Physical Activity Interventions in Faith-Based Organizations: A Systematic Review

Abstract

Objective:

To review and assess the effectiveness of physical activity interventions delivered in faith-based organizations.

Data Source:

We searched the Cochrane Library, DoPHER, EMBASE, LILACS, MEDLINE, PsycINFO, WHO ICTRP, and Clinicaltrials.gov databases until January 2016, without restriction of language or publication date.

Study Inclusion and Exclusion Criteria:

Randomized and nonrandomized controlled trials investigating physical activity interventions for adults delivered in faith-based organizations.

Data Extraction:

Two independent reviewers extracted data and assessed study methodological quality.

Data Synthesis:

We used relative risk and mean difference with 95% confidence interval to estimate the effect of the interventions on measures of physical activity, physical fitness, and health.

Results:

The review included 18 studies. Study participants were predominantly female, and the majority of trials were conducted in the United States. Study heterogeneity did not allow us to conduct meta-analyses. Although interventions delivered in faith-based organizations increased physical activity and positively influenced measures of health and fitness in participants, the quality of the evidence was very low.

Conclusion:

Faith-based organizations are promising settings to promote physical activity, consequently addressing health disparities. However, high-quality randomized clinical trials are needed to adequately assess the effectiveness of interventions delivered in faith-based organizations.

Keywords

faith-based organization religion physical activity motor activity systematic review review literature as topic

Introduction

Physical activity is a protective health behavior associated with a reduced risk of all-cause mortality,¹ chronic diseases, including heart disease, type 2 diabetes, and many cancers.² Although the benefits of physical activity are well established, the majority of the population does not meet current physical activity guidelines of 150 minutes of moderate to vigorous physical activity (MVPA) per week.³ In 2008, insufficient levels of physical activity accounted for 9% of global premature mortality, 6% of coronary heart disease, 7% of type 2 diabetes, 10% of breast cancer, and 10% of colon cancer worldwide.⁴ Therefore, physical activity programs are a promising strategy to promote health.

Physical activity programs that are enjoyable and culturally appropriate are more likely to have high adherence, and accessibility to health programs influences adherence, especially for minorities and uninsured populations. Faith-based organizations (organizations with religious components in their self-identity, organization, and decision-making)⁵ appear to have the ability to reach those in need of interventions.⁶ Faith-based organizations have been widely used to deliver health programs associated with a number of health outcomes,^6
-8 especially in the United States. Faith-based organizations have been suggested by the Centers for Diseases Control and Prevention to promote social support for physical activity, thus being a promising setting to promote physical activity behavioral change.⁹

Faith-based organizations can implement physical activity programs in 2 primary ways. Although many programs include religious or spiritual components (eg, prayers, Bible studies, faith, messages of love, and health messages enriched with scripture contents),⁸ others simply use the setting to deliver a program without faith components. The current review considers “faith-based” programs (those integrating faith components) and “faith-placed” programs (those only using the setting for delivery).^6,7 Several factors make these programs promising to promote health behavior change, including a familiar setting, psychosocial factors (eg, self-efficacy, locus of control, motivation, social support, and social networks), or environmental factors (eg, accessibility to a location and sports facilities). Also, behavior change theories, models, or constructs can be integrated when planning and implementing a physical activity program, such as the transtheoretical model, socioecological model, cognitive theory, and social cognitive theory.¹⁰

Reviews of interventions delivered in faith-based organizations suggest that these interventions can have a positive impact on participants’ health.^6,8 A review of physical activity interventions conducted in 2012⁷ concluded that interventions delivered in faith-based organizations show promise for increasing physical activity participation and promoting health, but the authors did not account for bias of the included studies or assess measures of health and fitness.

This review will update findings on the effectiveness of physical activity interventions delivered in faith-based organizations. Furthermore, we will account for measures of health and physical fitness and assess the risk of bias of included studies.

Methods

Data Sources

To identify potential studies, we searched the following electronic databases: Cochrane Central Register of Controlled Trials (CENTRAL), Database of Promoting Health Effectiveness Reviews (DoPHER), EMBASE, LILACS, MEDLINE, PsycINFO, International Clinical Trials Registry Platform from the World Health Organization (http://apps.who.int/trialsearch/), and Clinicaltrials.gov. Studies were included independent of publication status, year of publication, and language. We also searched reference lists of included studies, and, when necessary, we contacted authors of primary studies for further information. Table 1 shows a sample electronic search strategy. We conducted searches during March 2015, with a final update in January 2016.

Table 1.

Search Strategy Conducted in MEDLINE via Pubmed.

Field	Search strategy
#1	(“Religion”[Mesh] or “Religions”[Text Word] or “Religious Beliefs”[Text Word] or “Beliefs, Religious”[Text Word] or “Religious Belief”[Text Word] or “Religious Ethics”[Text Word] or “Prayer”[Text Word] or “Prayers”[Text Word] or “Religion and Medicine”[Mesh] or “Medicine and Religion”[Text Word] or “Faith Healing”[Mesh] or “Healing, Faith”[Text Word] or “Prayer Healing”[Text Word] or “Healing, Prayer”[Text Word] or “Religion and Science”[Mesh] or “Religion and Psychology”[Mesh] or “Psychology, Religion”[Text Word] or “Religion, Psychology”[Text Word] or “Psychology and Religion”[Text Word] or “Spirituality”[Mesh] or “Spiritual Therapies”[Text Word] or “Spiritualities”[Text Word] or “Temple”[Text Word] or “Church”[Text Word] or “Mosque”[Text Word] or “Synagogue”[Text Word] or “Faith-based”[Text Word])
#2	(“Exercise”[Mesh] or “Exercises”[Text Word] or “Exercise, Physical”[Text Word] or “Exercises, Physical”[Text Word] or “Physical Exercise”[Text Word] or “Physical Exercises”[Text Word] or “Exercise, Isometric”[Text Word] or “Isometric Exercises”[Text Word] or “Isometric Exercise”[Text Word] or “Exercise, Aerobic”[Text Word] or “Aerobic Exercises”[Text Word] or “Aerobic Exercise”[Text Word] or “Sports”[Mesh] or “Sport”[Text Word] or “Athletics”[Text Word] or “Athletic”[Text Word] or “Exercise Movement Techniques”[Mesh] or “Pilates-Based Exercises”[Text Word] or “Pilates Based Exercises”[Text Word] or “Pilates Training”[Text Word] or “Exercise Therapy”[Mesh] or “Therapy, Exercise”[Text Word] or “Exercise Therapies”[Text Word] or “Therapies, Exercise”[Text Word] or “Motor Activity”[Mesh] or “Activity, Motor”[Text Word] or “Motor Activities”[Text Word] or “Physical Activity”[Text Word] or “Activities, Physical”[Text Word] or “Activity, Physical”[Text Word] or “Physical Activities”[Text Word] or “Locomotor Activity”[Text Word] or “Activity, Locomotor”[Text Word] or “Locomotor Activities”[Text Word])
#3	#1 AND #2

