, anxiety, depression, self-efficacy) between men and women participants in these programs will provide invaluable information on which to base future program interventions to (a) increase psychological benefits and (b) offer insights into gender differences in exercise adherence during and following phase II CR.
Self-Efficacy and Phase II Cardiac Rehabilitation
To date, the majority of research examining the influence of phase II CR on self-efficacy has focused on what Maddux (7) has described as task efficacy (ie, confidence in one’s ability to perform the elemental aspects of a task such as walking for 30 minutes). Authors of 2 studies have examined task efficacy as a predictor of exercise adherence in phase II CR. One found a significant relationship between task efficacy and exercise adherence, (8) whereas the other found no relationship. (9) However, when examining task efficacy as an outcome variable, authors of 3 studies have shown that task efficacy significantly increased from pre- to postphase II CR. (9-11) Schuster and Waldron (4) found that men’s task efficacy upon entry into a phase II CR program was significantly higher than women’s, although they did not explore the change in task efficacy over time. However, some evidence suggests that men and women have similar increases in task efficacy from prebypass surgery to 1 week postsurgery that persists up to 12 months after surgery. (12,13) Nonetheless, men’s task efficacy was significantly higher at all time points.
Unfortunately, we know of no information regarding the change in task efficacy specifically associated with a structured phase II CR program that directly compares men and women. In a sample from the general population, McAuley, Courneya, and Lettunich (14) found that women had a larger increase than men had in task efficacy during a 20-week structured exercise program. This finding suggested that it is important to consider potential gender differences in a phase II CR context. In addition, it is important to examine the relationship between exercise adherence and the change in task efficacy in men and women separately because it is not clear whether increases in task efficacy reported in the literature (9-11) are actually related to exercise adherence during phase II CR or to continued exercise following the completion of a phase II CR program.
Barrier efficacy (ie, confidence that one can perform a task under challenging conditions) has also been considered an important outcome of phase II CR. (15) The notion is that individuals who are more effective in overcoming context-relevant barriers (eg, exercise barriers) will be more likely to persist in the face of those barriers than individuals who are less effective in doing so. (16) Researchers have consistently shown that barrier efficacy is a significant predictor of exercise adherence in cross-sectional (17,18) and prospective (19,20) designs in the general exercise population. Furthermore, Blanchard et al (15) found that patients with higher barrier efficacy had significantly higher exercise adherence during phase II CR than patients with lower barrier efficacy.
In the only study that examined barrier efficacy in a phase II CR context, Bock et al (21) found that barrier efficacy significantly increased from pre- to postphase II CR. However, those researchers did not examine gender differences. Therefore, although Blanchard et al (15) found that men had significantly higher barrier efficacy than women (cross-sectional) during phase II CR, it is not known whether phase II CR differentially influences barrier efficacy in men and women. Furthermore, it is not known whether efficacy for barriers associated with phase II CR has a similar influence on men’s and women’s exercise adherence during and after phase II CR.
Why Does Cardiac Rehabilitation Improve Mood?
Accumulating evidence suggests that phase II CR significantly reduces anxiety and depression (22-25) and increases positive mood (26,27) (eg, vigor) in men and women. However, research that attempts to explain these changes in mood is limited. Recently, some researchers have begun to use a social cognitive approach to explain mood changes in cardiac populations. (28,29) Social cognitive theory suggests that there is a bidirectional relationship between self-efficacy and mood such that a change in one might lead to a change in the other. (16)
To date, we know of no reports of research that directly examined this relationship in a phase II CR program. However, there is evidence to suggest that task efficacy is related to mood disturbance 2 weeks after a coronary angioplasty (28) and to vigor at 4, 12, and 24 weeks after bypass surgery. (29) Although this research provided initial insight on the task efficacy/mood relationship in a cardiac population, it did not demonstrate that a change in task efficacy was associated with a change in mood (ie, the researchers did not test the bidirectional relationship). Stewart et al, (30) however, showed that a significant increase in task efficacy was associated with a significant decrease in anxiety and depression in mildly hypertensive men after a 10-week training program. Therefore, it appears that an increase in task efficacy experienced during an exercise program is associated with a decrease in anxiety and depression. Nonetheless, this relationship has not been demonstrated in a phase II CR context.
With respect to barrier efficacy, there is also no information about the bidirectional relationship between barrier efficacy and mood in phase II CR. Although McAuley (31) has shown that barrier efficacy was significantly related to positive mood midway through an exercise program in sedentary middle-aged participants, there has yet to be a report of research that directly examines the bidirectional relationship between barrier efficacy and mood proposed by social cognitive theory. (16) If a bidirectional relationship between barrier efficacy and mood exists during phase II CR, this will underscore the importance of designing exercise interventions to increase barrier efficacy. In doing so, a phase II CR program that effectively increases barrier efficacy may increase exercise adherence during phase II CR (15) and may also improve the psychological well-being of the patients.
