| | Active for Life: Final Results from the Translation of Two Physical Activity ProgramsBackgroundMost evidence-based programs are never translated into community settings and thus never make a public health impact. DesignActive for Life (AFL) was a 4-year translational initiative using a pre–post, quasi-experimental design. Data were collected from 2003 to 2007. Analyses were conducted in 2005 and 2008. Setting/participantsNine lead organizations at 12 sites participated. Active Choices participants (n=2503) averaged 65.8 years (80% women, 41% non-Hispanic white). Active Living Every Day (ALED) participants (n=3388) averaged 70.6 years (83% women, 64% non-Hispanic white). InterventionIn AFL, Active Choices was a 6-month telephone-based and ALED a 20-week group-based lifestyle behavior change program designed to increase physical activity, and both were grounded in social cognitive theory and the transtheoretical model. The interventions were evaluated in Years 1, 3, and 4. An adapted shortened ALED program was evaluated in Year 4. Main outcome measureModerate- to vigorous-intensity physical activity, assessed with the CHAMPS self-reported measure. ResultsPosttest survey response rates were 61% for Active Choices and 70% for ALED. Significant increases in moderate- to vigorous-intensity physical activity, total physical activity, and satisfaction with body appearance and function, and decreases in BMI were seen for both programs. Depressive symptoms and perceived stress, both low at pretest, also decreased over time in ALED. Results were generally consistent across years and sites. ConclusionsActive Choices and ALED were successfully translated across a range of real-world settings. Study samples were substantially larger, more ethnically and economically diverse, and more representative of older adult's health conditions than in efficacy studies, yet the magnitude of effect sizes were comparable. Introduction  Translational studies focus on whether effective interventions work in real-world settings.1, 2, 3, 4, 5, 6 Their purpose is not only to examine whether evidence-based programs improve outcomes but also to understand how organizations implement these programs to reach greater numbers of people, to examine the social and environmental influences on program delivery, and even to study how practitioners may adapt these programs to better suit their needs and the needs of their constituents. Translational studies are critical in understanding key issues related to the adoption and dissemination of effective programs.7, 8 The Task Force on Community Preventive Services concluded that individually adapted behavior change programs (which can include telephone-delivered, group-based, and other program formats) are a strongly recommended approach for promoting physical activity in communities.9 The essential components of these interventions include strategies such as realistic goal setting, identification of barriers, problem solving, self-monitoring (tracking), and provision of feedback and reinforcement.9, 10 Furthermore, a recent review of telephone-based interventions for promoting physical activity and dietary change concluded that there is a solid evidence base to support the efficacy of this delivery method; however, the authors recommended that more studies be conducted to examine broader dissemination of this approach into population health practice.11 Active for Life (AFL), an initiative of the Robert Wood Johnson Foundation, was one of the first national studies to examine whether evidence-based physical activity programs could reach large numbers of adults aged ≥50 years, have similar impacts to earlier efficacy studies, and be sustained over time within existing community or clinical settings.5, 12 Results from the first year of the study found that the two physical activity programs tested—Active Choices, a 6-month telephone-delivered program; and Active Living Every Day (ALED), a 20-week group-based program—increased physical activity with effects comparable to results reported in the initial efficacy studies.13 The AFL initiative has accrued three additional years of experience (outcome data collected in two of these years). The purpose of this paper is to expand prior work by examining results over multiple years. The larger sample permits a more rigorous examination of several questions about the programs that are vital for translational research: 1.How do the characteristics of program participants differ across study years? 2.How effective are the programs in increasing physical activity and improving related outcomes for older adults? 3.How consistent are effect sizes over time and settings as different cohorts of study participants undergo the same intervention? 4.What is the impact of making a substantial site-initiated modification of one of the programs (i.e., shortening ALED from a 20-week to a 12-week program)? Methods Program Overviews In the AFL initiative, Active Choices was a 6-month program delivered through one face-to-face meeting and up to eight one-on-one telephone counseling calls.14, 15, 16, 17 During the face-to-face meeting, the health educator established rapport with the participant and covered key programmatic material including a review of program expectations; the formulation of a physical activity plan and goals; a discussion of interests, motivation, perceived benefits, and perceived barriers to physical activity; and a discussion of exercise safety. Participants were given an exercise log, a pedometer, and a local resource guide of physical activity programs. Finally, a call schedule was established. Participants then received biweekly telephone calls for the first 2 months and monthly telephone calls for the next 4 months. Counseling was tailored to the person's readiness for physical activity change18 and emphasized key constructs in social cognitive theory.19 During each call, the health educator inquired about potential changes in health and exercise-related injuries. Next, the participant's physical activity since the last call was discussed (focusing on days, minutes, and step counts) as well as the participant's current