Inclusion and Exclusion Criteria

Our review included randomized controlled trials (RCTs) and non-RCTs with a control or comparison group. Trial participants were 18 years of age or older and able to participate in physical activity interventions. Participants were included independent of gender, ethnicity, or presence of chronic diseases. We included interventions delivered in faith-based organizations containing at least 1 active physical activity component, that is, any type of physical activity conducted (ie, group dance classes, walking groups, or strength training sessions). Comparison groups were physical activity interventions delivered in a secular setting or participants from faith-based organizations who did not participate in a physical activity intervention.

Primary outcomes of interest were changes in physical activity level (self-reported or objectively measured) and measures of physical health; that is, health variables related to physical activity (ie, body mass index [BMI], blood pressure, body weight, and cholesterol levels). Secondary outcomes were measures of physical fitness, adherence to the intervention, and adverse events related to the intervention. We also collected qualitative data from process measures related to the implementation of programs, when available.

Data Extraction

Two review authors (M.T.P. and G.M.J.P.) independently screened titles and abstracts and assessed the full text of selected abstracts applying the inclusion criteria mentioned earlier. They extracted data from included studies using a standard data extraction form, previously piloted to assure that it captured relevant information. These data were entered into a Microsoft Excel spreadsheet and verified.

Data Synthesis

We assessed the risk of bias using the Cochrane Collaboration’s Risk of Bias Tool.¹¹ We judged studies as at “high risk,” “low risk,” or “unclear risk” of bias for each domain. When needed, we solved disagreements through discussion in consensus meetings. For treatment effect, we summarized dichotomous data using relative risk (RR), with 95% confidence intervals (CIs). We handled continuous data as mean differences (MDs) with 95% CIs. We assessed physical activity results according to the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach,¹² which is a system for rating the quality of a body of evidence in systematic reviews, to help the evidence-based decision-making process. We also analyzed treatment effects considering the clinical relevance of the result, that is, we tried to verify whether the physical activity change would influence participants’ health. Our review protocol is registered on the International prospective register of systematic reviews PROSPERO (CRD42015019534).¹³

Results

We identified 9888 references and assessed the title and abstracts of 9096 references after removing duplicates. We selected 80 references and assessed the full text of these studies. We excluded 47 studies that did not meet the inclusion criteria. The reasons for exclusions were interventions with inadequate physical activity components for our review, other study designs, or programs delivered in secular settings. Finally, 18 studies were included^{14

-31} in the review (Figure 1). One study was classified as “ongoing”;¹⁵ for this reason, we presented its available characteristics but we did not assess data. In cases where we could not calculate measures of effects, we reported results as presented by primary authors. We present results narratively due to considerable clinical and methodological heterogeneity among studies. This heterogeneity impeded us from conducting meta-analyses of outcomes.

Figure 1.

Flow diagram of the review process of physical activity interventions delivered in faith-based organizations.

Nine studies were non-RCTs^{19,20,25

-31} and 9 were RCTs,^{14

-18,21,22
-24} all randomized at the cluster level. We present general characteristics of the studies in Tables 2 and 3. Most programs were conducted in the United States,^{14

-18,21

-27,30,31} 3 were conducted in New Zealand,^19,28,29 and 1 was conducted in Australia²⁰. One study²⁵ compared delivering an intervention in a faith-based organization versus a secular setting, whereas all other studies investigated physical activity programs delivered in faith-based organizations compared to a control, comparison, or minimal intervention groups. Three studies were faith placed,^19,20,23 whereas all others were faith based.^{14

-17,18,21,22,24

-31} Study participants were primarily African Americans^{14,18,21

-24,27,30,31}; other ethnicities included Hispanics,^15,16 caucasians,¹⁷ Tongans,²⁹ and Samoans.²⁸ Most studies targeted only women,^{14,15,17,20,24,25,27,31} and others had primarily female participants, except for 1 study with 85.7% male participants.²² Hypertension,^{14,22,23,30,31} diabetes,^18,28,29 and obesity^{14,19,23

-26,30} were the most prevalent comorbidities at baseline, that is, affecting more than 50% of the samples.

Table 2.

Characteristics of Included Studies: Randomized Controlled Trials (RCTs).

