MELISSA BOPP, P H D , SARA WILCOX, P H D , MARILYN LAKEN, P H D , R N , & LOTTIE MCCLORIN, MS Abstract: The potential benefit of physical activity (PA) programs delivered through churches is largely unexamined. This study examined availability of PA programs, interpersonal support for PA, and PA participation in African-American churches. Individuals from a random sample of 20 churches in South Carolina participated in a telephone survey (N-571). Forty two percent of respondents reported PA programs at their churches. Walking programs (20%), aerobics (22%) or a combination of both (20%) were most common. Respondents who reported having these programs were more likely to meet PA recommendations than those who did not (p-0.05). Larger churches were more likely to offer PA programs (p-0.02) than small or medium sized churches. Only 247o of respondents had spoken with the health director at their church about participating in a PA program, and only 25% and 337o had ever spoken with another church member about a PA program or were encouraged to join a PA program, respectively. Individuals with more interpersonal support from other church members for PA were significantly more likely to meet PA recommendations (p=0.01). This study indicates that program and interpersonal supports within African American churches may offer a venue for increasing PA among members. Key Words: Physical Activity, Faith-Based, Social Support, Ethnic Minority

PHYSICAL ACTIVITY PARTICIPATION IN AFRICAN AMERICAN CHURCHES

R

egular physical activity (PA) participation has well-documented physical and mental health benefits for people of all ages(Physical Activity Guidelines Advisory Committee, 2008). Despite these known benefits, a large portion of the population remain underactive or sedentary, especially ethnic minorities. Caucasian adults are more likely to meet current PA guidelines (51.4% met recommendations) (Haskell et al., 2007) when compared with African Americans (40.4%) (Centers for Disease Control Prevention, 2009). These observed differences in PA participation between Caucasians and African Americans remain when educational levels and household Melissa Bopp, PhD, is an Assistant Professor in the Department ofKinesiology, Community Health Institute at Kansas State University. Sara Wilcox, PhD, is an Associate Professor at the University of South Carolina, Arnold School of Public Health Department of Exercise Science. Marilyn Laken, PhD, RN, is Professor and Director, Office of Special Initiatives, Medical University of South Carolina. Lottie McClorin, MS, Project Manager, University of South Carolina Arnold School of Public Health Department of Exercise Science.

Journal of Cultural Diversity • Vol. 16, No. 1

incomes are similar (Seefeldt, Malina, & Clark, 2002). It is critical to find effective ways to increase PA in African Americans, as this group experiences disproportionate disease burden associated with obesity and^ inactivity that PA could positively impact. For example, cardiovascular disease is the leading cause of death for African American males and females, and African Americans have higher incidence rates of cardiovascular disease when compared wifh Caucasians of similar ages and socioeconomic statuses (American Heart Association, 2008). African Americans also have almost twice the risk of both first-time stroke and incidence of diabetes mellitus compared with Caucasians (American Heart Association, 2008). Research has identified a number of demographic, psychological, social, and environmental correlates of PA in African Americans. For African American women, regular PA participation has been positively associated with higher education levels, being married, higher perceived health status, having social support for PA, having role models for PA, lower social strain, attempting weight loss, perceiving benefits of PA, and a favorable physical environment for PA (Ainsworth, Wilcox, Thompson, Richter, & Henderson, 2003; Eyler et al., 2002). Among African American men, regular participation in PA is positively correlated with higher

Spring 2009

education and income levels, having a physician discuss PA, attempting to lose weight, younger age, and living in a physical environment conducive to PA (Crespo, Ainsworth, Keteyian, Heath, & Smit, 1999; Macera, Croft, Brown, Ferguson, & Lane, 1995; Young, Miller, Wilder, Yanek, & Becker, 1998). Previous PA interventions in African Americans have used a variety of approaches: family oriented, community oriented, church-based, or home-based with telephone or mail components (Banks-Wallace & Conn, 2002; Taylor, Baranowski, & Young, 1998). Though most of these interventions considered PA to be a variable of interest, weight loss or other cardiovascular disease risk factors have been the main dependent variable in many interventions, particularly in faith-based interventions. Although many studies have targeted their interventions exclusively for African Americans, few include culturally tailored program elements. Faith-based health promotion programs have been shown to be successful, especially in the African American community where a large portion of adults are members of a church or report regular attendance at church (Baskin, Resnicow, & Campbell, 2001; DeHaven, Hunter, Wilder, Walton, & Berry, 2004). Health promotion programs that address health disparities within a church's congregation can draw strong support from church leaders and members by promoting a multi-dimensional, holistic image of health (Lasater, Carleton, & Wells, 1991). Interventions within the church offer a unique source of social support not found in other settings, which has been found to be a positive correlate of PA participation in ethnic minority populations (Eyler et al., 1999). Churches present members with the opportunity for fellowship with individuals who have similar values and beliefs and a strong support network for activities, including health-related programs. The degree to which Pastors and church lay leaders are involved in a health promotion program affect the level of success (Peterson, Atwood, & Yates, 2002). For example, the Black Churches United for Better Health Project (Campbell et al., 1999) focused on dietary outcomes, including increased fruit and vegetable consumption, and attempted to institutionalize the program by forming coalitions within the church and community, encouraging pastor support as well as incorporating educational materials on bulletin boards and in church bulletin inserts and was successful in changing behavior. The Los Angeles Mammography Promotion Program in Churches (Markens, Fox, Taub, & Gilbert, 2002) found that church leader participation was an important determinant of the success of the intervention. Ensuring that the church's philosophy and organization are consistent with the intervention design can also impact the success of the research-church partnership (Baskin et al., 2001; Lasater, Wells, Carleton, & Elder, 1986). Increasing PA participation has been the focus of some faith-based interventions, with varying levels of involvement from the church. Examples include Project Joy (Yanek, Becker, Moy, Gittelsohn, & Koffman, 2001), and^ Healthy Body/ Healthy Spirit (Resnicow et al., 2005). Project Joy saw improvements in PA behavior, though not significant, while Healthy Body/Healthy Spirit was successful in