stage of readiness for change. Based on this information, the health educator chose cognitive and behavioral topics for discussion (e.g., barriers/benefits, goal-setting, self-monitoring). Finally, the health educator assessed whether the participant wished to modify activity goals. Health educators could send tip sheets based on call content. Active Living Every Day is a 20-week physical activity intervention delivered in a small-group setting. The curriculum is a dissemination version of Project Active,20, 21 a randomized clinical trial that incorporated constructs from social cognitive theory19 and the transtheoretical model18 to help participants accumulate a minimum of 30 minutes of moderate-intensity physical activity on most days of the week. After the first 3 years, the lead organizations requested that the program be shortened to 12 weeks to enhance recruitment and foster partnerships with organizations that offered shorter programs. The shortened program maintained its essential elements by: (1) eliminating elements not being used, (2) eliminating or making optional material and activities not directly related to the program essential elements, (3) combining related topics that were previously addressed in separate sessions, (4) decreasing the amount of review and reinforcement of lifestyle skills covered in multiple sessions, and (5) extending the length of each session from 60 to 90 minutes. Lead Organizations and Intervention Staff Organizations responded to a competitive request for proposals. Prior to being selected, they chose the program (Active Choices or ALED) that they thought best suited the needs of their organization and community. Of the nine funded lead organizations, four selected the Active Choices program and implemented it at five distinct sites, and five selected the ALED program and implemented it at seven distinct sites from 2003 through 2007. Organizations are reported elsewhere13 and are listed at the end of this article. Initially, Active Choices staff at lead organizations was trained and approved by program developers. These staff became train-the-trainers and trained and approved their own staff. ALED staff was trained via online modules and a face-to-face or webcast workshop and certified by Human Kinetics. At several ALED sites, former participants became paid or volunteer facilitators. Program developers provided programmatic technical assistance as needed. The AFL National Program Office also provided technical assistance for program marketing, recruitment, budget management, and sustainability planning. A detailed process evaluation tracked intervention fidelity and the use of technical assistance. Program developers, the National Program Office, and the organizations received process evaluation feedback to help ensure that program adaptations were appropriate and did not compromise fidelity to essential elements of the programs. A paper describing the process evaluation and major findings is currently under review (SF Griffin, University of South Carolina & Clemson University, unpublished findings, 2008). Each lead organization aimed to recruit 100 participants in Year 1 (beginning March 2003); 200 participants in Year 2; and 300 participants in both Years 3 and 4. The lead organization tailored recruitment strategies to the communities it targeted (strategies reported elsewhere13). Over the 4 years, organizations relied on a similar set of recruitment strategies but became more focused and targeted. Participants Participants had to be aged ≥50 years (no upper limit); sedentary or under-active (i.e., ≤2 days/week and <120 minutes/week); and free of medical conditions or disabilities that required higher levels of supervision, as determined by the lead organization. Each lead organization was required to design and implement a risk management strategy that included risk assessment, participant education, and healthcare provider involvement as appropriate (prior to and during the program). All organizations were encouraged to use the revised Physical Activity Readiness Questionnaire (PAR-Q)22 for risk assessment. All organizations educated participants on risks of physical activity and methods to minimize risks (e.g., start low and go slow) and included safety tips and information on warning signs and symptoms. This approach is associated with few adverse events.23 At two sites, participants with risks based on PAR-Q items were required to obtain written medical approval before starting the program. At the other sites, participants were advised to discuss these risks with their healthcare provider, but were not routinely required to obtain written medical approval. Site directors had the discretion to require medical approval and were responsible for approving and overseeing the risk-management procedures. Design and Procedure The evidence supporting the efficacy of Active Choices14, 15, 16, 24, 25, 26 and ALED20, 21 comes from rigorous RCTs. AFL was designed to study research to practice translation by examining whether the magnitude of effects in community settings were comparable to what has been reported in controlled research settings and to examine the representativeness (or reach) of the sample (see Estabrooks and Gyurcsik27 for a discussion of the evaluation of translational projects). The study was not designed to compare the programs. Thus, a pre–post design was used, and the study was powered to examine change over time in each program separately. Study data were collected from 2003 to 2007. All participants completed an informed consent form that was approved by the two university IRBs and review boards or legal departments of the lead organizations. Participants then completed a brief questionnaire that collected personal and demographic information followed by a comprehensive survey in Year 1. Although sites recruited participants and delivered the intervention in Year 2, outcome data were not collected so that the evaluation team could analyze Year-1 findings and plan for modifications to the evaluation in the final 2 years. Sample size calculations indicated that it was not necessary for all participants to complete