摘要:Objectives. To evaluate the impact of a faith-based intervention to promote physical activity in Latinas. Methods. We randomized 16 churches in San Diego County, California, to a physical activity intervention or cancer screening comparison condition (n = 436). The intervention followed an ecological framework and involved promotoras . We examined 12-month intervention effects, including accelerometer-based moderate-to-vigorous physical activity (MVPA; primary outcome) and secondary outcomes. We conducted the study from 2010 to 2016. Results. Mixed effects analyses showed significant increases in accelerometer-based MVPA (effect size = 0.25) and self-report leisure-time MVPA (effect size = 0.38) among Latinas in the intervention versus comparison condition. Participants in the intervention condition had about 66% higher odds of meeting the 2008 Physical Activity Guidelines, had reduced body mass index (effect size = 0.23), and used more behavioral strategies for engaging in physical activity (effect size = 0.42). Program attendance was associated with increased self-reported leisure-time MVPA and the number of motivational interviewing calls was associated with meeting the 2008 Physical Activity Guidelines. Conclusions. A faith-based intervention was effective in increasing MVPA and decreasing body mass index among participants. Process analyses showed the value of program attendance and motivational interviewing calls. Fewer than half of Latina adults meet the aerobic 2008 Physical Activity Guidelines (2008 PAG), 1 and Latinas engage in less leisure-time physical activity per week than Latino men. 2 This low level of physical activity among Latinas may explain, in part, their high rates of cardiovascular disease and other chronic diseases such as cancer. 3 Health promotion programs are needed to improve the health profile of Latinos, the fastest-growing ethnic group in the United States. Ecological frameworks, which recognize the multiple factors that influence behavior, can be effective for designing comprehensive programs to promote physical activity. For example, individual-level factors such as motivation and self-efficacy for physical activity are important determinants of physical activity. 4 Individual-level strategies for promoting activity may include pedometers to self-monitor activity or counseling to help set goals and identify reinforcers. At the interpersonal level, social support helps individuals start and maintain physical activity 5,6 ; a program’s interpersonal-level components might include peer-led activities that promote social support for activity. Environmental correlates of physical activity include access to parks and destinations near the home. 4,7,8 Environmental-level strategies for promoting activity may include neighborhood improvement projects (e.g., enhanced sidewalks and crosswalks, park renovations). A key principle of ecological frameworks is that interventions that target multiple levels of influence (e.g., individual, interpersonal, organizational, and environmental) should be more effective in changing behavior than those that target only 1 level. 9 To our knowledge, no published multilevel interventions have targeted individual factors as well as the social and built environment to promote physical activity among Latino communities. An effective approach to promote physical activity among Latinos is involving community health workers or promotoras . 10–12 A number of factors may explain why promotora -led programs are effective. Promotora -led group exercise classes may facilitate social support, an important correlate of physical activity. 11 Promotoras may be effective in increasing physical activity because they can be role models for activity and help individuals understand health education messages, tailoring information to make it more relevant. 13 A promising setting for promoting physical activity among Latinas is through churches. Faith-based programs are likely to have large reach, credibility, and sustainability in Latino communities. According to a Pew Research Center report, 55% of Latinos identified as Catholic, and almost 40% of Latino Catholics reported attending church at least once a week. 14 Preliminary evidence suggests that faith-based health promotion programs can positively affect health behaviors such as physical activity among Latinos. 11,15–17 Almost all published studies have assessed physical activity through self-report, had short intervention periods (12 weeks or less), and targeted primarily African American women. 18 Research findings on physical activity among Latinas in a faith-based setting are limited to one 16-week intervention that assessed only knowledge of the benefits of physical activity. 15 These models and lines of evidence justify a rigorous evaluation of a faith-based multilevel physical activity intervention in Latino communities 18 targeting individual, interpersonal, organizational, and environmental influences on Latinas’ physical activity. 19,20 We tested the hypothesis that a multilevel physical activity intervention will increase physical activity among Latinas compared with the comparison condition. We report outcomes and process–outcome analyses from the first year of the intervention.