Purpose of the Present Study
Our first purpose in conducting this study was to examine the time-course changes of task and barrier efficacy, anxiety, depression, and vigor during and following phase II CR in men and women. We hypothesized that barrier efficacy, task efficacy, and vigor would significantly increase and that anxiety and depression would significantly decrease from pre- to postphase II CR. We further hypothesized that these postrehabilitation levels would be maintained at follow-up (ie, 6 to 10 weeks postrehabilitation). Finally, on the basis of previous research, (14) we hypothesized that the changes in task and barrier efficacy would be larger in women than in men, although we expected no gender differences for any of the mood changes. (22,27)
The second purpose of our study was to examine the relationships between the changes in task and barrier efficacy during phase II CR and exercise adherence in men and women. We hypothesized that both men and women with higher exercise adherence during phase II CR would experience larger changes in task and barrier efficacy. The third purpose of our study was to examine the relationships between task and barrier efficacy at the end of phase II CR and exercise adherence following completion of the program (ie, from postrehabilitation to follow-up). We hypothesized that task and barrier efficacy at the end of phase II CR would be significantly related to exercise adherence postrehabilitation in men and women. We also hypothesized that tenets of social cognitive theory increases in task and barrier efficacy during and after phase II CR would be significantly related to decreases in anxiety and depression and an increase in vigor in both men and women.
Phase II Cardiac Rehabilitation at the Glenrose Program
Phase II CR follows phase I, which is the hospitalization initially associated with a cardiac event. Phase II usually starts within 2 to 4 weeks after hospital discharge. (32) The Glenrose phase II program combines medically supervised exercise with nutrition and behavior-modification education to reduce risk factors associated with CHD. The program lasts from 4 to 8 weeks, depending on the severity of the patient’s condition. Exercise frequencies vary from 1 to 3 times per week and last approximately 1 hour. The program’s duration and exercise frequency are determined by each patient’s cardiac condition. Both can be altered, depending on the patient’s response to the rehabilitation. The physician, other medical staff, and the exercise specialists decide on details of the exercise program. The Glenrose program also includes weekly education classes on topics such as cardiac risk factors, sexual activity postcoronary event, stress management, and nutrition.
Participants Versus Nonparticipants
All patients entering the Glenrose rehabilitation program were eligible to participate in our study. Over the 9-month recruitment period, we sent an initial questionnaire package to 253 patients; 78 men and 29 women returned their questionnaires at their initial orientation meeting at the Glenrose. This yielded a response rate of 43%. The most common reasons for refusing to participate were lack of interest and inability to speak English.
To evaluate the representativeness of our sample, we compared the 146 individuals who were approached but did not participate in our study with the 107 initial participants on demographic (eg, marital status, education level, and employment status) and medical variables (eg, cholesterol levels and type of cardiac event). We found no differences on any of these variables.
Participants Versus Dropouts
During the study, 26 of the 107 participants (21 men and 5 women) dropped out, yielding a completion rate of 76%. Five of these participants never returned a questionnaire and remained in rehabilitation, 8 were discharged early, 4 did not attend any exercise sessions, 2 of the patients’ programs were put on hold for health reasons, and 7 patients missed their postrehabilitation follow-up stress test.
To further evaluate how representative our findings were, we compared the final participants with the dropouts on the same demographic and medical variables as those listed above and found no group differences. Finally, because we obtained prerehabilitation data on all of our 107 patients, we further compared the adherers and dropouts on the theoretical variables in question before the start of their phase II CR. Our findings showed that adherers and dropouts were similar in levels of task efficacy (duration and frequency), barrier efficacy, vigor, anxiety, and depression.
The final sample consisted of 57 men and 24 women who completed all 4 questionnaires. The men and women were similar in age t(79) = -1.10, p > .05 (men, age M = 58.68 y; women, age M = 61.75 y) and body mass index (BMI) t(79) = -.24, p > .05 (BMI men, M = 28.82; BMI women, M = 29.14). However, women had a larger percentage of body fat than men (M = 32.55% vs M = 24.74%), t(78) = -8.11, p < .01. Women also had more thyroid problems and were more likely than men to be homemakers (see Table 1).
The barrier efficacy scale consisted of 9 exercise barriers specific to phase II CR. (15) They were (a) fear of having a cardiac incident, (b) back pain, (c) medication side effects, (d) angina/chest pain, (e) health-related problems, (f) bad weather, (g) too much work, (h) lack of time, and (i) too expensive. The statement “How confident are you that you can exercise at some point during the day when … ?” preceded each exercise barrier. Patients rated their confidence on a scale from not at all confident (1) to very confident (10). We calculated the average of the 9 exercise barriers and used it as an indicator of overall barrier efficacy at all 4 time points. The scale showed acceptable internal consistency at (a) mail-out ([alpha] = .86), (b) prerehabilitation ([alpha] = .87), postrehabilitation ([alpha] = .64), and follow-up ([alpha] = .79).
We assessed 2 aspects of task efficacy. The first measured confidence in exercising for various durations. We asked patients, “How confident are you that you can exercise at a moderate but comfortable intensity for …” (a) 10 minutes, (b) 20 minutes, (c) 30 minutes, (d) 40 minutes, and (e) 50 minutes. After 50 minutes, the patients rated their confidence on a scale ranging from not at all confident (1) to very confident (10). We then averaged the 5 ratings to obtain the task efficacy (duration) score. Internal consistencies were acceptable at (a) mail-out ([alpha] = .96), (b) prerehabilitation ([alpha] = .95), postrehabilitation ([alpha] = .82), and follow-up ([alpha] = .85).