primary outcome measures. At the same time, we wanted to know how participants compared across time. Thus, a short survey was developed to collect several key outcomes more efficiently. In Years 3 and 4 we requested a priori that the first 100 participants per site complete the comprehensive survey and the remaining participants complete the short survey. In Year 4 of the study, the 12-week adaptation to ALED was evaluated. ALED participants completed a pretest and a 12- and 20-week posttest survey. The length and format of the pretest and the 20-week posttest surveys were as described above; however, to reduce burden for the sites and participants, the 12-week survey was a short survey for all participants. At Active Choices sites posttest surveys were sent directly to participants, along with postage paid envelopes, 2 weeks prior to the scheduled program end date. The same was true for Year 1 of ALED except that one site administered the posttest survey in sessions 19 or 20. In Year 3, ALED sites administered the posttest survey in sessions 19 or 20. In Year 4, the 12-week ALED posttest survey was administered by all sites in sessions 11 or 12 and the 20-week posttest survey was sent directly to participants with a postage-paid envelope. Participants not present during ALED sessions were sent the posttest survey to complete and return. For both programs, postcard reminders and a second survey were sent to nonresponders. Participants who returned their pretest and/or posttest survey entered a drawing for $20 gift cards to a local retail store. After the first year of the study, one site elected not to participate in the gift card drawing. Measures Sociodemograhics and health-related variables Participants reported their date of birth; race; ethnicity; marital status; gender; years of formal education; income (long survey only); height and weight (to compute BMI28 in kg/m2); health rating; and presence of chronic health conditions.29 Physical activity The comprehensive survey contained the 41-item Community Healthy Activities Model Program for Seniors (CHAMPS) physical activity measure.30 This self-report measure includes activities typically undertaken by older adults for exercise, activities undertaken in the course of their day that are physical in nature, and recreational activities that provide physical activity. Minutes/week spent in moderate- and vigorous-intensity physical activity (MVPA) and in light-, moderate-, and vigorous-intensity physical activity (total physical activity) were derived. The CHAMPS has strong psychometric properties, including demonstrated validity,31 test–retest reliability,31 and sensitivity to change.16, 17, 30, 32, 33 MVPA and total physical activity were highly positively skewed at pretest and somewhat skewed at posttest. A square root transformation successfully normalized these variables. A secondary three-item measure29 assessed participation, frequency, and duration of MVPA in all participants in order to classify participants as sedentary (<10 min/wk); regularly active (≥30 min/d, ≥5 d/wk); or under-active (not meeting criteria for sedentary or regularly active).34 Depressive symptoms The comprehensive survey included the widely used 10-item Center for Epidemiological Studies Depression Scale (CES-D).35, 36, 37 Participants rated the frequency with which they experienced symptoms of depression during the past week. Perceived stress The comprehensive survey included a 4-item version of the Perceived Stress Scale,38, 39 an extensively used questionnaire that was designed to measure the degree to which situations in one's life are appraised as stressful. Body satisfaction Participants rated their satisfaction over the past 4 weeks with nine aspects of body appearance (comprehensive survey) and function (both surveys) on a 7-point scale. Two subscales were derived from factor analysis40: satisfaction with body appearance (3 items) and satisfaction with body function (6 items). Statistical Analyses All analyses that refer to a bi-directional distribution used 2-sided p values. Statistical analyses were conducted in 2005 (Year-1 data) and 2008. Chi-square analyses tested differences in categoric baseline variables across years. ANOVAs tested differences in continuous baseline variables across years. To address changes from pretest to posttest in hours/week of MVPA, total physical activity, depressive symptoms, perceived stress, satisfaction with body function and appearance, and BMI, repeated measures ANCOVAs (using SAS PROC MIXED) were conducted that controlled for site clustering; race (non-Hispanic white, black, Latino, or other ethnic minority); gender; BMI; health rating; and education (high school or fewer years versus some college or college graduate). SAS PROC GLIMMIX, controlling for the same covariates, assessed change in the dichotomous outcome variable. Both types of analyses use a likelihood-based approach to accommodate data missing at random. Participants were recruited from five geographic sites for Active Choices and seven geographic sites for ALED; therefore, all change analyses took into account this site-level clustering and had 4 and 6 degrees of freedom (N – 1), respectively. Two types of change analyses were conducted. First, changes among participants were examined using all available data (i.e., no imputations were made for missing data). Second, because there is no control group in AFL, more conservative analyses were conducted and assumed no change in outcomes among those who did not return posttest surveys by carrying forward baseline values. An effect size (d=[posttest mean – pretest mean] / pretest standard deviation) was computed for each continuous dependent variable using means that were adjusted for site clustering, race, gender, BMI, health rating, and education.41 Effect sizes of d=0.2 were considered small, d=0.5 medium, and d=0.8 large.42 In order to examine consistency in outcomes across sites, analyses were also conducted for each site separately if the overall effect was significant for continuous variables. Results All Year-1 results were published elsewhere13 but are included in this manuscript to