Idea Transcript
Western Journal of Nursing Research, 2004, 26(1), 31-46 Western Journal of Nursing Research February 2004, Vol. 26, No. 1
ARTICLE 10.1177/0193945903259350
Self-Efficacy Intervention Effect on Physical Activity in Older Adults Molly J. Allison Colleen Keller
This study determined the effectiveness of a self-efficacy intervention designed to improve selfefficacy and physical activity in older adults postcardiac event. An experimental three-group design tested the intervention, with treatment groups receiving 1 of 2 supportive telephone protocols (theory-based self-efficacy coaching or attention control). Outcome variables included selfefficacy expectations for physical activity (PA), self-reported PA, and PA performance (distance walked in 6 minutes). The self-efficacy intervention was effective in demonstrating greater PA performance when compared to the attention control intervention, and PA self-efficacy was significantly correlated with both measures of PA. There were significant main effects of time for PA self-efficacy and distance walked, and a significant interaction effect on the distance walked because of time and treatment condition. Although the self-efficacy intervention did not show a direct effect on level of PA self-efficacy as hypothesized, there was an indirect interaction effect on distance walked and physical activity confidence. Keywords: physical activity; theory-based intervention; self-efficacy; older adults
As many individuals aged 65 and older have been excluded from most exercise and cardiac recovery studies (Messinger-Rapport & Sprecher, 2002), little is known about physical activity (PA) behaviors of older adults when they leave formal cardiac rehabilitation. Many of the health behavior intervention studies that report significant behavioral changes are grounded in social cognitive theory (Dunn, Madhukar, Kampert, Clark, & Chambliss, 2002; King, 2001; Resnick & Spellbring, 2000; U.S. Department of Health and Human Services, 1996). Self-efficacy, an individual’s confidence or belief that a person can perform or is capable of performing an action, is a Authors’ Note: We would like to acknowledge Dr. Trish Hutchinson for her editorial advice, Bruce Paper for his statistical assistance, and Rosalinda Pena for her technical support. Molly J. Allison, R.N., Ph.D., Assistant Professor, Angelo State University; Colleen Keller, R.N., Ph.D., F.N.P., Professor and Chair, Department of Family Care Nursing, The University of Texas Health Science Center at San Antonio School of Nursing. DOI: 10.1177/0193945903259350 © 2004 Sage Publications
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
31
32
Western Journal of Nursing Research
salient predictor of health behavior change and maintenance (Bandura, 1997). Although there are studies that demonstrate significant increases in self-efficacy as a result of exercise, only a few studies used strategies theoretically hypothesized to enhance self-efficacy as the framework of the intervention (Allen, 1996; Calfas, Sallis, Oldenberg, & French, 1997; Dunn et al., 2002; McAuley, Courneya, Rudolph, & Lox, 1994). PHYSICAL ACTIVITY AND SOCIAL COGNITIVE RESEARCH
The research addressing the social cognitive component of PA as a health promotion behavior has been predominantly descriptive and exploratory, and in a number of studies, self-efficacy and PA behaviors have been significantly correlated (.35 to .53) (Carroll, 1995; Conn, 1998; McAuley, Lox, & Duncan, 1993; Oka, Gortner, Stotts, & Haskell, 1996; Wilcox & Strandt, 1996). In pre- or quasi-experimental studies (Calfas et al., 1997; McAuley, Schaffer, & Rudolph, 1995) researchers have demonstrated positive relationships between self-efficacy and PA, and the dynamic nature of selfefficacy over time. Few studies have used strategies hypothesized to enhance self-efficacy as a framework for intervention (Allen, 1996; Calfas et al., 1997; McAuley et al., 1994). With few exceptions, self-efficacy/physical activity research studies have not systematically affected change in selfefficacy as a mediating variable for the enhancement of PA. Pre- or quasi-experimental studies implied that PA might be mediated through self-efficacy. Results from experimental studies using theory-based intervention strategies to enhance activity behavior by enhancing selfefficacy do not fully support the theoretical linkages. Although all these studies resulted in significantly improved PA, the role of self-efficacy as a mediating process was not fully explained. The present research examined the influences of self-efficacy on PA behavior of older adults following a cardiac event and explored the mediation role of self-efficacy in the promotion of increased levels of PA in older adults. PURPOSE