Study	Sample Characteristics	Physical Activity Intervention	Outcomes of Interest
Anderson¹⁴ United States, PASS: Physical activity with spiritual strategies	African American women N = 2 churches Intervention (n = 11, age 70 ± 5.5) Waiting list (n = 16, age 66 ± 5.47)	10 weeks, 1×/week, 90 min/session Strength training for major groups (1-2 sets of 8-10 reps). Health Promotion Model, Social Cognitive Theory, and Expectancy Value Theory. Delivered by researcher, faith based	Physical activity (7-DAI modified; % of individuals meeting Healthy People 2010 goal) Physical fitness (chair stand, arm curl, 6-minute walk tests)
Arredondo et al¹⁵ United States, Faith in Action (Fe en Acción)	Latino/Hispanic women N = 16 churches Physical activity intervention (n = 217, age = 44.5 ± 9.8) Cancer screening/attention-control (n = 219, age = 44.4 ± 9.4)	24 months, offered 6×/week, 60 min/session MVPA classes (walking groups, cardio dance, and strength training classes) Socioecological Model Trained community health workers (promotoras), faith based	Physical activity (Accelerometer and Global Physical Activity Questionnaire) Measures of health (BMI and waist circumference) Physical fitness (3-minute submaximal step test)
Bopp et al¹⁶ United States, Walk for life/Camina para vida	Latino/Hispanic (61.8% female) N = 3 churches Intervention (n = 24) Comparison (n = 23) Age = 42.5 ± 12.1	8-week walking contest (team based) Faith-based	None of interest; author’s assessed physical activity knowledge
Peterson et al¹⁷ United States, Heart and Soul Physical Activity Program	Caucasian women N = 4 churches Intervention (n = 22, age = 53.7 ± 9.7. Comparison (n = 20, age = 48.3 ± 7.7).	12 weeks, 1×/week, 60 min/session 15 minutes of group physical activity Social support (appraisal, belonging, tangible, and self-esteem domains) Delivered by researcher, faith based	Physical activity (7-day activity recall) Physical fitness (Rockport walk test to estimate VO₂ max)
Samuel-Hodge et al¹⁸ United States, A new DAWN-Diabetes Awareness and wellness network	African Americans with diabetes type II N = 24 churches Intervention (n = 117, 64% female, age = 57 ± 0.9) Minimal intervention (n = 84, 63% female, age = 61.3 ± 1.3)	12 months (8-month intensive intervention; 4-month reinforcement), 2×/month, 60 min/session 15 minutes of PA segment with chair exercises Model of chronic disease management, Social Cognitive Theory, Stages of Change Model, and Adult Learning Theory Church diabetes advisor (peer counselor with diabetes type 2 or having lived with someone with diabetes for at least 2 years), faith based	Physical activity (accelerometer) Measures of health (weight, blood pressure)
Wilcox et al²¹ United States, Health-E-AME faith-based PA initiative	African Americans (68.12% female) N = 20 churches Intervention (n = 311) Control (n = 260) Age: 27% were 18-34 years, 31.7% were 35-49 years, 32.7% were 50-64 years, and 24.2% were 65 years and older	3 years Action-oriented activities: praise aerobics, chair exercises, and walking contests. Step to fitness (8-week program): contained 20-30 minutes of physical activity Social Ecology, Transtheoretical Model, and Social Network Led by volunteers, faith based	Physical activity (Behavioral Risk Factor Surveillance System—physical activity module; proportion of participants meeting CDC-ACSM MVPA recommendations)
Wilcox et al²² United States, FAN: Faith, Activity and Nutrition	African Americans N = 74 churches Intervention (n = 749, 24.7% female, age = 53.8 ± 14) Delayed intervention (n = 508, 23.6% female, age = 54.6 ± 14.4)	15 months PA suggested as strategy to be implemented at churches Structured Ecologic Model and Social Cognitive Theory Delivered by a FAN committee, faith based	Physical Activity (CHAMPS questionnaire, accelerometer in a subsample) Measures of health (blood pressure)
Woods et al²³ United States, Living well by faith—Health and wellness program	African Americans (73% female) N = 5 churches Intervention (n = 74) Minimal intervention (n = 32) Age = 60% were 45-64 and 11% were ≥65	8 weeks, 2×/week, 90 min/session 30 minute PA (warm-up, stretch, aerobic activity, and cool down) Certified fitness specialists, faith placed	Physical activity (Behavioral Risk Factor Surveillance System—physical activity module) Measures of health (blood pressure, weight, BMI, % body fat) Physical fitness (step test—YMCA protocol)
Young and Stewart ²⁴ United States	African American women N = 11 churches Intervention (n = 123, age = 48.2 ± 2.2) Stretching group (n = 73, age = 48.4 ± 2.3)	6 months, 1×/week, 60 min/session Aerobic exercise: 5-10 minutes warm-up, 30-40 minutes of aerobic exercise (ie, dance), 5-10 minutes of floor exercise, and cool down Social cognitive theory Certified aerobics instructors from the community, faith based	Physical activity (Stanford 7-Day Physical Activity Recall and Yale Physical Activity Survey) Measures of health (cardiovascular risk factors, weight, BMI, blood pressure, blood lipids) Physical fitness (cardiorespiratory—oxygen uptake)

Abbreviations: ACSM, American College of Sports Medicine; BMI, body mass index; CDC, Centers for Disease Control and Prevention; CHAMPS, Community Healthy Activities Model Programs for Seniors Institute for Health & Aging; MVPA, moderate to vigorous physical activity; PA, physical activity; VO₂ max, maximum rate of oxygen consumption; YMCA, Young Men’s Christian Association; 7-DAI: 7-Day Activity Interview.

Table 3.

Characteristics of Included Studies: Non-RCTs.

Study	Sample	Physical Activity Intervention	Outcomes of Interest
Bell et al¹⁹ New Zealand, Samoan Ola Fa’autauta (Life-Wise) Study	Samoans N = 3 churches Intervention (n = 365, age = 43.9 ±13.7, 60% female) Control (n = 106, age = 39.1 ± 13.2, 64% female)	12 months, 1×/week Aerobic physical activity Trained instructors then trained church members, faith placed	Physical Activity (whether participating in moderate or vigorous leisure time activity during normal week) Measures of health (BMI, weight, waist and hip circumference, and blood pressure)
Jorna et al²⁰ Australia, embracing a healthy lifestyle	Women members of 1 church (ethnicity unclear) Intervention (n = 19, age = 40.2 ± 10.4) Nonhealth-related relationship education (n = 30, age = 38.3 ± 13)	8 weeks, 1×/week 30 minutes of moderate-intensity physical activity Social cognitive theory Trained staff qualified in a health discipline, faith placed	Physical activity (CHAMPS questionnaire)
Kim et al²⁶ United States, The WORD—Wholeness, Oneness, Righteousness, Deliverance	African Americans N = 4 churches Intervention (n = 36, age 57.7 ± 11.5, 69% female) Delayed intervention (n = 37, age 50.9 ± 14.2, 73% female)	8 weeks, 1×/week 15 minutes of physical activity with exercise tape Transtheoretical model and social support Lay health leaders, faith based	Physical Activity (16-item checklist measuring frequency and duration of types of activities) Measures of health (weight, waist and hip circumference)
Murrock and Gary²⁷ United States	African American women N = 2 churches Intervention (n = 66) Comparison (n = 60) Age = 36-82	8 weeks, 2×/week, 45 min/session Gospel music dance classes (5-minute warm-up, 30-minute dance, 10-minute cool down). Social cognitive theory Dance instructor from the community, faith based	Physical Activity (PASE—10-item scale assessing physical activity during the past 7-day time frame) Measures of health (body fat, BMI)
Simmons et al²⁸ New Zealand	Samoans at high risk of diabetes N = 2 Intervention (n = 67; 66% female; age = 37 ± 16) Control (n = 115; 61% female; age = 35 ± 17)	24 months, 1×/week on first year and 2×/week on second year Sitting exercises, low impact aerobics, traditional dance movements, walking, and sports. Trained Samoan health workers, faith based	Physical activity (exercise questions) Measures of health (weight, BMI, waist and hip circumferences, waist: hip ratio)
Simmons et al²⁹ New Zealand	Tongans at high risk of diabetes N = 2 Intervention (n = 167, 52% female, age = 13-79) Control (n = 86, 49% female, age = 15-82)	2 years; 1×/week on first year and 2×/week on second year Sitting exercises, low impact aerobics, traditional dance movements, walking, and sports Led by Bishop’s wife, faith based	Physical activity (exercise questions) Measures of health (weight, BMI, waist circumference)
Thomson et al³⁰ United States, Delta Body and Soul III	African Americans N = 8 churches Intervention (n = 287, 75.3% female, age = 47.3 ± 14.4) Control (n = 122, 63.9% female, age = 47 ± 16.4)	6 months, physical activity every 3 weeks (20 sessions of 60 minutes) Zumba set to gospel music, strength, and flexibility activities with bands and resistant tubes Transtheoretical model Certified fitness instructor, faith based	Physical activity (Rapid Assessment of Physical Activity survey) Measures of health (BMI, blood pressure, nonfasting blood lipids)
Yanek et al³¹ United States, Project Joy	African American women N = 16 churches Spiritual intervention (n = 267, age = 53.6 ± 9) Standard intervention (n = 188, age = 51.9 ± 9) Self-help (n = 74, age = 53.9 ± 10)	12 months (20 weeks of intensive intervention), 1×/week 30 minutes of moderate physical activity (brisk walking, water aerobics, or Tae Bo) Community Action and Social Marketing Model, Social Learning Theory, and Social Support Health educators and lay health leaders, faith based	Physical Activity (Yale Physical Activity Survey) Measures of health (weight, BMI, % of body fat, waist circumference, blood pressure, and blood lipids)
Dornelas et al²⁵ United States^a	Women at high risk of cardiovascular disease (48% Hispanic, 46% African American) N = 3 Spanish-speaking church (n = 23) African American Church (30) Women’s health clinic (n = 23) Age (median) = 40 years	10 weeks, 2×/week, 50 min/session Dance aerobics at moderate intensity (64%-76% of HR). Sessions: 5 minutes warm-up, 25 minutes of aerobic activity, 10 minutes of resistance training, 10 minutes of stretch/cool down Certified fitness instructors, faith based	Measures of health (weight, waist to hip ratio and blood pressure) Physical Activity (Baecke Questionnaire of Habitual Physical Activity) only reported at baseline