show data across years. Description of the Sample Across years (not including Year 2), 2503 and 3388 were enrolled into Active Choices and ALED, respectively, and completed a pretest survey. Participant characteristics for each year, separately by program, are shown in Table 1. Variables that differed significantly (p<0.05) by year, for each program, are described. For Active Choices, significant differences by year were found for race/ethnicity, education, income, diabetes, and physical activity. The percentage of ethnic minorities, those with less than a high school education, and those who were sedentary increased across years. The percentage with a household income less than $30,000 was highest in Year 3. More participants in Year 3 had diabetes. For ALED, significant differences by year were found for a greater number of variables. Compared to Year 1, a larger percentage of ALED participants in Years 3 and 4 were aged >75 years, had less than a high school education, had hypertension, and had arthritis. A smaller percentage of ALED participants in these 2 years, however, were completely sedentary. BMI was higher in Year 4 than Year 3. MVPA and total physical activity were higher in Year 4 than in Year 1. Satisfaction with body appearance was lower in Year 1 than in Year 4, and satisfaction with body function was lower in Year 1 than in Years 3 and 4. Finally, perceived stress was higher in Year 3 than in Year 4. Analysis of Survey Nonresponders and Participant Withdrawals A total of 275 (72%) Active Choices participants in Year 1, 687 (60%) in Year 3, and 558 (57%) in Year 4 returned their posttest surveys. For ALED these numbers were 333 (73%) and 1095 (76%) for Years 1 and 3. In Year 4, 1169 (78%) participants returned their posttest survey at 12 weeks, and 925 (62%) returned their follow-up survey at 20 weeks (i.e., 8 weeks after the program ended). For both programs, participants who returned posttest surveys were significantly (p<0.01) more likely than those who did not to be white, high school or college educated, and healthier (self-rated). Additionally, for Active Choices these participants were more likely to have a household income of $30,000–$59,000, report higher baseline physical activity, and report lower levels of baseline stress and depressive symptoms. For ALED, those free from coronary heart disease were more likely to return the 20-week survey, whereas women were more likely than men to return the 12-week survey. For Active Choices, 25 (7%) participants in Year 1, 83 (7%) in Year 3, and 79 (8%) in Year 4 actively withdrew from the program. For ALED, these numbers were 44 (10%), 233 (16%), and 127 (9%). For both programs, those who withdrew were significantly (p<0.01) more likely than those who did not to be white, older, and leaner. Additionally, for Active Choices, those with lower educational levels were more likely to withdraw, whereas for ALED those with coronary heart disease were more likely to withdraw. Change Analyses Adjusted mean scores on the outcome variables for pretest and posttest and effect sizes for these changes are summarized in Tables 2 (Active Choices) and 3 (ALED). As reported elsewhere,13 all of the changes from pretest to posttest in Year 1 were significant except depressive symptoms and perceived stress for Active Choices. The pattern was the same for Active Choices in Years 3 and 4. For ALED, all changes were significant in Years 3 and 4 except that the reduction in depressive symptoms approached statistical significance only in Year-3 carry-forward analyses. The effect sizes (analyses with available data) were somewhat larger for Active Choices in Years 3 and 4 than Year 1, but largest in Year 1 for ALED. The effect sizes based on carry-forward analyses were similar in magnitude across years and expectedly smaller than the analyses conducted with available data. | | |  | | Year 1 | Year 3 | Year 4 | |
|---|
| | Available data | Carry forward | Available data | Carry forward | Available data | Carry forward | |
|---|
| Moderate and vigorous PA, d | 0.62 | 0.43 | 0.66 | 0.36 | 0.75 | 0.43 | | |  Pretest h/wk | 2.89(0.77) | 2.96(0.42) | 2.89(0.44) | 2.73(0.44) | 2.31(0.38) | 2.20(0.35) | | | Posttest h/wk | 5.33(0.77) | 4.69(0.42) | 5.65(0.48) | 4.22(0.44) | 5.13(0.42) | 3.90(0.35) | | | t with 4 df (p value) | 8.83(<0.001) | 5.87(0.004) | 6.20(0.003) | 8.77(<0.001) | 7.22(0.002) | 6.37(0.003) | | | All PA, d | 0.55 | 0.41 | 0.60 | 0.32 | 0.63 | 0.37 | | | Pretest h/wk | 7.97(1.45) | 7.73(0.75) | 8.16(0.93) | 8.09(1.01) | 7.43(0.80) | 7.14(0.73) | | | Posttest h/wk | 12.13(1.45) | 10.75(0.75) | 12.98(0.99) | 10.70(1.01) | 12.14(0.87) | 9.99(0.73) | | | t with 4 df (p value) | 7.93(0.001) | 7.87(0.001) | 5.52(0.005) | 7.95(0.001) | 5.70(0.005) | 5.05(0.007) | | | Depressive symptoms, d | −0.04 | −0.03 | −0.08 | −0.03 | 0.02 | 0.02 | | | Pretest score (range: 0 to 30) | 7.69(0.85) | 6.82(0.48) | 6.74(0.49) | 6.89(0.51) | 6.15(0.50) | 6.13(0.49) | | | Posttest score (range: 0 to 30) | 7.52(0.85) | 6.69(0.48) | 6.33(0.51) | 6.74(0.51) | 6.26(0.53) | 6.23(0.49) | | | t with 4 df(p value) | −0.73(0.51) | −0.79(0.47) | 1.63(0.18) | 0.97(0.39) | 0.40(0.71) | 0.62(0.57) | | | Perceived stress, d | −0.01 | 0.00 | −0.07 | −0.03 | 0.00 | 0.03 | | | Pretest score (range: 0 to 16) | 4.93(0.56) | 4.72(0.33) | 4.80(0.32) | 4.76(0.32) | 4.20(0.35) | 4.16(0.33) | | | Posttest score (range: 0 to 16) | 4.91(0.56) | 4.72(0.33) | 4.56(0.34) | 4.68(0.32) | 4.19(0.37) | 4.25(0.33) | | | t with 4 df (p value) | −0.10(0.93) | 0.00(0.997) | 1.14(0.32) | 0.72(0.51) | 0.01(0.99) | 0.50(0.64) | | | Satisfaction with body appearance, d | 0.34 | 0.23 | 0.44 | 0.27 | 0.38 | 0.21 | | | Pretest score (range: −9 to +9) | −2.76(0.87) | −2.76(0.51) | −2.75(0.46) | −2.92(0.46) | −2.90(0.58) | −3.00(0.59) | | | Posttest score (range: −9 to +9) | −1.04(0.87) | −1.60(0.51) | −0.23(0.49) | −1.45(0.46) | −0.93(0.61) | −1.92(0.59) | | | t with 4 df (p value) | 5.37(0.006) | 3.67(0.02) | 8.27(0.001) | 7.67(0.001) | 4.55(0.01) | 3.67(0.02) | | | Satisfaction with body function, d | 0.49 | 0.36 | 0.56 | 0.33 | 0.55 | 0.29 | | | Pretest score (range: −18 to +18) | −1.90(1.64) | −1.03(0.95) | −1.41(0.86) | −1.64(0.84) | −1.24(0.88) | −1.60(0.83) | | | Posttest score (range: −18 to +18) | 2.74(1.64) | 2.32(0.95) | 4.48(0.91) | 1.81(0.84) | 4.17(0.93) | 1.28(0.83) | | | t with 4 df (p value) | 5.49(0.005) | 5.38(0.006) | 5.69(0.005) | 4.22(0.01) | 6.18(0.004) | 5.45(0.006) | | | BMI, d | −0.05 | −0.03 | −0.07 | −0.04 | −0.06 | −0.03 | | | Pretest kg/m2 | 32.74(1.61) | 31.27(1.27) | 29.90(0.67) | 29.96(0.68) | 30.44(0.69) | 30.46(0.68) | | | Posttest kg/m2 | 32.41(1.61) | 31.05(1.27) | 29.39(0.68) | 29.66(0.68) | 30.00(0.69) | 30.22(0.68) | | | t with 4 df(p value) | −2.85(0.046) | −2.78(0.0498) | 3.70(0.02) | 3.33(0.03) | 4.71(0.009) | 4.33(0.01) | | | | |