The specific aims of this study were to test the effects of a nursing selfefficacy coaching intervention (SECIE) directed toward enhancement of PA
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
February 2004, Vol. 26, No. 1
33
self-efficacy and to determine effects of SECIE on PA in older adults’ recovery postcardiac event. The theory-based intervention (SECIE) was expected to have higher levels of self-efficacy and PA over time than the attention control (ACE) or usual care group (UC C). METHOD Design
A three-group time-series design (3 × 3) was used to examine the effects of a self-efficacy coaching intervention (SECIE) on self-efficacy and PA of older adults at 6 weeks and 12 weeks postcardiac event. Volunteers were randomly assigned into one of three treatment conditions: an experimental group (SECIE) that received the self-efficacy-based intervention; an ACE group that received a telephone follow-up protocol; or the UCC group that received usual care. Participants completed (a) Self-Efficacy Expectation Scales (SES), (b) Physical Activity Scale for the Elderly (PASE), (c) 6Minute Walk Test (6MWT) at baseline, 6, and 12 weeks; and (d) a demographic questionnaire at baseline. Sample and Setting
The sample was drawn from Phase I cardiac rehabilitation programs from one of two medical centers serving a large, sparsely populated area in rural southwest Texas. For inclusion, participants met the following criteria: aged 65 to 80 years, diagnosed with coronary heart disease (CHD), and referred to Phase I cardiac rehabilitation. Exclusion conditions for the study were (a) termination of Phase I cardiac rehabilitation, (b) complex dysrhythmias, (c) neurological or musculoskeletal disorders that affected mobility, and (d) inability to understand English or Spanish. Protection of human participants was reviewed and approved by the Institutional Review Board. Sample size calculations for evaluating effects of the intervention were based on measurements of reported changes (in excess of 21%) in self-efficacy. A sample size of 30 in each group was necessary to detect a medium effect of .30 on self-efficacy (Kruger, 2001), an alpha level of .05, and a power of .85. To account for participant attrition, oversampling by 20% required 36 participants to be recruited for each group.
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
Western Journal of Nursing Research
34
Description of Participants
The final sample included 83 volunteers with diagnosed coronary heart disease, 68.7% men (n = 57) and 31% women (n = 26). Ages ranged from 65 to 80 years (M = 71.78, SD ± 4.38). Of the participants, 71% entered Phase I cardiac rehabilitation for coronary artery bypass graft surgery (CABG), the remainder had a nonsurgical diagnosis. The sample was predominately married (79.5%) and Caucasian (80.7%), with the majority (54%) having at least a high school education. Of the participants, 31% had college or special training, and 19% had graduate education. All but two of the minority participants were Hispanic (16.9%). Recruitment and Data Collection Procedure
Eligible patients received information about the research study during their orientation to the cardiac rehabilitation (Phase I) program. The investigator met with the interested participants to determine if the participant met inclusion criteria. If they were interested in participation, the investigator obtained written informed consent and assisted the participant in completing the questionnaires. Participants drew from a shuffled prenumbered deck of cards and were randomly assigned to one of three groups: SECIE, ACE, UCC. An appointment for the first telephone contact (within 1 week) was arranged, and a written reminder was given to the participant during recruitment. All participants attended standard discharge teaching and Phase I cardiac rehabilitation. Appointments for data collection were arranged by the investigator at the participant’s home, cardiac rehabilitation centers, malls, tracks, or parks. Participants were contacted by telephone approximately 1 week prior to the 6-week and 12-week data collection points. Intervention Procedure
The intervention nurse was carefully selected based on CHD knowledge and communication skills and trained so that the interventions were implemented as planned to control intervener bias. Training included record keeping and review of intervention objectives. Written treatment protocols for the SECIE group and the ACE group were used and strictly followed at 2week intervals for the 12-week treatment period.