Abbreviation: BMI, body mass index; CHAMPS, Community Healthy Activities Model Programs for Seniors Institute for Health & Aging; HR, heart rate; PASE: Physical Activity Scale for the Elderly; RCTs, randomized controlled trials.

^aComparison physical activity delivered in a faith-based organization versus secular setting.

Assessment of Bias

We assessed the risk of bias of 17 studies because the study by Arredondo et al¹⁵ is ongoing. Selection bias was unclear in 8 studies^{14,16
-18,21

-24} and high in 9 studies.^{19,20,25

-31} Due to the nature of physical activity interventions, it is impossible to blind participants from their allocation to the intervention or control group. From that standpoint, we judged performance bias as not applicable with the exception of interventions delivered by the principal investigator,^14,17 where performance bias might occur. Detection bias was low in 2 studies^23,24 with blinded assessors, unclear in 10 studies,^{16,18,20
-22,25
-27,30,31} and high risk in 5 studies.^{14,17,19,28,29} Attrition bias was low in 2 studies^18,28 and unclear in 7 studies^{14,17,19,20,23,24,26} because studies did not provide information for losses and withdrawals. Attrition was considerable from baseline to follow-up measures (higher than 20%) in 8 studies.^{16,21,22,25,27,29,30,31} Six studies had low risk of reporting bias,^17

-21,31 and all other studies presented reporting issues; thus, we classified them as high risk of reporting bias.^{14,16,22

-29} Four studies did not have additional issues; therefore, they were low risk for other sources of bias.^16,21,22,27

We found potential threats to validity in most studies. Nonrandomized studies were exposed to bias due to their study design.^{19,20,25

-31} Major baseline imbalances were identified in 3 studies.^14,17,26 We identified the following additional sources of bias: first, studies lacked a true control group.^{18,23,24,26,28,31} Second, in 1 study, participants were required to pay a fee to attend exercise classes.²⁰ Also, contamination bias potentially occurred in 1 study because intervention and control churches were in close proximity to each other (about 5 km apart).²⁹ Finally, 1 study was conducted in a region where previous interventions were conducted (Delta Body and Soul I and II). In this case, the same participants could have been recruited by similar studies more than once.³⁰ We summarize the risk of bias for all included studies in Figures 2 and 3.

Figure 2.

Risk of bias summary for each risk of bias item presented in percentages across all included studies.

Figure 3.

Risk of bias graph for each risk of bias item for each included study.

Measures of Physical Activity, Physical Health, and Physical Fitness

Physical activity

We present measures of changes in physical activity, physical health, and physical fitness in Table 4. Fourteen studies reported on changes in physical activity.^{14,17

-22,24,26

-31} Authors measured physical activity in various ways and using various tools/questionnaires. They also reported it differently, for example, physical activity was reported in time exercising in different intensities (light, moderate, and vigorous), total physical activity, leisure physical activity time, or percentage of participants meeting recommendations. We focused on changes in total physical activity and MVPA because those intensities have a greater impact on participants’ health.³ Five studies^{14,19,20,26,30} found significant changes in physical activity outcomes favoring the intervention, but only Bell et al¹⁹ found significance in participation of MVPA (RR: 0.66, 95% CI: 0.46 to 0.96) and in participation in vigorous leisure time (RR: 1.85, 95% CI: 1.16-2.25), but it was unclear if the questionnaire used in this study was a validated tool. Overall, the quality of evidence for physical activity outcome was very low according to the GRADE approach.

Table 4.

Measures of Changes in Physical Activity, Physical Health, and Physical Fitness.