The percentages of participants meeting CDC–American College of Sports Medicine (ACSM) recommendations are shown in Table 4. For both programs, these percentages increased significantly from pretest to posttest in all years. For ALED Year-4 analyses, increases were maintained from 12 weeks to 20 weeks in analyses with available data, but there was a significant reduction in carry-forward analyses. | | | | | Year 1 | Year 3 | Year 4 | |
|---|
| | Available data | Carry forward | Available data | Carry Forward | Available data | Carry forward | |
|---|
| AC: Met PA recommendations | | | | | | | | | Pretest, % | 9.92 | 8.43 | 9.76 | 8.93 | 11.19 | 9.51 | | | Posttest, % | 24.07 | 18.03 | 37.63 | 25.41 | 35.64 | 22.28 | | | F (1, 4) (p value) | 63.46(0.001) | 76.31(0.001) | 189.26(<0.001) | 113.39(<0.001) | 161.26(<0.001) | 88.99(<0.001) | | | ALED: Met PA recommendations | | | | | | | | | Pretest, % | 15.36 | 15.51 | 16.55 | 16.21 | 15.60 | 15.00 | | | 12 wks, % | | | | | 49.09 | 47.67 | | | 20 wks, % | 42.14 | 35.50 | 48.68 | 40.28 | 44.14 | 33.11 | | | F (p value) | 180.04(<0.001) | 200.35(<0.001) | 318.23(<0.001) | 274.66(0<0.001) | 201.95(<0.001) | 183.26(<0.001) | | | | |
Discussion Translational studies of evidence-based physical activity programs are needed to determine the feasibility of implementing physical activity programs in a variety of real-world settings with diverse populations over time, and once implemented, to determine if they produce comparable results.1, 3, 43, 44 AFL is an excellent example of a translational study and demonstrated the successful translation of two programs into a variety of real-world settings with a diverse older adult population. AFL also provided an opportunity to examine program effects after an adaptation reducing the ALED program length from 20 to 12 weeks. AFL represents a marked departure from the usual prevention research settings and populations.3, 7 Study Limitations Several study limitations should be considered. The primary study limitation is the absence of a control group. Although it is likely that effect sizes would have been somewhat attenuated if a control group had been included, we believe the potential for bias is more than offset by the advantages of studying the translation of these programs to community settings. There are several other limitations. To minimize response burden self-reported data were used. CHAMPS, the primary outcome measure, correlates moderately with objective measures of physical activity, physical functioning, and quality of life.30, 31 One of the original efficacy studies showed similar increases in subjective and objective physical activity measures.14 Self-reported behaviors may be influenced by social desirability; however, if social desirability were responsible for these outcomes, one might expect similar effect sizes across all the self-reported outcomes. Effect sizes varied by outcome and were consistent with the literature. An additional limitation is the problem of high nonresponse among participants at posttest, particularly for Active Choices, and the fact that the nonresponders were arguably at greater risk and higher need than responders. However, the carry-forward analyses indicated that even if these participants did not change at all, the public health and clinical benefits of the programs would still be substantial. Additionally, maintenance of behavior change over the long term has not been reported. Collecting Year-4 data at 12 weeks (immediately post) and 20 weeks after the shortened ALED program provided some preliminary indications of the maintenance of behavior. Six-month follow-up data are being collected at selected sites, but these data are not yet available. It is also not known whether smaller community-based organizations could successfully implement the models. Finally, a cost-effectiveness analysis was not conducted. This type of analysis is strongly recommended in future studies so as to better inform policy decisions. Despite these limitations, a number of key findings and their implications merit further discussion. Robust Effects in a Diverse Study Sample Community participants of several different ethnicities and educational and income levels successfully completed the program. Moreover, the majority of participants across settings were from populations that are typically understudied in the physical activity literature. Participants were less advantaged, more ethnically diverse, and at greater risk than in the original efficacy studies,14, 15, 16, 20, 21, 25, 45 and similar to a community-based Active Choices replication,46 but with a higher representation of African Americans and a lower representation of Latinos. In efficacy studies of Active Choices,15, 16, 25, 47 the percentage of ethnic minorities ranged from 5% to 14%, and effect sizes for physical activity–related outcomes were computed in the range of d=0.10 to 0.63, with generally larger effects for women and generally smaller effects for treadmill time. In efficacy studies of ALED,20, 21 effect sizes for physical activity–related outcomes were computed in the range of d=0.12 to 0.56. Thus, effect sizes in AFL were quite comparable to those reported in the original efficacy-based studies, despite the more diverse samples, and they were fairly consistent across years and sites, indicating robust and clinically meaningful changes. Also parallel to AFL findings, effect sizes for BMI and psychosocial measures were more