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
February 2004, Vol. 26, No. 1
35
SECIE protocol. The self-efficacy intervention was designed to incorporate the elements of the self-efficacy construct and was based on the reported success of multiple risk-factor behavioral changes made with telephone contact and a behavioral program based on social-cognitive theory (Allen, 1996; DeBusk et al., 1994). The SECIE protocol was designed to enhance confidence for PA using four sources of self-efficacy: performance accomplishments, verbal persuasion, physiological arousal, and vicarious experience through a telephone intervention. Participants in the SECIE group were interviewed according to the SECIE protocol that followed a four-point model of verbal self-efficacy enhancement: compliments on progress and encouragement for personal activity performance, observation of others with CHD performing physical activities, importance of including others in activities, and developing awareness of physiological responses to activity. Performance accomplishments were monitored and encouraged through individual goals, rehearsal of desired behaviors, and review of activity log. Verbal persuasion was used to encourage relative progress, attribute accomplishments to patient’s own abilities, and incorporate significant others into the activity plan for support and reinforcement of behaviors. Physiological arousal included questioning the participants about their responses to PA and helping the participant to interpret symptoms accurately. Vicarious experience drew attention to relative progress of other persons with CHD of similar age who are physically active. ACE telephone protocol. Participants in the ACE received a telephone call every 2 weeks, were asked by the intervention nurse to report their progress with the PA and exercise program prescribed in cardiac rehabilitation, and reminded to continue their physical activity/exercise program. Conversations were brief but supportive, and the four points covered in the self-efficacy intervention were not discussed. Usual Care
Participants assigned to the control group received the same discharge instructions and patient teaching as the intervention groups without the additional nurse-directed telephone support. Usual care participants were followed-up by their physician and may or may not have participated in Phase II cardiac rehabilitation.
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
36
Western Journal of Nursing Research
Instruments
SES. Self-efficacy expectations (Jenkins, 1989) were measured by a series of independent paper-and-pencil scales designed to measure self-efficacy expectation for behaviors related to recovery from a cardiac event. Selected scales included the following: (a) walking (16 items), (b) lifting (8 items), (c) general activities (17 items), (d) climbing stairs (7 items), and (e) work (14 items). Respondents were asked how confident they were about their ability to perform activities on a scale of 0 (no confidence) to 10 (total confidence). Each scale was scored independently. The numerical responses for each scale were summed and divided by the total number of activities. The resulting number was the level of confidence the individual had to perform the activity. In the Jenkins (1989) study, internal consistency for each efficacy scale was estimated by use of coefficient alpha at four data collection points, ranging from .97 to .67 across the four data collection points. Content validity, estimated by four expert raters, was .92. Recent revisions of this scale have included the addition of items relevant to other study groups, as well as deletions of repetitive items (Resnick & Jenkins, 2000). The PASE. Physical activity was assessed by the PASE (Washburn, Smith, Jette, & Janney, 1993), a norm-referenced scale that uses a Likert-type scale (never to 5 to 7 days) and yes or no answers to 12 types of PA including leisure, household, and work/volunteer activities. PASE scores were weighted and summed to provide an estimate of an older individual’s PA level. Stability of the PASE (N = 254) over time was assessed by the test-retest reliability correlation (.75, 95% confidence interval [CI] = .69 to .80) between baseline scores and follow-up scores (3 to 7 weeks). Validation measures, including content validity and PASE scores, were significantly associated with Sickness Impact Profile scores and perceived health status, grip strength, static balance, and leg strength showing strong convergent validity (Washburn et al., 1993). Test-retest reliability in our study was .72 (Allison, Keller, & Halfmann, 1998). The 6MWT was used to measure functional exercise capacity and was determined to be a suitable measure of outcome for clinical trials in patients with heart disease (Guyatt et al., 1986). Following careful Cardiac Rehabilitation Phase I assessment guidelines for activity progression that included electrocardiogram, blood pressure, pulse, and subjective signs and symptoms of cardiac instability, participants were instructed to walk continuously for 6 minutes. The Borg Perceived Exertion Scale (Borg, 1982) was used with each patient to protect against excessive or dangerous effort. All patients were asked to walk at a perceived exertion of 10 to 11 on the Borg
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
February 2004, Vol. 26, No. 1
37
scale (“Light” to “Somewhat light”) and to stop if for any reason they felt short of breath or light-headed. The distance covered in that time was measured precisely with a rolling measurement instrument (Rolatape Model 400, Spokane, WA). Time was measured with a digital sports stopwatch. Data Analysis
Descriptive statistics, univariate analysis, and frequency distributions were performed on all variables to describe the characteristics of the study population. Data were plotted for normal distribution and outliers. Multivariate analysis of variance and covariance techniques with repeated measures were used to test changes over time in outcome measures (selfefficacy, PA, and exercise capacity) between and within groups. Because there was no significant direct effect of the intervention on outcome variables, percentage change was used to describe outcome scores rather than differences between means. Post hoc analyses were performed whenever a significant difference (p < .05) was found (Groups × Time) on any of the dependent variables. RESULTS Group Description
Participants (n = 83) completing the study included SECIE = 28, ACE = 27, and UCC = 28. The attrition in this study was 23% (n = 25), and data were eliminated from the analysis, reducing the sample size of the study (n = 83). With this older-aged sample, the attrition was not surprising, with reasons for dropout including three deaths, progression of comorbidity of diabetes and respiratory disease, progression of cardiac disease, and change of residence or living with family. Because attrition rates greater than 20% are a bias concern (Polit & Hungler, 1999), t tests and chi-square tests were calculated on demographic variables to determine homogeneity of variance between participants that completed the study and those that were lost by attrition. There were no statistically significant differences between participants completing the study and those lost by attrition on the demographic variables of age, ethnicity, gender, education, or type of cardiac event. With a calculated effect size of .57, the sample size was large enough to detect true differences between groups with analysis of variance and covariance at the .05 level (Cohen, 1988).