Study and Sample Size	Physical Activity Changes	Measures of Physical Health	Physical Fitness
Anderson ¹⁴ I (n = 11), C (n = 16)	Time spent in moderate-intensity physical activity (min/week): MD, 76.36; 95% CI, −162.85 to 307.57. Time spent walking (min/week): MD, 56.60; 95% CI, 13.67-126.87. Time spent performing muscle strength (min/week)^a,b: MD, 50.19; 95% CI, 19.81-80.57. Participants reaching the Healthy People 2010 goal: RR, 1.06; 95% CI: 0.65-1.73	NA	Strength of upper body (repetitions): MD, 0.00; 95%CI, 3.80-3.80. Strength of lower body (repetitions): MD, 0.04; 95%CI −2.77 to 2.85. Cardiorespiratory fitness (yards change)^a,b: MD, 108.17; 95%CI, 21.74-194.60
Bell et al¹⁹ I (n = 365), C (n = 106)	Participation in moderate or vigorous physical activity^a,b: RR, 0.66; 95%CI, 0.46 to 0.96. Participation in brisk walking: RR, 0.83; 95%CI, 0.59 to 1.18. Participation in vigorous leisure time physical activity^a,b: RR, 1.85; 95% CI, 1.16 to 2.95	Weight (kg)^a,b,c: MD −1.40, 95% CI −6.84 to 4.04. BMI (kg/m²)^a,c: MD −0.10, 95% CI −2.03 to 1.83. Waist circumference (cm): MD 1.60, 95% CI −2.52 to 5.72. Hip circumference (cm): MD −0.90, 95% CI −4.23 to 2.43. Systolic blood pressure (mm Hg)^a,c: MD 6.00, 95% CI 0.46 to 11.54. Diastolic blood pressure (mm Hg)^a,c: MD 6.00, 95%CI 1.62 to 10.38	NA
Jorna et al²⁰ I (n = 19), C (n = 30)	Walking (h/wk change)^a,b: MD, 0.49; 95% CI, 0.11 to 0.87. Moderate to vigorous physical activity (h/wk change): MD, 0.61; 95% CI, −0.03 to 1.25	NA	NA
Kim et al²⁶ I (n = 36), C (n = 37)	Recreational physical activity (METs change)^a,b: MD, 6.40; 95% CI 2.08 to 10.72. Total physical activity (METs change): MD, 5.50; 95% CI, −1.45 to 12.45	Weight (lbs change)^a,b: MD, −2.78; 95% CI, −4.64 to −0.92. Waist circumference (cm change): MD, −1.80; 95% CI, −5.29 to 1.69. Hip circumference(cm change): MD, −2.30; 95% CI, −4.60 to 0.00	NA
Murrock and Gary²⁷ I (n = 66), C (n = 60)	Self-reported physical activity (PASE scale score): MD, 13.90; 95% CI, −15.16 to 42.96	BMI (kg/m2): MD, −2.50; 95% CI, −5.22 to 0.22. Body fat (%)^a,b: MD, −2.20; 95% CI, −4.39 to −0.01	NA
Peterson et al¹⁷ I (n = 22), C (n = 20)	Time spent in moderate physical activity (min/week): MD, 38.33; 95% CI, −37.65 to 114.36. Energy expenditure (kcal/week): MD, −13.05; 95% CI, −752.33 to 726.24	NA	Cardiorespiratory fitness (mL O₂/kg/min): MD, 4.45; 95%CI, −1.47 to 10.37
Samuel-Hodge et al¹⁸ I (n = 117), C (n = 84)	Total physical activity (h/d)^c: MD, 0.20; 95% CI, −0.4 to 0.8. Physical activity (total hours)^c: MD, 0.40; 95% CI, −6.3 to 7.1. Light physical activity (min)^c: MD, 6.90; 95% CI, −2.7 to 16.6. Moderate physical activity (min)^a: MD, −2.7; 95% CI, −0.5 to −4.9	Weight (kg)^c: MD, −0.2; 95% CI, −1.4 to 1.0 Systolic blood pressure (mm Hg)^c: MD, 0.8; 95% CI, −4.0 to 5.6. Diastolic blood pressure (mm Hg)^c: MD, 1.7, 95% CI, −1.0 to 4.5	NA
Simmons et al²⁸ I (n = 67), C (n = 115)	Number of days per week people exercised: MD, 0.50; 95% CI, −0.19 to 1.19. People exercising more than 3 days per week: RR, 1.39; 95% CI, 0.98 to 1.98	Weight (kg)^a,b: MD, −3.10; 95% CI, −5.50 to −0.70. BMI (kg/m²): MD, −2.00; 95% CI, −4.15 to 0.15. Waist circumference (cm): MD, −2.00; 95% CI, −6.81 to 2.81; Hip circumference (cm)^a,b: MD, −6.00; 95% CI, −10.73 to −1.27. Waist to hip ratio: MD, −0.01; 95% CI, −0.03 to 0.01	NA
Simmons et al²⁹ I (n = 167), C (n = 86)	People exercising more than 3 days per week: −2% for intervention versus −3% for control, P value nonsignificant	Weight (kg): MD, 2.30; 95% CI, −0.33 to 4.93. Waist circumference (cm): MD, 3.00; 95% CI, −1.39 to 7.39	NA
Thomson et al³⁰ I (n = 287), C (n = 122)	A significant change in strength/flexibility-related physical activity was seen in the intervention group, and control participants had a decrease in their physical activity after intervention (data not shown)	BMI (kg/m²): P = .1398. Systolic blood pressure (mm Hg): P = .9011 Diastolic blood pressure (mm Hg): P = .1747 Blood lipids—LDL cholesterol (mg/dL): P = 0.0843 Blood lipids—HDL cholesterol (mg/dL): P = 0.8284 Blood lipids—total cholesterol (mg/dL): P = .4534 Blood lipids—triglycerides (mg/dL): P = 0.3897	NA
Wilcox et al²¹ I (n = 311), C (n = 260)	No significant intervention impact in moderate physical activity participation or meeting CDC recommendations for physical activity (values not reported)	BMI (kg/m²): MD, −0.80; 95%CI, −3.05 to 1.45	NA
Wilcox et al²² I (n = 749), C (n = 508)	Leisure time moderate to vigorous physical activity (hrs/wk)^a: MD, −0.10; 95% CI, −0.16 to −0.04. Total moderate to vigorous physical activity (hrs/wk)^a: MD, −0.10; 95% CI, −0.18 to −0.02	Systolic blood pressure (mm Hg): MD, 3.45; 95% CI, −1.09 to 7.99. Diastolic blood pressure (mm Hg): MD, 1.14; 95% CI, −0.81 to 3.09	NA
Woods et al²³ I (n = 74), C (n = 32)	NA	Weight (lb): A decrease of 5 lb for intervention compared to no change in the control group (P < .02) BMI (kg/m²): Decrease for intervention in comparison to control group (P <.05) Body fat (%): Decrease of 3.5% for intervention compared to no change in control group (P <.03) Blood pressure (mm Hg): No changes for systolic and diastolic	Cardiovascular fitness (%): A significant increase of participants from the intervention group performing average or better (from 41% to 70%, P < .02) in the step test, compared to no change in the control group
Yanek et al³¹ I (n = 267), C (n = 188)	Energy expenditure: MD, 9.80; 95% CI, −87.53 to 107.13	Weight (lbs)^a,b: MD, −1.93; 95% CI, −3.24 to −0.62 BMI (kg/m²)^a,b: MD, −0.31; 95% CI, −0.53 to −0.09. Body fat (%): MD, −0.26; 95% CI, −0.55 to 0.03 Waist circumference (inches)^a,b: MD, −0.65; 95% CI, −1.10 to −0.21. Systolic blood pressure (mm Hg): MD −0.65, 95% CI −3.05 to 1.75 Diastolic blood pressure (mm Hg): MD 0.58, 95% CI −1.92 to 0.76. Blood lipids—LDL cholesterol (mg/dL): MD −0.58, 95% CI −1.92 to 0.76 Blood lipids—HDL cholesterol (mg/dL): MD −0.79, 95% CI −2.31 to 0.73	NA
Young Stewart²⁴ I (n = 123), C (n = 23)	Physical activity between aerobic exercise and stretching groups was not significantly different at end of the intervention (P = 0.3). Prevalence of physical inactivity: RR, 1.01; 95% CI, 0.66 to 1.54	Cardiovascular risk factors: Less than 50% of participants completed follow-up assessments, data not reported	Cardiovascular fitness: Less than 50% of the sample completed follow-up assessments, data not reported