The second task efficacy measure was confidence in exercising at various frequencies. We asked patients, “How confident are you that you can exercise for 30 continuous minutes at a moderate, but comfortable, intensity …” (a) 2 times per week, (b) 3 times per week, (c) 4 times per week, and (d) 5 times per week. Patients rated their confidence on a scale from not at all confident (1) to very confident (10). Again, we averaged the 4 ratings to obtain the task efficacy (frequency) score. Internal consistencies were acceptable at (a) mail-out ([alpha] = .97), (b) prerehabilitation ([alpha] = .97), postrehabilitation ([alpha] = .87), and follow-up ([alpha] = .85). The format used for these scales is similar to that used in previous studies. (9,11)
Anxiety, Depression, and Vigor
We assessed these 3 moods by using the abbreviated subscales of the Profile of Mood States developed by Grove and Prapavessis. (33) For anxiety, we used the following items: (a) restless, (b) nervous, (c) on edge, (d) tense, (e) uneasy, and (f) anxious; for depression, (a) hopeless, (b) helpless, (c) sad, (d) worthless, (e) miserable, and (f) uncertain; and for vigor (a) cheerful, (b) vigorous, (c) full of pep, (d) active, (e) energetic, and (f) lively. Each adjective was preceded by the statement, “Over the past week, I have felt …” Patients rated their moods on a scale from not at all (0) to extremely (4). Higher scores on anxiety and depression indicate greater mood disturbance, but the opposite is true for vigor. Internal consistencies were acceptable at (a) mail-out (anxiety [alpha] = .90, depression [alpha] = .92, vigor [alpha] = .91); (b) prerehabilitation (anxiety [alpha] = .90, depression [alpha] = .92, vigor [alpha] = .87); (c) postrehabilitation (anxiety [alpha] = .90, depression [alpha] = .95, vigor [alpha] = .83), and follow-up (anxiety [alpha] = .88, depression [alpha] = .94, vigor [alpha] = .84).
Exercise Adherence During Rehabilitation
To obtain an objective measure of the patients’ exercise adherence during their phase II CR, we used the following formula: number of exercise sessions attended divided by number of exercise sessions prescribed at the start of rehabilitation multiplied by 100. The percentage approach was necessary because the patients’ exercise frequencies varied from 1 to 3 times per week, depending on the severity of their condition. A percentage score was standardized across patients, and the Glenrose staff verified patients’ objective exercise attendance by medical information recorded during each exercise session. If a patient did not appear for his or her scheduled exercise session, he or she was marked absent for that day in the medical file.
We used the leisure score index (LSI) of the Godin Leisure-Time Exercise Questionnaire (34) to assess self-reported exercise. The LSI contains 3 questions about the frequency of mild, moderate, and strenuous exercise performed for at least 15 minutes during free time in a typical week. We calculated a total LSI score by adding the frequency of exercise in each of the categories. An independent evaluation of this measure found that its reliability and validity compared favorably with 9 other self-report measures of exercise related to various criteria, including test-retest scores, objective activity monitors, and fitness indices. (35) In the present study, we used the LSI to measure frequency of exercise from postrehabilitation to follow-up (ie, 6-10 weeks postrehabilitation).
Before beginning the study, we received approval from the Health Research Ethics Board of the University of Alberta. Once participants were referred to the Glenrose rehabilitation program, we sent them a questionnaire along with the information package that the Glenrose typically sent to patients. The questionnaire package included a letter from the Glenrose to support the research, a letter explaining the details of the study, and 2 informed-consent forms (1 for the patient and 1 for the researcher), as well as the questionnaire that included the LSI, task efficacy, barrier efficacy, and the 3 mood scales. The letter asked the patients who agreed to participate to complete the questionnaire immediately and to get in touch with the researchers if they had any questions or concerns. Patients were asked to return the completed questionnaire and informed-consent document at their initial orientation meeting at the Glenrose, which was 3 to 5 weeks later.
The researcher met patients who returned their completed questionnaires at the Glenrose orientation and gave them the same questionnaire to be completed and returned to their first scheduled exercise session, which was from 2 to 10 days later. The patients received their third questionnaire at their next-to-last exercise session and were asked to return it at their final session. Finally, patients completed their fourth questionnaire at their postrehabilitation follow-up stress test, which took place from 6 to 10 weeks after they completed their phase II CR program.
Once patients completed the final questionnaire, they were debriefed, a process that consisted of a conversation about the hypotheses of the study and answers to any questions they had about the study or their responses to the questionnaires. The researcher told them that they could get the results of the study any time.
Purpose 1: Self-Efficacy and Mood Over Time
To examine potential gender differences in the changes in task and barrier efficacy, and in anxiety, depression, and vigor throughout phase II CR, we conducted 3 separate analyses. All analyses used 2 (Gender) x 2 (Time) mixed-model analyses of variance (ANOVAs). The first analysis used mail-out to prerehabilitation as the time period to rule out history and maturation as potential threats to the internal validity of the study; we treated it as a control condition.