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
Western Journal of Nursing Research
38
TABLE 1: Pearson’s Product-Moment Correlation Matrix of Dependent Variables (N = 83)
SES2 SES3 PASE 1 PASE 2 PASE 3 6 MWT 1 6 MWT 2 6 MWT 3
r r r r r r r r
SES1 SES2
SES3 PASE 1 PASE 2 PASE 3 6MWT1 6MWT2
.307** .220* .090 .241* .063 .437** .220* .099
.473** .464** .623** .511** .872** .844**
.867** .363** .524** .576** .487** .842** .724**
.351** .460** .298** .454** .533**
.688** .374** .491** .437**
.425** .630** .608**
.576** .473**
.881**
NOTE: SES = Self-Efficacy Expectation Scales; PASE = Physical Activity Scale for the Elderly; 6MWT = 6-minute walk test. *Correlation is significant at the .05 level (two-tailed). **Correlation is significant at the .01 level (two-tailed).
Dependent Variables
To examine the strength of the relationships between the dependent variables, a correlation matrix of the dependent variables was created (Table 1). Self-efficacy scores were significantly correlated with measures of PA and distance walked at baseline, 6 weeks, and 12 weeks. Although the relationship between self-efficacy and PA was not significant at baseline, it was significant at 6-week and at 12-week postcardiac event. The relationship between self-efficacy and distance walked was highly significant at most data collection points. As self-efficacy scores increased, so did the distance walked in 6 minutes. Baseline self-efficacy scores, however, were not related to either the distance walked or the PA scores at 12 weeks. Self-Efficacy
Mean self-efficacy scores were similar at baseline and increased for all groups over the 12-week period. At the end of the 12-week period, the SECIE group had higher mean scores than either the ACE or the UCC group. At all data collection points, the ranges and standard deviations of scores were similar for all groups. Self-efficacy scores for the SECIE group increased 64.51% from baseline to 6 weeks, as compared to 55.22% for the ACE group and 42.70% for the UCC control group. Although less impressive, the SECIE group percentage change in SES scores between 6 weeks and 12 weeks was 9.81%, as compared to 6.68% for the ACE group and
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
February 2004, Vol. 26, No. 1
39
7.67% for the UCC control group. The overall percentage change in mean self-efficacy scores during the 12-week period was greater for the SECIE group (80.64%) than for the ACE group (65.75%) or the UCC control group (53.64%). Self-Reported Physical Activity
In contrast to self-efficacy scores, PA scores present a different pattern from either the self-efficacy or the distance walked. Mean PA scores’ ranges and standard deviations are similar for all groups at all data collection points. Self-reported PA scores declined dramatically from the baseline to the 6week data collection point and then returned to near baseline scores at the 12-week data collection point. The SECIE group declined 23.84% and the ACE and UCC group 18.86% and 28.82%, respectively, from baseline to the 6-week data collection point. The improvement from 6 weeks to 12 weeks in self-reported PA was slightly higher for the two intervention groups (SECIE = 39.83%, ACE group = 42.30%) compared to the UCC group (UCC group = 29.26%). The overall improvement of PA scores, from the baseline to the 12-week data collection point, was greater for the ACE group (15.46%) than for the SECIE group (6.49%) or for the UCC group (–7.99%). The UCC group did not reach their prehospitalization self-reported PA level at 12 weeks. Distance Walked in 6 Minutes