Abbreviations: BMI, body mass index; C, control; CDC, Centers for Disease Control and Prevention; CI, confidence interval; I, intervention; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MD, mean difference; METs, metabolic equivalent of task; NA, not available; PASE, Physical Activity Scale for the Elderly; RR, relative risk.

^aValues statistically significant for P < .05.

^bValues favoring the intervention.

^cValues were adjusted by primary studies, thus results might be slightly different in values.

Physical health

Studies reported various measures of physical health. Seven studies reported body weight,^{19,18,23,26,28,29,31} and 2 of those found that the intervention did not result in weight loss.^18,29 From the studies, which found that the intervention influenced body weight, Bell et al¹⁹ reported weight loss when the analysis was adjusted for baseline and sociodemographic characteristics (P = .039). The following 3 studies found that the intervention resulted in weight loss: Kim et al²⁶ (MD: −2.78, 95% CI: –4.64 to –0.92 lbs); Yanek et al³¹ (MD: –1.93, 95% CI: −3.24 to –0.62 lbs), and Simmons et al²⁸ (MD: −3.10, 95%CI: −5.50 to −0.70 kg). Additionally, Woods et al²³ noted a loss of 5 pounds in intervention participants.

Body mass index was reported by 7 studies.^{19,21,23,27,28,30,31} Changes favoring the intervention was seen by Yanek et al³¹ (MD: −0.31, 95% CI: −0.53 to −0.09) and Woods et al.²³ Three studies measured percentage of body fat.^27,23,31 Murrock and Gary²⁷ found a decrease favoring interventions (MD: −2.20, 95% CI: −4.39 to −0.01), whereas Woods et al²³ reported a decrease of 3.5% in intervention participants (P < .03). Several studies reported body circumference measures. Five studies reported waist circumference,^{19,26,28,29,31} and Yanek et al³¹ found a decrease favoring interventions (MD: −0.65, 95% CI: −1.10 to −0.21 inches). Three studies measured hip circumference,^19,26,29 and Simmons et al²⁸ found a decrease favoring interventions (MD: −6.00, 95% CI: −10.73 to −1.27 cm). In addition to anthropometric data, 8 studies reported metabolic-related outcomes to indicate health benefits. Six studies assessed blood pressure^{18,19,22,23,30,31} and 2 studies measured blood lipids,^30,31 but there were no changes favorable to interventions.

Physical fitness

Three studies assessed physical fitness.^14,17,23 Two studies found changes in cardiorespiratory fitness favoring interventions.^14,23 Intervention group participants¹⁴ walked more yards (MD: 108.17, 95% CI: 21.74 to 194.60) when compared to control participants, and intervention group participants performed better on the 6-minute walk test compared to the control group participants.²³

Intervention adherence

Measures of adherence to the physical activity intervention were reported by most studies.^{14,18
-20,23
-25,27

-31} Studies measured adherence as percentages of participants who attended intervention,^14,18,24 number of sessions attended,^23,27 and average number of participants per class.^19,25,28,31 Some studies only measured adherence in a subsample of participants.^28,29 Adherence to the intervention varied across studies, ranging from 81.8%¹⁴ to 24.3%.²⁴

Adverse events

Adverse events did not appear to be a concern. Although 1 study¹⁸ reported 2 participant deaths (caused by a stroke and by an advanced chronic obstructive pulmonary disease), authors noted that it was not related to the intervention. Murrock and Gary²⁷ attributed intervention group attrition rates to participants’ health concerns. Other studies did not report adverse events.

Discussion

Our review of physical activity interventions delivered in faith-based organizations included 18 studies. Although some of the interventions showed promise for increasing physical activity and measures of physical health and fitness, overall the low quality of scientific evidence available to date does not allow firm conclusions on the effectiveness of such interventions. Using the GRADE approach, we rated the physical activity data available for this review as very low quality. Two studies reported positive changes, but one of these studies did not present full data³⁰ and the other assessed physical activity participation with a questionnaire that lacked information about adequate validation.¹⁹ Also, both programs provided less than the recommended amount of physical activity deemed necessary to provide meaningful benefits to cardiovascular and musculoskeletal fitness³: The Bell et al intervention involved weekly sessions¹⁹ and the Thomson et al intervention involved 1 session every 3 weeks.³⁰ Of note, these programs included nutrition classes in addition to a physical activity intervention.^19,30 Thus, it is difficult to disentangle whether physical activity was sufficient to promote the behavior change alone or if additional components influenced effectiveness.