For example, if barrier efficacy significantly increased during phase II CR and was unchanged in the control condition, we could make a stronger case that phase II CR was responsible for this change.
The second analysis used prerehabilitation to postrehabilitation as the time period and examined the changes in self-efficacy and mood during phase II CR. For the third analysis, we used the time from postrehabilitation to follow-up and examined the changes in self-efficacy and mood after the phase II CR at the follow-up stress test. We followed up all interactions with 1-way repeated measure ANOVAs. See Table 2 for means and standard deviations for task and barrier efficacy, anxiety, depression, and vigor in men and women for each time.
From mail-out to prerehabilitation, we found no Gender x Time interaction for barrier efficacy, F(1, 78) = 0.51, p > .05; task efficacy (frequency), F(1, 77) = 0.01, p > .05; task efficacy (duration), F(1, 77) = 0.17, p > .05; vigor, F(1, 76) = 3.4, p > .05; anxiety F(1, 76) = 0.21, p > .05; or depression F(1, 76) = 0.18, p > .05. We also found no main effects for time for any of the variables. Therefore, it appears that the control condition may be effective in ruling out history and maturation as threats to the internal validity of our study.
Pre- to Postphase II CR
From pre- to postphase II CR, we found a significant Gender x Time interaction for barrier efficacy, F(1, 78) = 4.40, p < .05 ([[eta].sup.2] = .05); task efficacy (frequency), F(1, 78) = 6.58, p < .02 ([[eta].sup.2] = .08); and task efficacy (duration), F(1, 78) = 5.03, p < .05 ([[eta].sup.2] = .06). Additional 1-way repeated measure ANOVAs showed that men had significant increases in barrier efficacy, F(1, 56) = 75.96, p < .01 ([[eta].sup.2] = .52); task efficacy (frequency), F(1, 56) = 64.81, p < .05 ([[eta].sup.2] = .53); and task efficacy (duration), F(1, 56) = 94.28, p < .05 ([[eta].sup.2] = .62). However, the magnitude of change in women was larger for barrier efficacy, F(1, 22) = 53.85, p < .01 ([[eta].sup.2] = .71); task efficacy (frequency), F(1, 22) = 64.16, p < .05 ([[eta].sup.2] = .75); and task efficacy (duration), F(1, 22) = 58.31, p < .05 (112 = .73). Figures 1 to 3 show the time course changes of the task (frequency and duration) and barrier efficacy throughout the CR process.
We found no Gender x Time interactions for any of the mood variables. However, main effects of time showed that vigor significantly increased, F(1, 77) = 92.12, p < .01 ([[eta].sup.2] = .55), whereas anxiety significantly decreased, F(1, 77) = 65.12, p .05.
Postrehabilitation to Follow-up
We found no Gender x Time interactions for barrier or task efficacy (frequency and duration), anxiety, depression, or vigor. However, there were significant main effects for time for barrier efficacy, F(1, 78) = 38.29, p < .01 ([[eta].sup.2] = .33); task efficacy (frequency), F(1, 78) = 28.69, p < .01 ([[eta].sup.2] = .27); task efficacy (duration), F(1, 78) = 25.34, p < .01 ([[eta].sup.2] = .25); and vigor, F(1, 77) = 17.11, p .05, or depression, F(1, 76) = 0.03, p > .05. It is important to note, however, that the self-efficacy levels reported at follow-up were significantly higher than the levels reported at prerehabilitation for barrier efficacy, F(1, 79) = 46.30, p < .01 ([[eta].sup.2] = .37); task efficacy (duration), F(1, 79) = 95.37, p < .01 ([[eta].sup.2] = .55); and task efficacy (frequency), F(1, 79) = 61.33, p < .01 ([[eta].sup.2] = .44). This was also the case for vigor, F(1, 76) = 43.07, p < .05 ([[eta].sup.2] = .36).
Purpose 2: Exercise Adherence During Phase II CR and Changes in Self-Efficacy
To examine the influence of exercise adherence on the changes in task and barrier efficacy, we used a change-score approach. Although pre- to postchange scores are acceptable for those physiological variables in which reliability is high, such a strategy is less advisable when one uses psychosocial variables. (36) Therefore, to represent changes in the self-efficacy variables, we created residual scores by regressing each of the postrehabilitation scores on their own prerehabilitation scores. Then we performed zero-order correlations (1-tailed) between exercise adherence and each residual score separately for men and women because they had differential increases in all 3 self-efficacy variables.
We found that exercise adherence was significantly related to changes in (a) barrier efficacy for men (r = .58, p < .01) and women (r = .37, p < .05); (b) task efficacy (frequency) for men (r = .33, p < .01) and women (r = .48, p < .01); and (c) task efficacy (duration) for men (r = .32, p < .01) and women (r = .58, p < .01).