Journal of Cultural Diversity • Vol. 16, No. 1

increasing PA participation (Resnicow et al, 2005; Yanek et al., 2001). Other interventions have used PA as a component within the intervention, though PA participation was not a major outcome variable (Kumanyika & Charleston, 1992; McNabb, Quinn, Kerver, Cook, & Karrison, 1997; Oexmann, Ascanio, & Egan, 2001). There were two primary goals of this study. The first was to determine the degree of church involvement in PA promotion in a sample of African American churches in South Carolina, and the relationship to current PA participation in members. A second purpose was to examine whether support for PA within these churches was related to PA participation among members. METHODS

The Health-e-AME Physical-e-Fit program was developed to specifically target PA in African Methodist Episcopal Churches (AME) in South Carolina. This paper reports on baseline findings, prior to the implementation of any intervention. The intervention and outcomes are described in extensive detail elsewhere (Bopp et al, in press; Wilcox, Laken, Anderson et al., 2007; Wilcox, Laken, Bopp et al, 2007). Brief Description of the Intervention

The intervention was based on the Social Ecological Model (McLeroy, Bibeau, Steckler, & Glanz, 1988) and the Transtheoretical Model (Prochaska & DiClemente, 1983). All materials were developed with input from church leaders and members and were culturally and spiritually tailored, incorporating scripture and elements of religiosity. Churches interested in receiving the program sent a Health Director to be trained to deliver the intervention components. The Health Director could choose components of the program to deliver to their church, allowing them to design a tailored program based on the needs and preferences of their congregation. Program components included educational activities as well as action-oriented activities (chair aerobics, walking programs, behavioral and skill based classes, and aerobics set to gospel music). Participants & Procedure

Participants were adult members of African Methodist Episcopal (AME) congregations in South Carolina who agreed to participate in a baseline telephone survey (Wilcox, Laken, Anderson et al., 2007). The sampling frame for the survey consisted of all AME churches in SC stratified by conference (the AME church is organized into six geographical conferences). Churches were the primary sampling unit, and church members within selected churches were the secondary sampling unit. The only inclusion criterion for the churches was that the Pastors were willing to submit a complete roster of adult church members to the research staff (after announcing that members could remove their name from the list, if desired). The church rosters were transferred to the Institute for Public Service and Policy Research Survey Research Laboratory at the University of South Carolina (USC). All surveys were conducted using Computer-Assisted Telephone Interviewing software (CÍ3-CATI). The field period for the basehne survey was May to September 2003. To be eligible for the survey, participants had to be at least 18 years of age, report that Spring 2009

they attended church services at least twice per month and provide verbal informed consent. The project was reviewed and approved by the Institutional Review Boards at the Medical University of South Carolina (MUSC) and USC and a Planning Committee comprised of AME, MUSC, and USC representatives. Measures Sociodemographics._The survey assessed respondents' age, education, martial status, income, height, weight, and gender with questions from the Behavioral Risk Factor Surveillance System (BRFSS). Physical Activity. Participant's PA levels were assessed with the CDC's 2001 BRFSS PA module, assessing moderate physical activity (frequency and duration), vigorous physical activity (frequency and duration), and walking (frequency and duration), with examples of moderate and vigorous physical activities provided. Individuals were considered to be meeting recommendations if they participated in moderate intensity PA for 30 minutes a day on 5 or more days a week or vigorous intensity activity for 20 minutes a day on 3or more days a week. Participants were considered to be insufficiently active if they participated in more than 20 minutes of PA a week, but did not meet recommendations, and were considered to be inactive if they participated in less than 10 minutes of PA a week. Recently, a study comparing an objective physical activity monitoring technique with the BRFSS physical activity module reported 80% agreement between the two methods of classifying individuals who met the current CDC-ACSM physical activity recommendations (Strath, Bassett, Ham, & Swartz, 2003). Church PA programs._Respondents were asked to describe any PA programs that their church had offered in the past year. Participants taking part in programs were asked if their pastor or the pastor's spouse was involved in the program. Interpersonal Support for PA. Participants were asked if they had spoken with either the Health Director at their church, or any other church members, about joining a PA program, and a third question asking if anyone from their church had encouraged them to join a PA program. A combined variable was used to represent these three interpersonal support mechanisms from fellow AME nnembers, with possible scores of 0,1, or 2. Participation in Church Activities. Participants were asked to indicate (yes or no) whether they participated in choir, men's or women's auxiliary. Sons of Allen or Daughter of Sarah, Young People Department, or served as an usher, class leader, missionary, steward or stewardess, or trustee at their church. Statistical Analyses

Basic frequencies were conducted to assess distribution patterns for the variables, and Pearson product moment correlations were used to assess relationships between the variables. Chi-square analyses assessed the differences between groups for variables of interest. All significance levels were set at p