Participants who engaged in physical activity in faith-based organizations benefited from anthropometric changes. Interventions were somewhat effective in promoting decreases in body weight, BMI, body fat, and waist and hip circumference. Although the reported anthropometric changes were statistically significant, readers should interpret this finding with caution, as the magnitude of change was small and unlikely to result in significant health benefits. However, the results point to a positive direction, and future studies could potentially promote changes with clinically significant impacts on participants’ health. Interventions assessed in this review did not benefit blood lipids or blood pressure. But studies that assessed blood lipids and blood pressure offered physical activity sessions only once a week³¹ or every 3 weeks,³⁰ a frequency unlikely to significantly benefit cardiovascular fitness.³

Engaging in physical activity can produce positive changes in measures of physical fitness, including cardiorespiratory endurance, muscular endurance, muscular strength, body composition, and flexibility.³² Our review noted 3 studies that reported physical fitness outcomes.^14,17,23 Only Anderson found benefits to cardiorespiratory fitness but did not improve muscle strength,¹⁴ despite offering weekly strength classes for 10 weeks. This frequency of strength training was likely insufficient to show benefits, as the American College of Sports Medicine (ACSM) recommends strength training 2 to 3 times per week to improve and maintain muscle fitness.³ This example highlights one of the complexities of delivering and evaluating physical activity interventions, as the amount of exercise (intensity and frequency) necessary to produce meaningful fitness improvements varies among individuals and depends on their baseline activity levels. Although measures of fitness increase the robustness of a physical activity program in terms of delivery (valuable feedback for participants and researchers) and evaluation (to assess the effectiveness of the intervention), we were concerned that many studies did not assess physical fitness.

Primary studies reported no adverse events related to the intervention, and only 1 study reported health concerns as reason for participant’s attrition.²⁷ Based on these studies, physical activity interventions delivered in faith-based organizations appear to be safe for participants. Adherence across programs varied substantially; but given the variability in interventions and program delivery, it is not possible to draw conclusions about adherence to physical activity interventions in faith-based organizations; also, because most interventions included additional components, it is uncertain whether adherence would have been different if programs offered only physical activity.

Our review had positive aspects and limitations. We conducted a comprehensive search for studies, and 2 reviewers independently assessed studies and extracted data to minimize errors. We assessed risk of bias and calculated measures of effects from primary studies whenever possible. The high clinical and methodological heterogeneity found between studies did not allow us to conduct meta-analyses of outcomes. Also, because interventions differed greatly from one another, it was challenging to adequately assess the risk of bias. Due to the nature of the intervention (physical activity), double blinding was not applicable, and we judged blinding as low risk of bias for all studies, understanding that it would not cause selection bias. We encountered incomplete correspondence with trial authors, and as a consequence, our review lacked additional information. We came across studies that met our inclusion criteria, but we were unable to extract data for them.^27,16 The inclusion of nonrandomized studies could be a strength or weakness in our review: Including these studies allows readers to be aware of the relevant literature, but the lack of randomization decreases confidence in results from these studies.

Regarding the generalizability of evidence, readers should note that the reviewed studies included primarily middle-aged women (mean age 49.2 years); therefore, our review findings may not apply to interventions in men or younger/older populations. In addition, the majority of the interventions were conducted in US faith-based organizations, and findings may not apply to interventions conducted in other countries. Differences in intervention programming, delivery, and focus of the interventions are also important to note. Many interventions included components other than physical activity (eg, nutrition classes and motivation calls); in some cases, community volunteers helped set up programs, and local leaders actively participated in the planning and development of programs. Researchers applied different theories and constructs for behavior change. Readers should consider these aspects when interpreting our results and while planning a physical activity program in faith-based organizations.

Our results corroborate and update an earlier comprehensive review⁷ showing that faith-based organizations show potential as a promising setting for physical activity interventions. Although the literature on physical activity interventions in faith-based organizations is growing, our review found many sources of bias in the included studies, underscoring the need for more rigorous research in this area.

Conclusion

Faith-based organizations are promising, low-cost settings to promote physical activity that can help members overcome barriers to adopting new health behaviors and consequently address health disparities. We identified modest improvements in physical activity levels, physical health, and physical fitness. There is a need for high-quality randomized clinical trials to adequately assess the effectiveness of interventions delivered in faith-based organizations. We recommend more rigorous study designs, physical fitness measures, validated evaluation criteria, and comprehensive reporting to help advance our understanding of promoting physical activity in faith-based organizations. We recommend the use of fewer components in interventions to control for potential confounding factors. Studies conducted outside the United States with diverse ethnicities, balanced in gender and focusing on younger and older age-groups, would help provide better evidence of effectiveness of physical activity programs in faith-based organizations. These findings inform public health practitioners developing faith-based physical activity programs about characteristics and effectiveness available in the literature.

So What? Implications for Health Promotion Practitioners and Researchers

What is already known on this topic?

Physical activity is an important behavior to prevent disease and to promote health. A previous review⁶ showed that physical activity programs in faith-based organizations are promising.

What does this article add?

Our results update the literature in this area. We performed a broader search of studies, judged the quality of evidence by assessing various sources of study bias, and calculated measures of effects of interventions. Physical activity programs did not promote major changes in physical activity behavior. Yet, they promoted positive changes in measures of health and cardiorespiratory fitness, and we did not identify serious adverse events. All findings of this review had a low quality of evidence.

What are the implications for health promotion practice or research?

Interventions in faith-based organizations have the potential to improve health outcomes and physical fitness, but there is limited evidence about the effectiveness to change physical activity behavior. There is a need for more rigorous research to identify best practices. New research should aim to reduce sources of bias to allow for a better evaluation of the effectiveness the intervention. Additionally, researchers should preferentially use physical fitness outcomes to reflect effects of physical activity interventions.

Footnotes

Acknowledgments

The authors thank Sheila Kealey for providing great assistance revising the final version of the manuscript.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: E.A. Principal Investigator of “Faith in Action,” an ongoing cluster randomized faith-based intervention; M.T.P. collaborated with “Faith in Action” study as research assistant.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was supported by CAPES Foundation, Ministry of Education of Brazil, Brasília – DF 70040-020, Brazil, providing a Doctoral “sandwich” study abroad scholarship for M.T.P.

References

Clays

Lidegaard

De Bacquer

. The combined relationship of occupational and leisure-time physical activity with all-cause mortality among men, accounting for physical fitness. Am J Epidemiol. 2014;179(5):559–566. doi:10.1093/aje/kwt294.

Zethelius

Gudbjornsdottir

Eliasson

Eeg-Olofsson

Cederholm

. Level of physical activity associated with risk of cardiovascular diseases and mortality in patients with type-2 diabetes: report from the Swedish National Diabetes Register. Eur J Prev Cardiol. 2014;21(2):244–251. doi:10.1177/2047487313510893.

Garber

Blissmer

Deschenes

; American College of Sports Medicine. American College of Sports Medicine position stand. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Med Sci Sports Exerc. 2011;43(7):1334–1359. doi:10.1249/MSS.0b013e318213fefb.