Purpose 3: Self-Efficacy and Exercise Adherence Following Phase II CR
Before establishing a relationship between the self-efficacy variables and exercise adherence postrehabilitation (ie, at follow-up), we performed a between-subjects ANOVA (ie, gender) on exercise adherence. Results showed that the frequency of exercise was similar between men (M = 4.11 times/wk) and women (M = 4.28 times/wk), F(1, 78) = .09, p > .05. Therefore, we collapsed the sample in our subsequent analysis. Here, we correlated the self-efficacy scores postphase II CR to the frequency of exercise reported at follow-up (ie, LSI). Zero-order correlations (1-tailed) showed that frequency of exercise was not correlated to barrier efficacy (r = .05, p > .05) or task efficacy (duration r =. 14, p > .05), but it was correlated with task efficacy (frequency r = .33, p < .01).
Purpose 4: Bidirectional Relationship Between Self-Efficacy and Mood
Pre- to Postrehabilitation
To explain the vigor/anxiety changes from the pre- to postphase II CR that we had found in our first analysis, we created residual scores using the same procedure outlined under purpose 2. Then we performed zero-order correlations (1-tailed) on the mood and self-efficacy residualized scores. Because no gender differences in the mood changes were shown in the first analysis (purpose 1), we again collapsed the sample for analysis. The findings indicated that the increase in vigor was positively correlated with the increase in barrier efficacy (r = .31, p < .01); task efficacy (frequency r = .26, p < .01); and task efficacy (duration r = .22, p < .05). Furthermore, the decrease in anxiety was negatively correlated with the increase in barrier efficacy (r = -.38, p < .01); task efficacy (frequency r = -.20, p < .05); and task efficacy (duration r = -.30, p < .01).
Postrehabilitation to Follow-Up
As was shown with the ANOVAs from the first analysis (purpose 1), only vigor changed significantly from postrehabilitation to follow-up, whereas all 3 self-efficacy variables significantly decreased at the same level in men and women. Therefore, to explain the decrease in vigor, we created residualized scores (from postrehabilitation to follow-up) for all 4 variables in question and performed zero-order correlations (1-tailed) on the newly created scores. These showed that the significant decrease in vigor was positively associated with a significant decrease in task efficacy (duration; r = .20, p .05) or task efficacy (frequency; r = .05, p > .05).
Our study had 4 main purposes. The first was to examine the time-course changes of task and barrier efficacy, anxiety, depression, and vigor during and following phase II CR in men and women. The second was to examine the relationship between the changes in task and barrier efficacy during phase II CR and exercise adherence. The third was to examine the relationship in men and women between task and barrier efficacy at the end of phase II CR and of exercise adherence following completion of the program. The fourth purpose was to examine the bidirectional relationships between task/barrier efficacy and mood during and after phase II CR in men and women.
Purpose 1: Self-Efficacy and Mood Over Time
As we had anticipated, task (duration and frequency) and barrier efficacy significantly improved from pre- to postphase II CR. An interesting finding was that the size of these changes was significantly larger in women than in men. This finding is in accord with McAuley and associates’ finding that task efficacy increased with exercise in middle-aged adults. (14) However, the current study is the first to demonstrate a differential increase in barrier efficacy between men and women. In examining the means from Table 2, one can see that women’s self-efficacy was significantly lower than men’s for all 3 variables when they entered the rehabilitation program, which is consistent with previous research, (10,15) However, women’s task and barrier efficacy increased to levels that were similar to those of the men at the end of phase II CR. All 3 self-efficacy variables, however, significantly decreased at postrehabilitation follow-up. It is important to note that further analyses revealed that these levels of self-efficacy at postrehabilitation follow-up were significantly higher than their prerehabilitation levels. Therefore, the data suggest that although men and women experienced significant gains in all 3 self-efficacy variables during phase II CR, it appears that women may have more to gain from this phase of rehabilitation than men do, in terms of increasing task and barrier efficacy that persists once they leave the program. This has important implications for physicians because physicians have consistently been shown to refer more men than women to phase II CR programs. (3,37) Our findings also suggest the need to modify this bias because women may actually benefit more from phase II CR than men do in strengthening their confidence to exercise and their confidence to overcome barriers so they will continue to exercise.
Our findings are consistent, with previous literature, which showed that phase II CR significantly decreased anxiety (22) and significantly increased vigor (26,27) in both men and women. Furthermore, the patients who participated in the study sustained the reduced level of anxiety at the postrehabilitation follow-up, although vigor significantly decreased. It should be noted that further analyses showed that vigor was significantly higher at follow-up than at prerehabilitation.
The lack off, change in depression from pre- to postphase II CR in men and women, was an interesting finding that is inconsistent with previous literature. (22) However, as one can see from Table 2, a floor effect appears in the present study, with the sample showing very low depression scores at the beginning of the study. This is problematic because a floor effect does not leave much room for a significant decrease in depression over time. Nonetheless, that men and women had similar levels of depression before starting a CR program is consistent with previous research (38) and may suggest that one should not expect a gender difference on depression levels in phase II CR.