Similar to the self-efficacy scores, the distance walked increased over the 12-week data collection period. The mean distance walked scores were higher for the SECIE group at 6 weeks and 12 weeks than for the ACE group or the UCC group. Ranges between minimum and maximum scores vary somewhat among groups. The telephone intervention group (ACE) had greater variation in scores at 6 weeks and 12 weeks than the other two groups. From baseline to the 6-week data collection point, the percentage increase in distance walked scores were greater for both experimental groups (ACE = 117.07%, SECIE = 114.55%) than for the control group (UCC = 76.66%). Percentage increases between the 6-week and the 12-week data collection points were small for all treatment conditions (SECIE = 9.21%, ACE = 6.61%, UCC = 4.05%). The overall percentage increases in distance walked scores for the 12-week period were almost identical for the two experimental groups (SECIE = 134.30%, ACE = 131.41%). Both experi-
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
Western Journal of Nursing Research
40
mental treatment groups’ 12-week percentage increase in distance walked score was considerably higher than the control group’s 12-week 83.89% increase in distance walked score. Mean PA scores at baseline were not significantly related to the mean PA scores at 12 weeks (r = .063), nor were the self-efficacy and PA scores significantly related at base line (r = .090). Medical conditions related to heart disease preceding baseline were highly variable. Baseline self-efficacy scores were less strongly associated with PA scores (r = .623, p = .01) than they were to the mean distance walked scores (r = .844, p = .01) at 12 weeks. Descriptive statistics indicated that distance walked and self-efficacy scores covaried; as confidence increased so did distance walked in 6 minutes, which was not the case with self-reported PA scores. These relationships indicated that PA baseline scores were representative of a PA history and not necessarily an indication of the ability of the older adult at the data collection points. For this reason, the remainder of the data analysis does not include PA scores. Multiple Analysis of Variance
A two-factor (Treatment × Time) MANOVA with repeated measures was performed to test the differences between the groups (SECIE, ACE, or UCC) in self-efficacy and distance walked. The results of the MANOVA are presented in Table 2. To prevent a Type I error, a separate one-way analysis of variance with repeated measures was used. The main effects of time for self-efficacy (F= 166.29, df = 1.35, p < .0001) scores were statistically significant indicating that self-efficacy related to distance walked increased over time for all groups. Post hoc tests indicate that these differences were significant (p < .0001) at each measurement level. Contrary to expected effects, there was no main effect of treatment for self-efficacy (F = .962, df = 2, p = .387) or interaction effects due to time and treatment (F = 1.83, df = 2.71, p = .151). The main effects of time on distance walked scores were also significant. Distance walked by older adults in 6 minutes increased significantly over the 12-week data collection period. Post hoc tests indicate significant differences between all data collection periods (p < .0001). As with self-efficacy, there was no main effect of treatment group for distance walked (F = 1.83, df = 2, p = .166); however, there was an interaction effect of time and treatment for the distance walked scores. None of the post hoc pairwise comparisons among treatment conditions reached significance. The mean distance walked in 6 minutes in the self-efficacy treatment group (SECIE) was greater
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
February 2004, Vol. 26, No. 1
41
TABLE 2: 6-Minute Walk and Self-Efficacy Measures Over Time by Research Groups Repeated Measures: MANOVA
Source
Measure
Time
SES 69752.61 6MWT 21042885.56 SES 529.60 6MWT 315842.57
Treatment Time × Treatment
SS
SES 1536.95 6MWT 529508.38 Error (time) SE Score 33557.64 6Mwalk 6569308.49
df 1.35 1.473 2 2 2.71 2.95 108.29 117.86
MS
F
p
51529.37 14282764.95 264.80 157921.28
166.29 256.26 .962 1.834
.000 .000 .387 .166
567.71 179700.73 309.88 55736.11
1.83 3.224
.151 .026
NOTE: SES = Self-Efficacy Expectation Scales; 6MWT = 6-minute walk test.