Lee

Shiroma

Lobelo

Puska

Blair

Katzmarzyk

; Lancet Physical Activity Series Working Group. Effect of physical inactivity on major non-communicable diseases worldwide: an analysis of burden of disease and life expectancy. Lancet. 2012;380(9838):219–229. doi:10.1016/S0140-6736(12)61031-9.

Ebaugh

Pipes

Chafetz

Daniels

. Where’s the religion? Distinguishing faith-based from Secular Social Service Agencies. J Sci Study Relig. 2003;42(3):411–426. doi:10.1111/1468-5906.00191.

DeHaven

Hunter

Wilder

Walton

Berry

. Health programs in faith-based organizations: are they effective? Am J Public Health. 2004;94(6):30–36.

Bopp

Peterson

Webb

. A comprehensive review of faith-based physical activity interventions. Am J Lifestyle Med. 2012;6(6):460–478. doi:10.1177/1559827612439285.

Newlin

Dyess

Allard

Chase

Melkus

. A methodological review of faith-based health promotion literature: advancing the science to expand delivery of diabetes education to Black Americans. J Relig Health. 2012;51(4):1075–1097. doi:10.1007/s10943-011-9481-9.

Centers for Disease Control and Prevention. Strategies to Prevent Obesity and Other Chronic Diseases: The CDC Guide to Strategies to Increase Physical Activity in the Community. Atlanta: U.S. Department of Health and Human Services; 2011.

10.

McKenzie

Neiger

Thackeray

. Planning, Implementing, & Evaluating Health Promotion Programs: A Primer. 6th ed. Glenview, IL: Pearson Education, Inc; 2013.

11.

Higgins

Altman

Gotzsche

; Cochrane Bias Methods Group; Cochrane Statistical Methods Group. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928. doi:https://dx-doi-org.web.bisu.edu.cn/10.1136/bmj.d5928.

12.

Guyatt

Oxman

Vist

; GRADE Working Group. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008;336(7650):924–926. doi:10.1136/bmj.39489.470347.AD.

13.

Parra

Porfírio

GJM

Arredondo

Atallah

. Interventions in faith-based organizations for increasing physical activity: a systematic review. PROSPERO Int Prospect Reg Syst Rev. 2015;CRD42015019534.

14.

Anderson

. An intervention to increase physical activity behavior among older African American women. Dissertation Abstr Int Sec B Sci Eng. 2010;71:2306.

15.

Arredondo

Haughton

Ayala

. Fe en Accion/Faith in Action: Design and implementation of a church-based randomized trial to promote physical activity and cancer screening among churchgoing Latinas. Contemp Clin Trials. 2015;45(pt B):404–415. doi:10.1016/j.cct.2015.09.008.

16.

Bopp

Fallon

Marquez

. A faith-based physical activity intervention for Latinos: outcomes and lessons. Am J Health Promot. 2011;25(3):168–171. doi:10.4278/ajhp.090413-ARB-138.

17.

Peterson

Yates

Atwood

Hertzog

. Effects of a physical activity intervention for women. West J Nurs Res. 2005;27(1):93–110.

18.

Samuel-Hodge

Keyserling

Park

Johnston

Gizlice

Bangdiwala

. A randomized trial of a church-based diabetes self-management program for African Americans with type 2 diabetes. Diabetes Educ. 2009;35(3):439–454. doi:10.1177/0145721709333270.

19.

Bell

Swinburn

Amosa

Scragg

. A nutrition and exercise intervention program for controlling weight in Samoan communities in New Zealand. Int J Obes Relat Metab Disord. 2001;25(6):920–927. doi:10.1038/sj.ijo.0801619.

20.

Jorna

Ball

Salmon

. Effects of a holistic health program on women’s physical activity and mental and spiritual health. J Sci Med Sport. 2006;9(5):395–401. doi:10.1016/j.jsams.2006.06.011.

21.

Wilcox

Laken

Bopp

. Increasing physical activity among church members: community-based participatory research. Am J Prev Med. 2007;32(2):131–138. doi:10.1016/j.amepre.2006.10.009.

22.

Wilcox

Parrott

Baruth

. The faith, activity, and nutrition program: a randomized controlled trial in African-American churches. Am J Prev Med. 2013;44(2):122–131. doi:10.1016/j.amepre.2012.09.062.

23.

Woods

Levinson

Jones

. The Living Well by Faith health and wellness program for African Americans: an exemplar of community-based participatory research. Ethn Dis. 2013;23(2):223–239.

24.

Young

Stewart

. A church-based physical activity intervention for African American women. Fam Community Health. 2006;29(2):103–117.

25.

Dornelas

Stepnowski

Fischer

Thompson

. Urban ethnic minority women’s attendance at health clinic vs. church based exercise programs. J Cross Cult Gerontol. 2007;22(1):129–136. doi:10.1007/s10823-006-9023-1.

26.

Kim

Linnan

Campbell

Brooks

Koenig

Wiesen

. The WORD (wholeness, oneness, righteousness, deliverance): a faith-based weight-loss program utilizing a community-based participatory research approach. Health Educ Behav. 2008;35(5):634–650. doi:10.1177/1090198106291985.

27.

Murrock

Gary

. Culturally specific dance to reduce obesity in African American women. Health Promotion Pract. 2010;11(4):465–473. doi:10.1177/1524839908323520.

28.

Simmons

Fleming

Voyle

Fou

Feo

Gatland

. A pilot urban church-based programme to reduce risk factors for diabetes among Western Samoans in New Zealand. Diabet Med. 1998;15(2):136–142.

29.

Simmons

Voyle

Fou

Feo

Leakehe

. Tale of two churches: differential impact of a church-based diabetes control programme among Pacific Islands people in New Zealand. Diabet Med. 2004;21(2):122–128.

30.

Thomson

Goodman

Tussing-Humphreys

. Diet quality and physical activity outcome improvements resulting from a church-based diet and supervised physical activity intervention for rural, southern, African American adults: Delta Body and Soul III. Health Promot Pract. 2015;16(5):677–688. doi:10.1177/1524839914566851.

31.

Yanek

Becker

Moy

Gittelsohn

Koffman

. Project Joy: faith based cardiovascular health promotion for African American women. Public Health Rep. 2001;116(suppl 1):68–81.

32.

Caspersen

Powell

Christenson

. Physical activity, exercise, and physical fitness: definitions and distinctions for health-related research. Public Health Rep. 1985;100(2):126–131.