Purposes 2 and 3: Self-Efficacy and Exercise Adherence During and After Phase II CR
We found it interesting that task efficacy (frequency) was the only self-efficacy variable related to men’s and women’s exercise adherence once they left their phase II CR program. During phase II CR, however, task and barrier efficacy were related to exercise adherence, which suggests that task and barrier efficacy are important determinants of exercise adherence during the initial phase of CR (ie, phase II), but that only task efficacy (frequency) may be the important efficacy variable in a more maintenance phase of rehabilitation (ie, phase III). That is, in the early stages of cardiac rehabilitation (phase II), the frequency of exercise is related to the patient’s general beliefs about his or her capabilities to exercise for various durations and frequencies and his or her confidence of continuing to exercise in the face of exercise barriers.
During this phase, more efficacious patients are likely to adhere more to their exercise programs and to reach a point at which daily exercise has become routine. At this point, exercise barriers may play a smaller role in influencing the frequency of patients’ exercise. Therefore, once patients make the transition from adopting exercise during phase II CR to maintaining it after they leave the program (ie, phase III rehabilitation), it is their confidence in performing the behavior on a regular basis that is important to continued exercise behavior. This supports McAuley’s (19) notion that different efficacy cognitions play a more salient role at different stages of the exercise-rehabilitation process.
Purpose 4: Bidirectional Relationship Between Self-Efficacy and Mood
Another purpose of our study was to explain the mood changes during and after phase II CR. We used a social cognitive perspective in an attempt to explain these mood changes. We found that increases in task and barrier efficacy were significantly related to a decrease in anxiety and an increase in vigor during phase II CR, which supported the tenets of social cognitive theory (16) and previous research. (28,30) Yet, only the decrease in task efficacy (duration) was significantly related to the significant decrease in vigor at postrehabilitation follow-up. This finding is consistent with an earlier point that different types of self-efficacy may be more important at various phases of a rehabilitation program. (19) More specifically, it appears that task efficacy (frequency), task efficacy (duration), and barrier efficacy are all related to exercise adherence and changes in mood during phase II CR, although only frequency is related to exercise adherence once patients leave the program. Furthermore, task duration appears to have the only significant relationship with the decrease in vigor once patients leave phase II CR. Therefore, we recommend that future studies in CR continue to include various task efficacy measures along with a barrier efficacy measure to magnify the potential to explain relationships among self-efficacy and other psychosocial (eg, mood) and behavioral variables (eg, exercise adherence).
Despite our promising findings, this study does have limitations that need to be considered when one interprets the findings and plans future research.
First, the sample consisted of 57 men and 24 women. Although the sample size in each cell was adequate to demonstrate gender differences, future studies should include samples with equal numbers of men and women to avoid violating the homogeneity of variance assumption between groups resulting from unequal cell sizes.
Second, the patients we sampled may not have previously experienced the particular exercise barriers on the barrier efficacy scale and may have over- or underestimated their efficacy judgment as a result. (39)
Third, the barriers we assessed were specific to phase II CR (15) and might not generalize to phase III CR. Therefore, additional research to discover barriers to continued exercise involvement in phase III CR is called for.
Fourth, we did not obtain information on the patients’ use of medications. Some medication could have been responsible for the decrease in anxiety and the increase in vigor.
Fifth, although we included a control condition to help control for threats of history and maturation, the length of this condition varied from 3 to 5 weeks. Despite this variation, we found no change in any of our psychological variables, which supports the usefulness a control condition. Future studies, however, should use randomized clinical trials that would allow one to have control and experimental conditions of similar lengths to strengthen the validity of the findings.
Finally, the problem of time may also have existed in the postrehabilitation follow-up assessment, which varied from 6 to 10 weeks. However, the number of weeks from postrehabilitation to the follow-up assessment was correlated to the task efficacy (duration), task efficacy (frequency), barrier efficacy, and vigor residual scores (ie, the 4 variables that significantly changed during this period) and found no significant relationships. Therefore, the variation in the length of time between postphase II CR and the follow-up assessment does not appear to be a confounding variable in the present study. Nonetheless, future researchers will want to standardize the timing of assessments to strengthen the internal validity of the study.
Despite these limitations, we found that women had significantly larger increases in task and barrier efficacy during phase II CR than men did, but men and women had similar decreases at postrehabilitation follow-up. Furthermore, changes in anxiety and vigor throughout the rehabilitation process were associated with changes in task and barrier efficacy in men and women. In future studies, researchers should continue to compare men and women on various psychosocial variables throughout the rehabilitation process to obtain a thorough understanding of gender differences during and following phase II CR.