variable and generally smaller in the efficacy studies. An interesting divergence in the programs was that for Active Choices, effects were typically larger in the later 2 years (for analyses with available data) whereas for ALED, the pattern was just the opposite. Several key changes in the participant population over time were noted for ALED. Participants in later years were older, more active, and less educated with lower income levels and higher rates of diabetes and hypertension. This change in population characteristics, as well as the adaptation of the program from 20 to 12 weeks in Year 4, may account for the smaller physical activity effect sizes in later years. Because the population was more active initially, there may have been less room for change. It is important to underscore, however, that the effect sizes remained medium in size and clinically meaningful. Another difference between programs was that reductions in depressive symptoms and perceived stress were found for ALED but not Active Choices. The differing populations enrolled in the two programs makes it difficult to speculate on why these differences were observed, and it is important to note that baseline levels of these symptoms were very low and pre–post effect sizes very small. Nonetheless, the possibility that individuals with higher levels of stress and depressive symptoms may respond favorably to group-based formats is an interesting finding that merits further study in more carefully controlled RCTs. Implications for Practice The robust outcomes seen in this study give community programs some confidence that if they adopt Active Choices or ALED they can achieve comparable results. Of course, external validity relies on inductive, not deductive, methods and tests of generalization can never cover all the potential combinations of populations, settings and treatment adaptations.44 When a wide variety of these combinations are tested, however, practitioners can reduce uncertainty about the likely benefits of the programs in their own settings and populations. The AFL initiative provided an infrastructure that allowed the lead organizations to strategize for recruitment, train staff in the program approaches, maintain quality control through attention to the essential elements of the interventions, share information about legitimate adaptations within the overall models, obtain feedback from the process and outcome evaluation to share with stakeholders, and plan for sustainability. This context and approach, which included a great deal of interaction and communication between community sites and researchers, is consistent with core components of effective knowledge translation.48, 49, 50 We believe that the success of AFL was promoted, in part, because research activities were tailored to the needs of the sites, and the sites helped to guide the research activities and define needed adaptations in an interactive manner.48, 50 These and other related issues will be presented in forthcoming papers that focus on the high implementation fidelity in AFL along with challenges and adaptations. By partnering with trusted organizations that were embedded in the communities, AFL also succeeded in recruiting and retaining groups that do not ordinarily consent to participate in research. Furthermore, the profiles of participants who were less likely to complete the program allow for efforts to retain participants by targeting those most at greatest risk. Conclusion Better studies of the translation of research to practice are required in order to assess the external validity of evidence-based interventions.43 The AFL initiative demonstrates that a range of community-based organizations can not only implement evidence-based programs, but that substantial numbers of diverse older adults can receive physical and mental health benefits by participating in them. The Active for Life initiative is funded by the Robert Wood Johnson Foundation. The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Robert Wood Johnson Foundation or other institutions affiliated with the authors. We gratefully acknowledge the many participants who took part in the Active Choices and ALED programs and evaluation. We also acknowledge the involvement and significant contribution of the following program managers, facilitators/health educators, and staff at each of the lead organizations (listed alphabetically by organization name and alphabetically by person). Berkeley Public Health Department: Kate Clayton, Jennifer Divers, Erica Frechman, Maria Guerrero, Justine Kaplan, Elise Krumholz, Madhuri Sudan, Kristin Tehrani; Blue Shield of California: Deborah Kosmont, Carrie Krutzner, Mindy Morgen, Juan Rivas; Church Health Center: Teresa Cutts, Mia Earl-Clemmons, Doris Grider, Sterling McNeal, Michelle D. Rice; Council on Aging of Southwestern Ohio: Susan Binkert, Kelly Lake, Alisa Phillips, Karen Schwamberger, Roberta Wickerham; FirstHealth of the Carolinas: Lisa G. Hartsock, Melissa Watford; Greater Detroit Area Health Council: Vernice Davis Anthony, Esther Bishop-Files, Paul Bridgewater, Karen Calhoun, Marvin Cato, Mary Cocanougher, Maude Freeman, Pastor Michael C. Nabors, Alberta Smith-Plump, Debbie Stefanides, Theresa Taylor, John B. Waller, Jr., Shirley Williams, Carolyn Wilson-Hall; Human Kinetics, Inc.: H. Michelle Maloney; Jewish Council for the Aging of Greater Washington: Carol Ames, Christine Bruchac, Beverly Bryant, Sharlene P. Hirsch, Dorothy Morgan-Quelch; The OASIS Institute: Fabiana Cheistwer, Marcia Kerz, Cindy Merrins, Cheryl Roberts Oliver, Shirley Pogue, Brenda Schmachtenberger, Tracy Slate, Gail Weisberg; San Mateo County Health Department: Edith Cabuslay, Doris Y. Estremera-Rohleder, Isabel R. Guerrero, Katya Henriquez, Gabriela Lemus, Martha Milk, Scharlette Parker, Sam Stebbins, Cristina Ugaitafa, Jaslin Yu; Texas A&M Health Science Center: Vanessa Batt, Lisa Groce, Amanda Laughlin, Brigid Sanner, Amber Schickedanz, William Smith, University of South Carolina: Barbara E. Ainsworth, Corrie Barnett, Peter Bense, Julie Freelove-Charton, Nancy Chase, Shanikwha Jones, Samira Khan, Diana Lattimore, Angela Merlo, Russell Pate, Patricia A. Sharpe, Dennis Shepard, Winifred W. Thompson, M. Renée Umstattd; and YMCA of Metropolitan Chicago: Jan Arnold, Martha Beard, Annemarie DeFazio, Mary Ganzel, Karen Keenley, Jassen Lanfair, Fran Lewickyj. We recognize Robin Mockenhaupt from the Robert Wood Johnson Foundation for her pioneering role in Active for Life. We also thank Peter Hannan from the University of Minnesota for his statistical consultation on this project. We thank the National Advisory Committee for their valuable contributions to Active for Life. Finally, we thank the coalitions, partnering organizations, and advisory boards associated with each of the lead organizations for their meaningful contributions and support of the program. No financial disclosures were reported by the authors of this paper. Author order beyond the first 3 is alphabetical and does not reflect contribution. References 1. 