than the mean distance walked in 6 minutes for the telephone group (ACE) and even greater than for the usual care group (UCC). Although there is a statistically significant interaction effect of time and group on distance walked, we cannot determine the differential effects of the treatments (SECIE, the ACE, or the UCC). The 6MWT (minute walk test) scores rise sharply for all groups up to 6 weeks. The rate of increase in distance walked after 6 weeks lessens for all groups between 6 and 12 weeks; however, the self-efficacy treatment group (SECIE) maintained higher scores than the ACE group or the UCC group. The significance of the changes produced by the treatment conditions (SECIE, ACE, or UCC) on PA of older adults after a cardiac event was further evaluated using a two-factor (Treatment × Time) MANCOVA with repeated measures, with self-efficacy scores as a covariate of distance walked (see Table 3). Confirming the results of the MANOVA test, the analysis yielded highly significant effects of time on distance walked scores (F = 18.70, df = 1.78, p < .0001). Post hoc tests indicate that participants gained significant increases in distance walked at both data collection periods (Level 1 vs. Level 2, F = 28.81, df = 1, p < .0001; Level 2 vs. Level 3, F = 8.509, p = .005). There was no significant treatment effect over time on distance walked scores (F = 1.17, df = 3.56, p = .325). None of the three treatment conditions could account for the increase in the distance walked in 6 minutes over time. There was, however, a highly significant interaction effect between the covariate self-efficacy scores and the distance walked scores (F = 48.63, df = 1.78, p < .0001).
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
Western Journal of Nursing Research
42
TABLE 3: 6-Minute Walk Over Time by Research Groups With Self-Efficacy as Covariate: MANCOVA
Source
SS
Time Time × Treatment Time × SES 1 to 3 Time × Treatment × SES 1 to 3 Error (6MW)
df
MS
F
955598.48 1.78 5366809.72 18.70 119785.034 3.56 33644.76 1.17 2484759.15 1.78 1395819.38 48.63 313791.64 3.56 88136.60 3.07 3934644.12 137.07 28705.14
p .000 .325 .000 .023
NOTE: SES = Self-Efficacy Expectation Scales; 6MW = 6-minute walk.
The treatment and time interaction with self-efficacy as covariate on the distance walked was significant (F = 3.07, df = 3.56, p = .023). In the intragroup contrasts, using the Bonferroni procedure, no differences gained were significantly different in distance walked in pairwise comparisons of groups. The interaction effect was present but could not be attributed to one of the treatment groups. There was a highly significant interaction effect on distance walked because of time and self-efficacy. There was also a significant three-way interaction effect of treatment condition, distance walked, and the covariate self-efficacy scores (F = 3.030, df = 3.56, p = .023). Post hoc intragroup contrasts using the Bonferroni procedure did not show any significant contrasts between SECIE, the ACE, or the UCC. DISCUSSION
This social cognitive theory-based nursing intervention did not show a direct effect on older adults’ self-efficacy for PA postcardiac event; however, there was an indirect interaction effect on increased distance walked related to treatment conditions and the older adults’ self-efficacy for PA over the 12-week period. These findings are encouraging and lead us to ask important questions about clinical relevance and research design. All the self-efficacy sources described by Bandura (1997) were used in the theory-based intervention to reinforce the importance of PA with participants, reinforce PA behavior, and help participants interpret their physical response to activity by telephone over a 3-month period. To control for the motivational effect of the telephone call by a CHD nurse, we also had a group that received the attention of the telephone call but did not follow the social cognitive protocol.