Demographic Characteristics of the Sample
Variable n % n %
Married/common law 41 80.4 18 75.0
Divorced/separated 8 14.3 1 4.2
Single/widowed 3 5.4 5 20.8
[less than or equal to] Grade 9 9 15.8 2 8.3
High school 27 47.4 10 41.7
Postsecondary 21 36.8 12 50
Retired 27 49.1 14 58.3
Homemaker — 0 5 20.8
Employed 24 43.6 5 20.8
Unemployed 4 7.3 — 0
Myocardial infarction 32 57.1 13 54.2
Angina 2 3.6 1 4.2
Bypass surgery 12 21.4 3 12.5
Angioplasty/angiogram 9 16.1 7 29.1
Arthritis 20 35.1 10 41.7
Asthma 8 14.0 5 20.8
High blood pressure 25 44.0 10 41.7
Diabetes 5 8.8 3 20.8
Stomach problems 7 12.3 5 20.8
Gallbladder problems 5 8.8 6 25.0
Thyroid problems 3 5.3 9 37.5
Currently smoker 15 26.3 8 33.3
Never smoked 4 7.0 1 4.2
Quit 38 66.7 15 62.5
Variable [chi square](df) p
Single/widowed [chi square](5) = 7.14 NS
[less than or equal to] Grade 9
Postsecondary [chi square](2) = 1.52 NS
Unemployed [chi square](3) = 15.85 < .01
Angioplasty/angiogram [chi square](5) = 6.07 NS
Arthritis [chi square](1) = 0.31 NS
Asthma [chi square](1) = 0.58 NS
High blood pressure [chi square](1) = 0.03 NS
Diabetes [chi square](1) = 0.26 NS
Stomach problems [chi square](1) = 0.97 NS
Gallbladder problems [chi square](1) = 3.79 NS
Thyroid problems [chi square](1) = 13.9 < .01
Quit [chi square](2) = 0.56 NS
Means and Standard Deviations for the Self-Efficacy
and Mood Variables Across Time, by Gender
Variable Mail-out litation litation Follow-up
M 6.99 6.82 8.49 7.71
SD 1.63 1.69 0.73 1.12
M 5.60 5.63 8.09 7.36
SD 1.93 1.63 0.86 1.52
Task efficacy (duration)
M 6.53 6.64 9.04 8.72
SD 2.67 2.44 0.97 0.94
M 4.99 5.23 8.74 8.23
SD 2.43 2.72 1.38 1.53
Task efficacy (frequency)
M 6.04 6.21 8.55 8.12
SD 2.89 2.64 1.17 0.93
M 4.36 4.49 8.21 7.49
SD 2.88 2.43 1.91 1.97
M 1.25 1.20 0.77 0.79
SD 0.86 0.92 0.73 0.71
M 1.21 1.13 0.73 0.73
SD 0.99 0.91 0.84 0.59
M 0.60 0.58 0.48 0.52
SD 0.82 0.78 0.76 0.78
M 0.79 0.79 0.61 0.56
SD 0.98 0.97 0.91 0.85
M 2.03 1.97 2.49 2.31
SD 0.70 0.65 0.57 0.57
M 1.41 1.5 2.11 2.06
SD 0.66 0.54 0.38 0.44
Chris M. Blanchard received a doctoral research award from the Canadian Institutes of Health Research to support this research.
For further information, please address correspondence to Chris M. Blanchard, PhD, Director of Cancer Prevention, Behavioral Research Center, American Cancer Society, 1599 Clifton Road NE, Atlanta, GA 30329 (e-mail: firstname.lastname@example.org).
(1.) Cardiovascular Disease Statistics. Dallas, TX: American Heart Association; 2000.
(2.) Thomas RJ, Houston N, Lamendola C, et al. National survey on gender differences in cardiac rehabilitation programs. J Cardiopulm Rehabil. 1996;16:402-412.
(3.) Halm M, Penque S, Doll N, Beahrs M. Women and cardiac rehabilitation: Referral and compliance patterns. J Cardiovasc Nurs. 1999;13(3):83-92.
(4.) Schuster P, Waldron J. Gender differences in cardiac rehabilitation patients. J Adv Nurs. 1991;16:248-253.
(5.) Oldridge NB, Ragowski B, Gottlieb M. Use of outpatient cardiac rehabilitation services. Factors associated with attendance. J Cardiopulm Rehabil. 1992;12:25-34.
(6.) Ades PA, Waldmann ML, Polk DM, Coflesky JT. Referral patterns and exercise response in the rehabilitation of female coronary patients aged greater than or equal to 62 years. Am J Cardiol. 1992;12:25-31.
(7.) Maddux JE. Looking for common ground: A comment on Kirsch and Bandura. In: Maddux J, ed. Self-efficacy, Adaptation and Adjustment: Theory, Research, and Application New York: Plenum; 1995:377-385.
(8.) Ewart CK, Taylor CB, Reese LB, DeBusk RF. Effects of early postmyocardial infarction exercise testing on self-perception and subsequent physical activity. Am J Cardiol. 1983;51:1076-1080.
(9.) Jeng C, Braun L. The influence of self-efficacy on exercise intensity, compliance rate and cardiac rehabilitation outcomes among coronary artery disease patients. Rehabilitation. 1997; 12:13-24.
(10.) Schuster P, Wright C, Tomich P. Gender differences in the outcomes of participants in home programs compared with those in structured cardiac rehabilitation programs. Rehabilitation Nursing. 1995;20:93-101.
(11.) Foster C, Oldridge N, Dion W, et al. Time course recovery during cardiac rehabilitation. J Cardiopulm Rehabil. 1995;15: 209-215.
(12.) Carroll D. The importance of self-efficacy expectations in elderly patients recovering from coronary artery bypass surgery. Heart & Lung. 1995;24(1):50-59.
(13.) Jenkins LS, Gortner SR. Correlates of self-efficacy expectation and prediction of walking behavior in cardiac surgery elders. Ann Behav Med. 1998;20:99-103.
(14.) McAuley E, Courneya KS, Lettunich J. Effects of acute and long term exercise on self-efficacy responses in sedentary, middle-aged males and females. Gerontology. 1991;31:534-542.
(15.) Blanchard C, Rodgers W, Courneya KS, Daub B, Knapik G. Does barrier efficacy mediate the gender/exercise adherence relationship during phase II cardiac rehabilitation? Rehabil Psychol. In press.
(16.) Bandura A. Self-efficacy: The Exercise of Control. New York: Freeman; 1997.
(17.) Sallis JF, Pinski RB, Grossman RM, Patterson TL, Nader PR. The development of self-efficacy scales for health-related diet and exercise behaviors. Health Education Research. 1988; 3:283-292.
(18.) Home TE. Predictors of physical activity intentions and behaviour for rural homemakers. Can J Public Health. 1994;85: 132-135.
(19.) McAuley E. Self-efficacy and the maintenance of exercise participation in older adults. J Behav Med. 1993;16:103-113.
(20.) Sallis J, Hovell M, Hofstetter R, Barrington E. Explanation of vigorous physical activity during two years using social learning variables. Soc Sci Med. 1992;34(1):25-32.
(21.) Bock B, Albrecht A, Traficante R, et al. Predictors of exercise adherehce following participation in a cardiac rehabilitation program. Int J Behav Med. 1997;4(1):60-75.
(22.) Kugler J, Seelbach H, Kruskemper G. Effects of rehabilitation programmes on anxiety and depression in coronary patients: A meta-analysis. Br J Clin Psychol. 1994;33:401-410.
(23.) Lavie CJ, Milani CV. Effects of cardiac rehabilitation and exercise training programs in patients 75 years of age. Am J Cardiol. 1996;78:675-677.
(24.) Lavie CJ, Milani CV. Benefits of cardiac rehabilitation and exercise training in elderly women. Am J Cardiol. 1997;79:664-666.
(25.) Milani R, Lavie C, Cassidy M. Effects of cardiac rehabilitation and exercise training programs on depression in patients after coronary events. Am Heart J. 1996;132(4):726-732.
(26.) Oldridge NB, Streiner D, Hoffmann R, Guyatt G. Profile of mood states and cardiac rehabilitation after acute myocardial infarction. Med Sci Sports Exerc. 1995;900-905.
(27.) Engebretson T, Clark M, Niaura R, Phillips T, Albrecht A, Tilkemeier P. Quality of life and anxiety in a phase II cardiac rehabilitation program. Med Sci Sports Exerc. 1999;31(2) 216-223.
(28.) Perkins S, Jenkins L. Self-efficacy expectation, behavior performance, and mood status in early recovery from percutaneous transluminal coronary angioplasty. Heart & Lung. 1998;27(1):37-46.
(29.) Gortner S, Jenkins L. Self-efficacy and activity level following cardiac surgery. J Adv Nurs. 1990;15:1132-1138.
(30.) Stewart K, Kelemen M, Ewart C. Relationships between self-efficacy and mood before and after exercise training. J Cardiopulm Rehabil. 1994;14:35-42.
(31.) McAuley E. Efficacy, attributional, and affective responses to exercise participation. J Sport Exerc Psychol. 1991;13: 382-393.
(32.) Guidelines for Cardiac Rehabilitation Programs. Champaign, IL: American Association of Cardiovascular Pulmonary Rehabilitation. Human Kinetics; 1995.
(33.) Grove R, Prapavessis H. Preliminary evidence for the reliability and validity of an abbreviated profile of mood states, Int J Sport Psychol. 1992;23:93-109.
(34.) Godin G, Jobin J, Bouillon J. Assessment of leisure time exercise behaviour by self-report. A concurrent validity study. Can J Public Health. 1986;77:359-361.
(35.) Jacobs DR, Ainsworth BE, Hartman TJ, Leon AS. A simultaneous evaluation of ten commonly used physical activity questionnaires. Med Sci Sports Exerc. 1993;25:81-91.
(36.) Cohen J, Cohen P. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. Hillsdale NJ: Erlbaum; 1985.
(37.) Richardson L, Buckenmeyer P, Bauman B, Rosneck J, Newman I, Josephson R. Contemporary cardiac rehabilitation: Patient characteristics and temporal trends over the past decade. J Cardiopulm Rehabil. 2000;50:57-64.
(38.) Brezinka V, Dusseldorp E, Maes S. Gender differences in psychosocial profile at entry into cardiac rehabilitation. J Cardiopulm Rehabil. 1998;18:445-449.
(39.) DuCharme KA, Brawley LR. Predicting the intentions and behavior of exercise initiates using two forms of self-efficacy. J Behav Med. 1995;18:479-497.
Dr Blanchard is Director of Cancer Prevention, Behavioral Research Center, American Cancer Society, Atlanta; Drs Rodgers and Courneya are with the University of Alberta, Edmonton, Canada; and Mr Daub and Dr Black are with The Northern Alberta Cardiac Rehabilitation Unit, Alberta.
COPYRIGHT 2002 Heldref Publications
COPYRIGHT 2002 Gale Group