1Glasgow RE, Lichtenstein E, Marcus AC. Why don't we see more translation of health promotion research to practice? (Rethinking the efficacy-to-effectiveness transition). Am J Public Health. 2003;93:1261–1267. MEDLINE |
CrossRef
2. 2Dunn AL, Bettinghaus EP. Diffusion and dissemination for increasing physical activity in world populations. Am J Prev Med. 2006;31(4S):S94–S96. MEDLINE 3. 3Glasgow RE, Emmons KM. How can we increase translation of research into practice? (Types of evidence needed). Annu Rev Public Health. 2007;28:413–433. MEDLINE |
CrossRef
4. 4Brownson RC, Kreuter MW, Arrington BA, True WR. Translating scientific discoveries into public health action: how can schools of public health move us forward?. Public Health Rep. 2006;121:97–103. MEDLINE 5. 5Ory MG, Evashwick CJ, Glasgow RE, Sharkey JR. Pushing the boundaries of evidence-based research: enhancing the application and sustainability of health promotion programs in diverse populations. In: Browning CJ, Thomas SA editor. Behavioral change: an evidence-based handbook for social and public health. Edinburgh: Elsevier Churchill; 2005;p. 267–293. 6. 6Brownson RC, Ballew P, Dieffenderfer B, et al. Evidence-based interventions to promote physical activity: what contributes to dissemination by state health departments. Am J Prev Med. 2007;33(1S):S66–S73. Abstract | Full Text |
Full-Text PDF (216 KB)
|
CrossRef
7. 7Green LW. From research to “best practices” in other settings and populations. Am J Health Behav. 2001;25:165–178. MEDLINE 8. 8Estabrooks PA, Glasgow RE. Translating effective clinic-based physical activity interventions into practice. Am J Prev Med. 2006;31(4S):S45–S56. MEDLINE 9. 9Task Force on Community Preventive Services. Recommendations to increase physical activity in communities. Am J Prev Med. 2002;22(4S):S67–S72. 10. 10Ory MG, Jordan PJ, Bazzarre T. The Behavior Change Consortium: setting the stage for a new century of health behavior-change research. Health Educ Res. 2002;17:500–511. MEDLINE |
CrossRef
11. 11Eakin EG, Lawler SP, Vandelanotte C, Owen N. Telephone interventions for physical activity and dietary behavior change: a systematic review. Am J Prev Med. 2007;32:419–434. Abstract | Full Text |
Full-Text PDF (153 KB)
|
CrossRef
12. 12Yancey AK, Ory MG, Davis SM. Dissemination of physical activity promotion interventions in underserved populations. Am J Prev Med. 2006;31(4S):S82–S91. MEDLINE 13. 13Wilcox S, Dowda M, Griffin SF, et al. Results of the first year of Active for Life: translation of 2 evidence-based physical activity programs for older adults into community settings. Am J Public Health. 2006;96:1201–1209.
CrossRef
14. 14King AC, Haskell WL, Young DR, Oka RK, Stefanick ML. Long-term effects of varying intensities and formats of physical activity on participation rates, fitness, and lipoproteins in men and women aged 50 to 65 years. Circulation. 1995;91:2596–2604. MEDLINE 15. 15King AC, Baumann K, O'Sullivan P, Wilcox S, Castro C. Effects of moderate-intensity exercise on physiological, behavioral, and emotional responses to family caregiving: a randomized controlled trial. J Gerontol. 2002;57:M26–M36. 16. 16King AC, Pruitt LA, Phillips W, Oka R, Rodenburg A, Haskell WL. Comparative effects of two physical activity programs on measured and perceived physical functioning and other health-related quality of life outcomes in older adults. J Gerontol. 2000;55:M74–M83. 17. 17Stewart AL, Mills KM, Sepsis PG, et al. Evaluation of CHAMPS, a physical activity promotion program for older adults. Ann Behav Med. 1997;19:353–361. MEDLINE |
CrossRef
18. 18Prochaska JO, DiClemente CC, Norcross JC. In search of how people change (Applications to addictive behaviors). Am Psychol. 1992;47:1102–1114.
CrossRef
19. 19Bandura A. Social foundations of thought and action: a social cognitive theory. Englewood Cliffs NJ: Prentice-Hall; 1986;. 20. 20Dunn AL, Marcus BH, Kampert JB, Garcia ME, Kohl HW, Blair SN. Reduction in cardiovascular disease risk factors: 6-month results from Project Active. Prev Med. 1997;26:883–892. MEDLINE |
CrossRef
21. 21Dunn AL, Marcus BH, Kampert JB, Garcia ME, Kohl HW, Blair SN. Comparison of lifestyle and structured interventions to increase physical activity and cardiorespiratory fitness: a randomized trial. JAMA. 1999;281:327–334. MEDLINE |
CrossRef
22. 22Thomas S, Reading J, Shephard RJ. Revision of the Physical Activity Readiness Questionnaire (PAR-Q). Can J Sport Sci. 1992;17:338–345. MEDLINE 23. 23Ory M, Resnick B, Jordan PJ, et al. Screening, safety, and adverse events in physical activity interventions: collaborative experiences from the behavior change consortium. Ann Behav Med. 2005;29(Suppl):20–28.
CrossRef
24. 24King AC, Haskell WL, Taylor CB, Kraemer HC, DeBusk RF. Group- vs home-based exercise training in healthy older men and women (A community-based clinical trial). JAMA. 1991;266:1535–1542. MEDLINE 25. 25King AC, Oman RF, Brassington GS, Bliwise DL, Haskell WL. Moderate-intensity exercise and self-rated quality of sleep in older adults (A randomized controlled trial). JAMA. 1997;277:32–37. MEDLINE 26. 26King AC, Friedman R, Marcus B, et al. Ongoing physical activity advice by humans versus computers: the Community Health Advice by Telephone (CHAT) trial. Health Psychol. 2007;26:718–727.