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
February 2004, Vol. 26, No. 1
43
The findings of this study add to the literature by confirming a strong relationship of self-efficacy to PA over time during the rehabilitation process following a cardiac event and are parallel to those results reported others (Kruger, 2001; Perkins & Jenkins, 1998). The comparable relationships are most likely due to the specificity of cardiac rehabilitation prescription regarding exercise, personal motivation, and confidence gained from learning new behaviors for the period assessed. We may find that social cognitive theory-based interventions may affect categories of patients differently. For example Oka, DeMarco, and Haskell (1999) found that high and low levels of self-efficacy were associated with different outcomes. Resnick and Spellbring (2000) found a significant difference in self-efficacy expectations between those participants who adhered to the protocol and those who did not. Although self-efficacy has been reported to be a strong predictor of PA, in this study baseline self-efficacy did not correlate to either self-reported PA or distance walked at the completion of the 3-month period. Calfas et al. (1997) also found this to be true. An older adult’s confidence level at baseline after a cardiac event may be of limited use in explaining PA in the future, which can be explained by or attributed to the social-cognitive principle reciprocal determinism, a cognitive process influenced by cognitive, affective, and biological events (Bandura, 1986, 1997). This finding has important clinical relevance. Older adults’ beliefs about their ability to perform health-promoting activities are formational in nature. Their confidence at the time of an event may not be a good predictor of future behavior. The results of this study showed a significant interaction effect of time and treatment condition on PA, when self-efficacy and PA were considered as separate outcome variables. Over time, the self-efficacy coaching treatment made a significant difference in the distance walked but was not affected by the intervention. In this study, improvement in distance walked in 6 minutes for the theory-based intervention group was greater than that of the ACC group who received a less structured supportive treatment or the control group that did not receive any additional support. This finding was interesting, in that although the intervention increased the desired outcome, it did not increase the mediator, self-efficacy, suggesting that perhaps it was the specified attention that contributed to the outcome or that the measure of self efficacy was not theoretically sensitive. The treatment groups had superior PA outcomes, which supports clinical use of telephone communication with older adults through the rehabilitation process. The group that received telephone coaching based on the sources of self-efficacy did have higher levels of self-efficacy and PA. Because the
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
Western Journal of Nursing Research
44
difference in the PA outcomes between groups could not be attributed to the direct effect of treatment on self-efficacy, the appropriateness of relying solely on the cognitive factor as the total basis for intervention should be questioned. Even though telephone-counseling methods are effective and cost-effective support strategies (Castro, King, & Brassington, 2001), other factors should be considered. A logical extension of this study would be to explore environmental and behavioral mediators of PA that could be modified through intervention. King’s (2001) work with community-based settings and the socially and racially relevant intervention described by Fleury and Walton (2002) suggests that the environment has a great deal of influence on motivation and behavior. In this study, the method of PA measurement made a difference in the strength of the relationship between self-efficacy and PA. The 6MWT scores were more strongly correlated to the self-efficacy scores than were the self-reported PA scores. The ease of use and the precision of the 6MWT made it a very good instrument for nurses to use in the field. A few limitations should be noted. First, the ACC protocol may have served as proxy reinforcement similar to that of verbal persuasion in the selfefficacy protocol, suggesting that the ACC treatment protocol may not have been sufficiently different from the theory-driven treatment. Second, treatment integrity was ensured as is possible in field testing of interventions. Although careful records were kept by the intervention nurse and monitored by the investigator to ensure that telephone support protocols were delivered as intended, the ACC group might have served as a treatment group. Last, the measurement of self-efficacy, using the SES, might have been strengthened with corresponding instruments. NOTE 1. The study was supported in part by grants from Sigma Theta Tau International Beta Alpha Chapter. Results of the study were presented at the American Heart Association Scientific Sessions 2000 and the Southern Nursing Research Conference 2001.
REFERENCES Allen, J. K. (1996). Coronary risk factor modifications in women after coronary artery bypass surgery. Nursing Research, 45, 260-265. Allison, M., Keller, C., & Halfmann, P. L. (1998). Selection of an instrument to measure physical activity of elders. Rehabilitation Nursing, 23, 309-314.
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
February 2004, Vol. 26, No. 1
45
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman. Borg, G.A.V. (1982). Psychophysical basis of perceived exertion. Medicine and Science in Sports and Exercise, 14, 377-381. Calfas, K. J., Sallis, J. F., Oldenburg, B., & French, M. (1997). Mediators of change in physical activity following an intervention in primary care: PACE. Preventive Medicine, 26, 297-304. Carroll, D. L. (1995). The importance of self-efficacy expectations in elderly patients recovering from coronary artery bypass surgery. Heart and Lung, 24, 50-59. Castro, C. M., King, A. C., & Brassington, G. S. (2001). Telephone versus mail interventions for maintenance of physical activity in older adults. Health Psychology, 20, 438-444. Cohen, J. (1988). Statistical power analysis (2nd ed.).Hillsdale, NJ: Lawrence Erlbaum. Conn, V. S. (1998). Older adults and exercise: Path analysis of self-efficacy related constructs. Nursing Research, 47, 180-189. DeBusk, R. F., Miller, N. H., Superko, R., Dennis, C. A., Thomas, R. J., Lew, H. T., et al. (1994). A case-management system for coronary risk factor modification after acute myocardial infarction. Annals of Internal Medicine, 120, 721-729. Dunn, A. L., Madhukar, H. T., Kampert, J. B., Clark, C. G., & Chambliss, H. O. (2002). The DOSE study: A clinical trial to reexamine efficacy and dose response of exercise as a treatment for depression. Controlled Clinical Trials, 23, 584-603. Fleury, J., & Walton, M. C. (2002, October). Community-based physical activity intervention to promote cardiovascular health in rural-dwelling African Americans [Abstract No. 10]. Proceedings of the American Heart Association 4th Scientific Forum of Outcomes Research in Cardiovascular Disease and Stroke. Washington, DC. Guyatt, G. H., Sullivan, M. J., Thompson, P. J., Fallen, E. L., Pugsley, S. O., Taylor, D., et al. (1986). The 6-minute walk: A new measure of exercise capacity in patients with chronic heart failure. Canadian Medical Association Journal, 132, 919-923. Jenkins, L. S. (1989). Jenkins instruments: Self-efficacy expectation scales and activity checklists for selected cardiac recovery behaviors. (Available from Louise S. Jenkins, Ph.D., R.N., The Walter Schroeder Chair in Nursing Research, School of Nursing, University of Wisconsin, 1909 E. Hartford Avenue, P.O. Box 413, Milwaukee, WI 53201) King, A. C. (2001). The coming of age of behavioral research in physical activity. Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine, 23, 227-228. Kruger, J. (2001). A meta-analysis of factors related to self-efficacy for exercise in older adults. Dissertation Abstracts International, 62(03B), 1353 [Dissertation Abstracts Online AA3008081]. McAuley, E., Courneya, K. S., Rudolph, D. L., & Lox, C. L. (1994). Enhancing exercise adherence in middle-aged males and females. Preventive Medicine, 23, 498-506. McAuley, E., Lox, C., & Duncan, T. E. (1993). Long-term maintenance of exercise, self-efficacy, and physiological change in older adults. Journal of Gerontology, 48, P218-P224. McAuley, E., Schaffer, S. M., & Rudolph, D. (1995). Affective responses to acute exercise in elderly impaired males: The moderating effects of self-efficacy and age. International Journal on Aging and Human Development, 41, 13-27. Messinger-Rapport, B. J., & Sprecher, D. (2002). Prevention of cardiovascular diseases. Coronary artery disease, congestive heart failure, and stroke. Clinics in Geriatric Medicine, 18, 463-483. Oka, R. K., DeMarco, T., & Haskell, W. L. (1999). Perceptions of physical fitness in patients with heart failure. Progress in Cardiovascular Nursing, 14(3), 97-102.
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016
46
Western Journal of Nursing Research
Oka, R. K., Gortner, S. R., Stotts, N. A., & Haskell, W. L. (1996). Predictors of physical activity in patients with chronic heart failure secondary to either ischemic or idiopathic dilated cardiomyopathy. American Journal of Cardiology, 77, 159-163. Perkins, S., & Jenkins, L. (1998). Self-efficacy expectation, behavior performance, and mood status in early recovery from percutaneous transluminal coronary angioplasty. Heart and Lung, 27, 37-46. Polit, D. F., & Hungler, B. P. (1999). Nursing research: Principles and methods (6th ed.). Philadelphia: Lippincott. Resnick, B., & Jenkins, L. S. (2000). Testing the reliability and validity of the self-efficacy for exercise scale. Nursing Research, 49, 154-159. Resnick, B., & Spellbring, M. (2000). Understanding what motivates older adults to exercise. Journal of Gerontological Nursing, 26(3), 23-42. U.S. Department of Health and Human Services. (1996). Physical activity and health: A report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion. Washburn, R. A., Smith, K. W., Jette, A. M., & Janney, C. A. (1993). The physical activity scale for the elderly (PASE): Development and evaluation. Journal of Chronic Epidemiology, 46(2), 153-162. Wilcox, S., & Strandt, M. (1996). Relations among age, exercise, and psychological variables in a community sample of women. Health Psychology, 15, 110-113.
Downloaded from wjn.sagepub.com at PENNSYLVANIA STATE UNIV on May 11, 2016