CrossRef
27. 27Estabrooks P, Gyurcsik NC. Evaluating the impact of behavioral interventions that target physical activity: issues of generalizability and public health. Psychol Sport Exerc. 2003;4:41–55. 28. 28Expert Panel. Executive summary of the clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults. Arch Intern Med. 1998;158:1855–1867. MEDLINE |
CrossRef
29. 29CDC. Behavioral Risk Factor Surveillance System, 2005 Survey Questions. www.cdc.gov/brfss/questionnaires/english.htm. 30. 30Stewart AL, Mills KM, Kingz AC, Haskell WL, Gillis , Ritter PL. CHAMPS Physical Activity Questionnaire for Older Adults: outcomes for interventions. Med Sci Sports Exerc. 2001;33:1126–1141. MEDLINE 31. 31Harada ND, Chiu V, King AC, Stewart AL. An evaluation of three self-report physical activity instruments for older adults. Med Sci Sports Exerc. 2001;33:962–970. MEDLINE |
CrossRef
32. 32Stewart AL. Community-based physical activity programs for adults age 50 and older. J Aging Phys Act. 2001;9(Suppl):S71–S91. 33. 33Stewart AL, Verboncoeur CJ, McLellan BY, et al. Physical activity outcomes of CHAMPS II: a physical activity promotion program for older adults. J Gerontol. 2001;56:M465–M470. 34. 34Pate RR, Pratt M, Blair SN, et al. Physical activity and public health (A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine). JAMA. 1995;273:402–407. MEDLINE 35. 35Kohout FJ, Berkman LF, Evans DA, Cornoni-Huntley J. Two shorter forms of the CES-D (Center for Epidemiological Studies Depression) depression symptoms index. J Aging Health. 1993;5:179–193. MEDLINE |
CrossRef
36. 36Irwin M, Artin KH, Oxman MN. Screening for depression in the older adult: criterion validity of the 10-item Center for Epidemiological Studies Depression Scale (CES-D). Arch Intern Med. 1999;159:1701–1704. MEDLINE |
CrossRef
37. 37Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. App Psych Measure. 1977;1:385–401. 38. 38Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav. 1983;24:385–396. MEDLINE 39. 39Pbert L, Doerfler LA, DeCosimo D. An evaluation of the perceived stress scale in two clinical populations. J Psychopathol Behav Assess. 1992;14:363–375. 40. 40Reboussin BA, Rejeski WJ, Martin KA, et al. Correlates of satisfaction with body function and body appearance in middle- and older aged adults: The Activity Counseling Trial (ACT). Psychol Health. 2000;15:239–254. 41. 41Wolf FM. Meta-analysis (Quantitative methods for research synthesis). Newbury Park CA: Sage; 1986;. 42. 42Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed.. Hillsdale NJ: Lawrence Erlbaum; 1988;. 43. 43Steckler A, McLeroy KR. The importance of external validity. Am J Public Health. 2008;98:9–10.
CrossRef
44. 44Leviton LC. External validity. In: Smelser NJ, Baltes PB editor. The international encyclopedia of the social and behavioral sciences (Vol 8, Logic of inquiry and research design). Elsevier; 2001;p. 5195–5200. 45. 45Smolander J, Blair SN, Kohl HW. Work ability, physical activity, and cardiorespiratory fitness: 2-year results from Project Active. J Occup Environ Med. 2000;42:906–910. MEDLINE |
CrossRef
46. 46Hooker SP, Seavey W, Weidmer CE, et al. The California Active Aging Community Grant Program: translating science into practice to promote physical activity in older adults. Ann Behav Med. 2005;29:155–165. MEDLINE |
CrossRef
47. 47King AC, Taylor CB, Haskell WL. Effects of differing intensities and formats of 12 months of exercise training on psychological outcomes in older adults. Health Psychol. 1993;12:292–300. MEDLINE |
CrossRef
48. 48Bowen S, Martens P. Demystifying knowledge translation: learning from the community. J Health Serv Res Policy. 2005;10:203–211. MEDLINE |
CrossRef
49. 49Graham ID, Logan J, Harrison MB, et al. Lost in knowledge translation: time for a map?. J Contin Educ Health Prof. 2006;26:13–24. MEDLINE |
CrossRef
50. 50Jacobson N, Butterill D, Goering P. Development of a framework for knowledge translation: understanding user context. J Health Serv Res Policy. 2003;8:94–99. MEDLINE |
CrossRef
a University of South Carolina (USC), Department of Exercise Science, Columbia, South Carolina b USC Prevention Research Center, Columbia, South Carolina c Robert Wood Johnson Foundation, Princeton, New Jersey d Church Health Center, Memphis, Tennessee e Department of Family Medicine and Public Health Sciences, Wayne State University, Detroit, Michigan f The Cooper Institute, Dallas, Texas g Stanford University School of Medicine, Stanford Prevention Research Center, Stanford, California h Department of Health Research & Policy, Stanford, California i Texas A&M Health Science Center, Department of Social and Behavioral Health, College Station, Texas j Klein-Buendel, Inc., Golden, Colorado k Clemson University, Clemson, South Carolina l University of Illinois at Urbana-Champaign, The Center on Health, Aging, and Disability, Champaign, Illinois m The OASIS Institute, St. Louis, Missouri n Council on Aging of Southwestern Ohio, Cincinnati, Ohio  Address correspondence and reprint requests to: Sara Wilcox, PhD, Department of Exercise Science, Arnold School of Public Health, University of South Carolina, 921 Assembly Street, PHRC 3rd Floor, Columbia SC 29208
The full text of this article is available via AJPM Online at www.ajpm-online.net; 1 unit of Category-1 CME credit is also available, with details on the website. PII: S0749-3797(08)00605-3 doi:10.1016/j.amepre.2008.07.001 © 2008 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT