Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Sep 25.
Published in final edited form as: J Behav Med. 2018 Apr 28;41(4):537–549. doi: 10.1007/s10865-018-9926-7

Project SHINE: Effects of a Randomized Family-based Health Promotion Program on the Physical Activity of African American Parents

Sara M St George 1, Dawn K Wilson 2, M Lee Van Horn 3
PMCID: PMC7517616  NIHMSID: NIHMS1621937  PMID: 29705935

Abstract

This study examined the effects of a family-based health promotion intervention on the moderate-to-vigorous physical activity (MVPA), light physical activity (LPA), sedentary behavior, and fruit and vegetable intake of African American parents. Eighty-nine African American parents (41.5±8.5 yrs; 92% females; 74% obese; 64% < $40K income) and adolescents (12.5±1.4 yrs; 61% girls; 48% obese) were randomized to a six-week behavioral skills plus positive parenting and peer monitoring intervention grounded in Social Cognitive, Self-Determination, and Family Systems Theories or a general health comparison program. Parents wore accelerometers for seven days and completed three 24-hour dietary recalls at baseline and post-intervention. Multilevel regression models (controlling for baseline variables) demonstrated a significantly greater increase in parent MVPA for those in the intervention versus comparison condition (b=9.44, SE=4.26, p<.05). There were no other significant effects. Family-based approaches that include African American parents and youth may increase parent MVPA and hold promise for preventing chronic diseases.

Keywords: family-based, health promotion, intervention, African American, parents, physical activity

African Americans are at disproportionately greater risk for developing and/or dying from numerous preventable chronic diseases, including obesity, cancer, cardiovascular disease, and diabetes, than are non-Hispanic Whites (American Cancer Society, 2016; Centers for Disease Control and Prevention, 2017; Mozaffarian et al., 2016). Engaging in regular physical activity (PA), reducing sedentary behavior, and consuming a healthy diet have been shown to attenuate risks associated with chronic diseases (e.g., Biswas et al., 2015; Pandey et al., 2015). However, data from the National Health and Nutrition Examination Survey indicate that just under 8% of African American adults meet PA guidelines according to accelerometry estimates (Tucker, Welk, & Beyler, 2011), and only 15% and 6% consume the recommended daily servings of fruits and vegetables, respectively (Kirkpatrick, Dodd, Reedy, & Krebs-Smith, 2012). Furthermore, it has been estimated that African American men and women spend over 80% of their waking time sedentary (Hooker et al., 2016). Developing efficacious health promotion interventions for African American adults is thus an important national public health priority.

Interventions that target healthy lifestyle behaviors, including PA, sedentary behavior, and diet, often involve the family system – namely, parents and their children (Kitzman-Ulrich et al., 2010; Wilson, Sweeney, Kitzman-Ulrich, Gause, & St. George, 2017). Despite the traditional focus on youth outcomes, involving parents in youth health promotion efforts may be mutually beneficial for parents and children. Previous family-based healthy lifestyle interventions have shown improvements in the physical activity, sedentary behavior, and/or fruit and vegetable consumption of participating parents (e.g., Anderson, Newby, Kehm, Barland, & Hearst, 2015; Haire-Joshu et al., 2008). In one intervention study, an increase in parent fruit and vegetable consumption significantly predicted an increase in child fruit and vegetable intake (Haire-Joshu et al., 2008). Consistent with Social Cognitive Theory (Bandura, 2004), these effects may be explained by improved self-regulation for both parents and youth as well as increased parental modeling and social support, which have been associated with healthy lifestyle behaviors (Buckley, Cohen, Kramer, McAuley, & Mullen, 2014; McClain, Chappuis, Nguyen-Rodriguez, Yaroch, & Spruijt-Metz, 2009). For example, in a sample that included nearly 2,500 girls (50% African American) followed for 10 years, Madsen and colleagues found that girls who perceived their parents to be physically active were 50% more active than those who perceived their parents to be sedentary (Madsen, McCulloch, & Crawford, 2009). Overall, given the majority of U.S. adults are failing to meet national recommendations for healthy lifestyle behaviors, family-based approaches may facilitate parents in making their own lifestyle changes to model for and support their children’s health behavior change efforts.

Previous family-based health promotion interventions developed specifically for African American families have been delivered using either separate or joint sessions for parents and children, have included culturally-relevant content (e.g., music, dance, food/recipes), and have targeted behavioral skill-building (e.g., goal-setting, self-monitoring, positive reinforcement) and changes to the home environment (e.g., increasing the availability of healthy foods, decreasing screen time) (e.g., Baranowski et al., 2003; Klesges et al., 2010; Robinson et al., 2010; Stolley & Fitzgibbon, 1997; Wilson, Kitzman-Ulrich, et al., 2015). As noted in a review of family-focused interventions targeting physical activity, diet, and obesity interventions in African American girls and their parents, while these interventions generally show promise for youth outcomes, particularly when participating family members are encouraged to change their own health behaviors (Barr-Anderson, Adams-Wynn, DiSantis, & Kumanyika, 2013), very little is known regarding caregiver outcomes. Of the 27 studies reviewed by Barr-Anderson et al. (2013), only 37% targeted a change in the participating family members’ own behaviors, and these effects do not seem to be consistently reported in the literature. It is therefore important to target, assess, and report these effects so as to more fully appreciate the intervention’s impact on the family system.

Project SHINE (Supporting Health Interactively through Nutrition and Exercise) is a six-week family-based behavioral skills plus positive parenting and peer monitoring intervention for African American adolescents and their caregivers that targeted improvements in PA, sedentary behavior, and diet (St. George, Wilson, McDaniel, & Alia, 2016; St. George, Wilson, Schneider, & Alia, 2013). The intervention significantly reduced adolescent self-reported sedentary behavior for those randomized to the intervention condition compared to those in a general health education comparison condition (St. George et al., 2013). There was also a significant interaction effect between the intervention condition and parent-adolescent communication predicting adolescent self-reported sedentary behavior. Specifically, parents in the intervention condition who reported more positive communication around health behaviors had adolescents who reported lower sedentary behavior at post-intervention than did those with less positive communication or those in the comparison condition (St. George et al., 2013). No other intervention effects were observed for youth participating in the study. This paper extends previous work by evaluating the effects of Project SHINE on parent healthy lifestyle behaviors, specifically moderate-to-vigorous PA (MVPA), light PA (LPA), sedentary behavior, and fruit and vegetable intake. It was hypothesized that parents in to the intervention condition would increase their MVPA, LPA, and fruit and vegetable intake and reduce their sedentary behavior from baseline to post-intervention relative to those in a general health education comparison condition.

Methods

Participants

Participants were African American caregiver-adolescent dyads (n = 89) who volunteered to participate. Families were eligible if: (a) they had an 11-15 year old African American adolescent, and (b) a primary caregiver living in the same household as the adolescent willing to participate. Families were excluded if: (a) they were enrolled in a health promotion program, (b) the adolescent had a chronic disease, psychiatric condition or developmental delay, or (c) the parent or adolescent had a physical disability (e.g., wheelchair bound) or medical condition that would interfere with making changes to their physical activity levels. Parents who responded “no” to a question about whether they were physically capable of going on a brisk 30-minute walk were excluded from the study.

The study CONSORT diagram is depicted in Figure 1. Participants were recruited using convenience sampling at community events, through advertisements (e.g., flyers, radio), or through referrals from pediatric clinics and school nurses. Passive consent procedures were also implemented at a university-affiliated pediatric clinic. Letters mailed by the clinic informed patients of the study and provided directions for opting out of having their contact information released to study staff. Families who did not opt out within two weeks were contacted. Approval from the University of South Carolina IRB was obtained, and HIPAA procedures for patient privacy were followed. Parent consent and adolescent assent were obtained before data collection.

Figure 1.

Figure 1.

Open in a new tab

Participant flow diagram based on the Consolidated Standards for Reporting Trials (CONSORT)

Study Design and Procedures

Project SHINE used a multiple cohort design that included ten groups of 5-12 families across five cohorts (one intervention, one comparison group per cohort). Families were stratified by adolescent sex and weight status and randomly assigned to one of two evenings (Tuesdays or Thursdays) by the first author using a digital coin flip (www.random.org). Evenings were then randomized to a condition by a staff member blind to each family’s assigned evening using a digital coin flip. This process ensured allocation concealment.

All families participated in a two-week “run-in” period prior to being randomized to a condition, a strategy previously shown to maximize retention (Ulmer, Robinaugh, Friedberg, Lipsitz, & Natarajan, 2008). Attendance at the two run-in sessions was required (though make-up sessions were permitted) for continued participation throughout the remaining six weeks. During the run-in, facilitators provided an overview of the program, and a local African American dietitian provided education on national recommendations for healthy lifestyle behaviors (e.g., 60 minutes/day of PA for youth). All group sessions for both the intervention and comparison conditions were held at a university setting and led by two facilitators who were graduate students in clinical psychology or public health (n=6, 83% female, 50% African American, 33% Caucasian, 17% Hispanic). To reduce facilitator effects, the two facilitators who led intervention group sessions also led the concurrently occurring comparison group sessions across all five cohorts. Prior to leading sessions, facilitators completed 10 hours of face-to-face training that included didactics, group discussions, role-plays, and activities related to the theoretical model. Lead facilitators for each group were doctoral students in clinical psychology and received required cultural competency training as part of their degree program. Trained staff (blind to randomization) collected all measures at baseline (Week 0) and immediately following the six-week intervention period (Week 8). Participants received $20 at each measurement time point.

SHINE Intervention Condition

A detailed intervention description, including a curriculum matrix, essential elements, and process evaluation data, has been previously published (St. George et al., 2016; St. George et al., 2013). The theoretical framework used to develop the intervention integrated behavioral strategies from Social Cognitive Theory (e.g., self-monitoring, goal-setting), elements involved in facilitating intrinsic motivation from Self-Determination Theory (i.e., autonomy, competence, belongingness), and positive parenting practices from Family Systems Theory (e.g., parent-adolescent communication, parental monitoring of peers) (St. George et al., 2013; Wilson et al., 2017). The intervention conceptual framework is depicted in Figure 2. Overall, the intervention targeted behavioral, motivational, and parenting variables through interactive sessions designed to teach parents and adolescents how to change their healthy lifestyle behaviors and parents how to refine their positive parenting practices (e.g., communication skills, social support, autonomy support, parental monitoring and management of peer relationships) to support adolescent health behavior change. Cultural targeting strategies were used to enhance the intervention’s relevance for African American families (Kreuter, Lukwago, Bucholtz, Clark, & Sanders-Thompson, 2003). Specifically, the intervention used peripheral strategies (e.g., featuring photos of African American families in study materials), evidential strategies (e.g., providing national data related to African American youth), and constituent-involving strategies (e.g., having an African American dietician and community leader as outside speakers).

Figure 2.

Figure 2.

Open in a new tab

Project SHINE intervention conceptual framework

Parents and adolescents jointly attended six 1.5-hour weekly face-to-face group sessions. Each week, facilitators challenged parents and adolescents to work on a healthy lifestyle behavior of their choice (either PA, sedentary behavior, or diet-related) by self-monitoring, setting goals, and implementing new skills learned. Facilitators met individually with families for 5-15 minutes immediately before or after weekly group sessions to review progress towards their goals by discussing relevant barriers (e.g., “What, if anything, got in the way of you meeting your goal this week?”) and reinforcing change through action planning (e.g., “What do you think you can do differently this week to meet your goal?). These brief, individualized feedback sessions were part of the process of greeting families as they arrived to group sessions.

During the first group session, facilitators reviewed positive communication skills (e.g., active listening, using “I” statements, taking turns making brief statements, using a neutral tone of voice, making eye contact), and families collaboratively set group ground rules (e.g., “allow time for everyone to express their point,” “listen with an open mind”) to reinforce communication skills. Families were also encouraged to use positive communication skills to discuss how they were doing with regard to their healthy lifestyle behaviors during a “Family Walk & Talk.” During the second and third group sessions, parents and adolescents participated in separate discussions related to how family members and friends influenced healthy lifestyle behaviors. Families learned strategies for effective problem solving (e.g., define the problem, brainstorm all possible solutions, make a joint decision, discuss a plan for follow through) and were encouraged to negotiate two family rules related to their ongoing behavior change goals. Parents were also provided with specific behaviorally-based parenting strategies (e.g., descriptive praise, shared decision-making) associated with supporting their adolescent in meeting behavior change goals. The fourth group session emphasized the importance of peer relationships during adolescence and the role parents play in managing peer relationships around healthy lifestyle behaviors. Adolescents were encouraged to bring a friend to this session, and friends were integrated into activities. During the fifth group session, families examined their progress and readjusted their goals as needed. A local African American community leader and author specializing in youth development shared his personal health testimonial and discussed parenting concepts (e.g., parental monitoring) from his book on parenting. The sixth and final group session focused on maintaining health behavior changes through relapse prevention strategies. “Family Bonding Activities” (e.g., engage in a healthy activity together with your family and a friend) were assigned as weekly take-home activities to reinforce skill building.

Participants in the intervention condition attended an average of 4.39 ± 1.51 sessions (not including make-up sessions) and 4.92 ± 1.48 sessions (including make-up sessions) (St. George et al., 2016). Ratings of facilitator adherence (i.e., how well intervention was delivered in accordance with the theoretical framework) and competence (i.e., clinical skill) made by a trained, independent evaluator, indicated that the intervention was delivered with high fidelity and have been described in detail elsewhere (St. George et al., 2016).

General Health Education Comparison Condition

The general health education comparison groups met weekly for six weeks following the two-week run-in period as an attention control for the intervention condition. Parents and adolescents jointly attended six 1.5-hour weekly face-to-face group sessions. Sessions covered topics including sleep, stress management, hypertension, positive self-concept, diabetes, and cancer prevention. Sessions were information-based (e.g., symptoms and long-term effects of diabetes), and facilitators encouraged group members to share their existing knowledge on the topics. Interactive activities were used to keep families engaged in the content (e.g., a “fact or fiction” quiz about sleep, relaxation exercise for dealing with stress). These activities promoted social interactions between families more so than within families, and there was no focus on behavioral or positive parenting skills in this condition. Notes were made to ensure that facilitators followed the session guide and did not cover behavioral and positive parenting skills. Participants in this condition attended an average of 4.31 ± 1.56 sessions (not including make-up sessions) and 4.79 ± 1.51 sessions (including make-up sessions).

Measures

Demographic data.

Demographic data were collected from parents at baseline and included items such as sex (male or female), date of birth, and household income. Participants reported income using the following response options: <$10,000; $10,000-24,999; $25,000-$39,999; $40,000-$54,999; $55,000-$69,999; $70,000-$84,999; ≥$85,000.

Anthropometric measures.

Two measures of height, weight, and waist circumference were taken on parents using a Shorr Height Measuring Board, SECA 880 digital scale, and tape measure, respectively. Average scores of height and weight were used in anthropometric calculations of body mass index (BMI).

Seven-day accelerometry estimates (MVPA, LPA, sedentary behavior).

Seven-day accelerometry estimates for MVPA, LPA, and sedentary behavior were assessed with the Actical omni-directional accelerometers worn on the right hip. Actical has been previously validated as a measure of adult PA (Trumpeter et al., 2012). Data were recorded in one-minute epochs, and consistent with previous PA trials with large samples of African American adults, 60 minutes of consecutive zeros were used to define non-wear (Wilcox et al., 2013; Wilson, Van Horn, et al., 2015). Raw activity data were converted into time spent in MVPA, LPA, and sedentary behavior based on Actical-specific count thresholds for adults (sedentary behavior<100, LPA=100 to <1,075, and MVPA≥1,075) (Trumpeter et al., 2012).

Multiple imputation methods (see statistical analysis) were applied to missing data (Catellier et al., 2005; Xu et al., 2016). Days were divided into time intervals (6-9am, 9am-2pm, 2-5pm, 5-8pm, 8pm-12am) to enhance the precision of the imputation procedures by providing additional information for the imputation model (i.e., correlations in PA between time blocks across days) that would otherwise be absent. This has been done in previous PA trials using multiple imputation procedures with accelerometry data (Wilson et al., 2011; Wilson, Van Horn, et al., 2015). Despite the use of these time intervals for the imputation procedures, the outcomes remained average overall daily minutes of PA across different intensities. Valid wear was defined as time intervals during which the accelerometer was worn for at least 80% of the time interval (Catellier et al., 2005; Wilson, Van Horn, et al., 2015). Otherwise stated, a time block for a given participant was considered missing if 20% of that time block was non-wear After imputation, minutes per day of MVPA, LPA, and sedentary behavior were calculated by summing time points for each day and averaging the seven days.

Fruit and vegetable intake.

The Automated Self-Administered 24-hour Recall (ASA24), a free online dietary tool modeled after the validated Automated Multiple Pass Method, was used to collect dietary data (Subar et al., 2012). Because the ASA24 youth version (released September 2012) was unavailable for this study, administration protocols using the adult self-administered “Beta” and “ASA24-2011” versions were modified for use with families. Specifically, rather than having participants enter their own dietary information, trained and certified study staff contacted adolescents and parents (separately) via phone on three random days (two weekdays, one weekend day) and entered dietary information on their behalf. Prior to conducting recalls, staff completed extensive training on the ASA24, including a live demo and practice sessions.

Statistical Analysis

Missing Data.

Missing data were dealt with using multiple imputation (m=20) (Schafer & Olsen, 1988). Analyses with multiply imputed data involve a three-step process wherein multiple data sets are first generated, analyst models are estimated separately for each of the data sets, and pooled estimates of parameters and standard errors across analyst models are computed (Acock, 2005). Because this study contains a nested data structure (individuals within groups), an imputation function that modelled multilevel data was used (van Buuren & Groothuis-Oudshoorn, 2011). Specifically, the “mice.impute.2l.norm” function of the package “mice” in the statistical program R (Version 2.15.1) allowed for specification of the class variable (i.e., group) as well as inclusion of a random effect in each of the imputation models. Importantly, participants’ data are weighted according to the amount of valid information in the imputation model; as such, information from those who provide less data is down-weighted. Plots of imputed values indicated adequate convergence, and fractions of missing information are reported.

Accelerometry data were imputed in two phases at each time point to enhance the precision of the estimates. In the first phase, data were imputed at the interval level. Participants who had complete missing data were thus removed from this phase. In the second phase, those who had complete missing data were added back in to the dataset and their data were imputed at the summary level instead of the interval level. This process resulted in 20 datasets (1 per imputation) with complete data for all 89 participants for our analyses. Accelerometry data from 83 parents and 70 parents were included in the phase 1 imputations at baseline and post-intervention, respectively. For those whose data was included in the phase 1 imputations, the percentage of interval-level accelerometry data missing due to non-compliance was 13% for parents at baseline and 20% for parents at post-intervention. These rates of missing data due to non-compliance in studies that have used multiple imputation procedures are similar to those found in previous PA trials conducted with African American samples (Catellier et al., 2005; Wilson, Lawman, Segal, & Chappell, 2011).

In terms of dietary recalls, 62% of parents completed at least three recalls at baseline, and 57% from the full sample did so at post-intervention. There were no significant differences in outcomes of interest at baseline for those missing data at post-intervention.

Intervention Effects.

Random intercept multilevel regression models, controlling for baseline demographic variables (i.e., sex, age, income, BMI), cohort, and the baseline value of the outcome of interest, were used to test study outcomes. Variables were either contrast coded (sex, cohort), mean centered (age, income, BMI, baseline value of outcome), or dummy coded (treatment; 0=comparison, 1=intervention) to facilitate model interpretation, with the intercept representing the mean of the outcome across groups for the average parent in the intervention condition. All analyses were conducted using R (Version 2.15.1).

Results

Participant Characteristics at Baseline

See Table 1 for participant demographic characteristics by condition. Parents were 41.52 ± 8.54 years old, primarily female (92%), obese (74%), and did not have a college degree (60%). The average adolescent was 12.53 ± 1.42 years old, female (61%) and above the 85th percentile for BMI (13% overweight, 48% obese). There were no significant demographic differences between conditions at baseline.

Table 1.

Participant Demographic Characteristics

Variable Intervention Comparison Total
Sample Size, n (%) 49 (55%) 40 (45%) 89 (100%)
Sex, n (%)
 Female 44 (90%) 38 (95%) 82 (92%)
 Male 5 (10%) 2 (5%) 7 (8%)
Age (yrs; M ± SD) 42.51 ± 8.96 40.30 ± 7.93 41.52 ± 8.54
Body Mass Index (BMI; M ± SD) 36.03 ± 9.17 36.72 ± 9.42 36.34 ± 9.24
Weight Status, n (%)
 Underweight (BMI < 18.5) 1 (2%) 0 (0%) 1 (1%)
 Normal Weight (BMI 18.5 – 24.9) 5 (10%) 3 (8%) 8 (9%)
 Overweight (BMI 25.0 – 29.9) 6 (12%) 8 (20%) 14 (16%)
 Obese (BMI ≥ 30.0) 37 (76%) 29 (73%) 66 (74%)
Waist Circumference (cm; M ± SD) 105.30 ± 19.99 105.13 ± 20.62 105.22 ± 20.16
Relationship to Adolescent, n (%)
 Mother 41 (84%) 35 (88%) 76 (85%)
 Father 5 (10%) 1 (3%) 6 (7%)
 Other 3 (6%) 4 (10%) 7
Relationship Status, n (%)
 Married 16 (33%) 16 (40%) 32 (36%)
 Separated or Divorced 16 (32%) 7 (18%) 23 (25%)
 Widowed 1 (2%) 0 (0%) 1 (1%)
 Never Married or In Unmarried Couple 16 (33%) 17 (43%) 33 (37%)
Education, n (%)
 Some High School 1 (2%) 0 (0%) 1 (1%)
 High School Degree or GED 6 (12%) 5 (13%) 11 (12%)
 Some College 25 (51%) 17 (43%) 42 (47%)
 College Graduate 12 (24%) 9 (23%) 21 (24%)
 Graduate Training or Professional Degree 5 (10%) 9 (23%) 14 (16%)
Household Yearly Income, n (%)
 <$10,000 6 (12%) 6 (15%) 12 (13%)
 $10,000 to $24,999 10 (20%) 7 (18%) 17 (19%)
 $25,000 to $39,999 17 (35%) 11 (28%) 28 (31%)
 $40,000 to $54,999 8 (16%) 3 (8%) 11 (12%)
 $55,000 to $69,999 5 (10%) 2 (5%) 7 (8%)
 $70,000 or more 3 (6%) 11 (28%) 14 (16%)
Household Size (# people; M ± SD) 3.76 ± 1.56 3.58 ± 1.43 3.67 ± 1.50
Open in a new tab

Note. Frequency totals may not equal 100% due to rounding. There were no significant between-group differences on demographic characteristics at baseline.

See Table 2 for participant outcome variables by condition. At baseline, parents in the intervention vs. comparison condition engaged in 19.0 ± 1.9 vs. 20.9 ± 2.5 minutes/day of MVPA, 194.6 ± 10.7 vs. 186.7 ± 42.8 minutes/day of LPA, 743.8 ± 13.1 vs. 778.1 ± 11.4 minutes/day of sedentary behavior, and consumed 1.0 ± 0.2 vs. 0.8 ± 0.2 cups of fruit and 1.3 ± 0.2 vs. 1.6 ± 0.3 cups of vegetables, respectively.

Table 2.

Mean Levels of Outcome Variables by Condition for Parents

Intervention Condition (n=49)
Baseline (Week 0) Post-Intervention (Week 8)
Mean (SE) 95% CI Mean (SE) 95% CI
MVPA (min/day) 19.0 (1.9) 15.2- 22.8 27.3 (2.6) 22.1-32.4
LPA (min/day) 194.6 (10.7) 173.5- 215.64 189.0 (9.3) 170.7-207.2
Sedentary behavior (min/day) 743.8 (13.1) 718.1- 769.4 738.4 (16.6) 705.7-771.1
Fruits (cups) 1.0 ( 0.2) 0.6 -1.3 0.9 (0.3) 0.4-1.4
Vegetables (cups) 1.3 (0.2) 1.0 - 1.6 1.3 (0.2) 1.1-1.6
Comparison Condition (n = 40)
MVPA (min/day) 20.9 (2.5) 16.0- 25.8 18.8 (3.6) 11.7- 25.8
LPA (min/day) 186.7 (42.8) 101.4- 271.9 169.0 (15.3) 139.0- 199.0
Sedentary beahvior (min/day) 778.1 (11.4) 755.8- 800.4 753.8 (14.3) 725.6- 781.9
Fruits (cups) 0.8 (0.2) 0.5- 1.2 1.0 (0.3) 0.3- 1.6
Vegetables (cups) 1.6 (0.3) 0.9- 2.3 1.5 (0.3) 1.0- 2.0
Open in a new tab

Note. MVPA = moderate-to-vigorous PA; LPA = light physical activity. To adjust for clustering within groups and for use of multiple imputation procedures, a series of unconditional random intercept models were used to calculate means and standard errors (SE) for MVPA, LPA, sedentary behavior, fruit intake, and vegetable intake.

Intervention Effects on Parent Outcomes

MVPA.

At post-intervention, parents in the intervention vs. comparison condition engaged in 27.3 ± 2.6 vs. 18.8 ± 3.6 minutes/day of MVPA, respectively. This reflects an average increase of 8.3 minutes/day of MVPA in the intervention condition and an average decrease of 2.1 minutes/day in the comparison condition from baseline to post-intervention. Multilevel regression analyses indicated a significant effect of the intervention on parent MVPA (Table 3, Figure 2), such that those in the intervention condition engaged in greater minutes per day of MVPA at post intervention (controlling for baseline MVPA) than did those in the comparison condition (b=9.44, SE=4.26, t=2.21, p < 0.05). No control variables in this model were significantly associated with parent MVPA.

Table 3.

Multilevel Model Predicting Parent MVPA

Parameter Estimate (SE) df p FMI Lower CI Upper CI
Intercept 19.30** (4.24) 89 0.00 0.17 10.97 27.62
Female −2.71 (6.05) 89 0.66 0.10 −14.56 9.16
Age −0.01 (0.20) 89 0.97 0.20 −0.41 0.39
Income 0.73 (1.21) 89 0.55 0.43 −1.67 3.13
Parent BMI 0.04 (0.19) 89 0.84 0.20 −0.33 0.40
Cohort1.con 1.31 (6.86) 89 0.85 0.21 −12.18 14.79
Cohort2.con −4.98 (6.67) 89 0.46 0.21 −18.08 8.12
Cohort3.con −4.60 (7.80) 89 0.56 0.31 −19.98 10.78
Cohort4.con 4.71 (7.13) 89 0.51 0.24 −9.31 18.74
Baseline MVPA 0.22 (0.14) 89 0.12 0.27 −0.06 0.493
Intervention 9.44* (4.26) 89 0.03 0.15 1.09 17.79
Open in a new tab

Note. MVPA = Moderate-to-vigorous physical activity; BMI = Body mass index; SE = Standard error of the parameter estimate adjusted for the use of multiple imputations; df = Estimated degrees of freedom adjusted for use of multiple imputations and capped at sample size; FMI = Fraction of missing information; CI = 95% confidence intervals

**

p < .01,

*

p < .05

LPA.

At post-intervention, parents in the intervention vs. comparison condition engaged in 189.0 ± 9.3 vs. 169.0 ± 15.3 minutes/day of LPA, respectively. This reflects an average decrease of 5.6 minutes/day of LPA in the intervention condition and of 17.7 minutes/day in the comparison condition from baseline to post-intervention. There was no significant effect of the intervention on parent LPA. Parents’ baseline LPA (b= 0.35, SE= 0.14, t=2.50, p < .05) was the only control variable significantly associated with LPA at post-intervention (Table 4).

Table 4.

Multilevel Models Predicting Parent LPA and Sedentary Behavior

Parameter Estimate (SE) df p FMI Lower CI Upper CI
Parent LPA
Intercept 168.48** 14.47 89 0.00 0.16 140.08 196.87
Female 7.80 21.92 89 0.72 0.12 −35.20 50.79
Age −0.26 0.75 89 0.72 0.19 −1.74 1.21
Income −1.89 3.75 89 0.61 0.25 −9.27 5.48
Parent BMI −0.44 0.73 89 0.54 0.32 −1.88 1.00
Cohort1.con −7.35 22.35 89 0.74 0.19 −51.26 36.55
Cohort2.con −17.36 21.24 89 0.41 0.16 −59.05 24.34
Cohort3.con 11.51 25.86 89 0.66 0.27 −39.40 62.41
Cohort4.con 20.14 23.83 89 0.40 0.26 −26.75 67.03
Baseline LPA 0.35* 0.15 89 0.02 0.60 0.06 0.64
Intervention 16.88 14.68 89 0.25 0.22 −11.98 45.74
Parent Sedentary Behavior
Intercept 440.56** (122.40) 89 0.00 0.43 197.97 683.14
Female 13.60 (34.06) 89 0.69 0.09 −53.20 80.40
Age −0.023 (1.26) 89 0.99 0.31 −2.50 2.46
Income −1.21 (6.20) 89 0.85 0.34 −13.45 11.03
Parent BMI 1.00 (1.16) 89 0.39 0.34 −1.29 3.29
Cohort1.con −28.04 (33.02) 89 0.40 0.27 93.05 36.96
Cohort2.con −10.01 (37.31) 89 0.79 0.47 −84.14 64.12
Cohort3.con −3.06 (36.73) 89 0.93 0.29 −75.43 69.31
Cohort4.con −44.17 (35.12) 89 0.21 0.32 −113.44 25.09
Baseline Sedentary Behavior 0.40* (0.15) 89 0.01 0.42 0.09 0.70
Intervention −2.99 (23.42) 89 0.90 0.36 −49.25 43.27
Open in a new tab

Note. LPA = light physical activity; BMI = Body mass index; SE = Standard error of the parameter estimate adjusted for the use of multiple imputations; df = Estimated degrees of freedom adjusted for use of multiple imputations and capped at sample size; FMI = Fraction of missing information; CI = 95% confidence intervals

**

p < .01,

*

p < .05

Sedentary behavior.

At post-intervention, parents in the intervention vs. comparison condition engaged in 738.4 ± 16.6 vs. 753.8 ± 14.3 minutes/day of sedentary behavior, respectively. This reflects an average increase of 5.4 minutes/day of sedentary behavior in the intervention condition and of 24.3 minutes/day in the comparison condition from baseline to post-intervention. There was no significant effect of the intervention on parent sedentary behavior. Parents’ baseline sedentary behavior (b=0.40, SE=0.15, t=2.67, p < 0.05) was the only control variable significantly associated with sedentary behavior at post-intervention (Table 4).

Fruit and vegetable intake.

At post-intervention, parents in the intervention vs. comparison condition consumed 0.9 ± 0.3 vs. 1.0 ± 0.3 cups of fruit and 1.3 ± 0.2 vs. 1.5 ± 0.3, cups of vegetables, respectively. For fruit intake, this reflects an average decrease of 0.1 cups in the intervention condition and an average increase of 0.2 cups in the comparison condition from baseline to post-intervention. For vegetable intake, this reflects no change in intake in the intervention condition from baseline to post-intervention and a decrease of 0.1 cups in the comparison condition. There were no significant effects of the intervention on either parent fruit or vegetable intake. Parents’ baseline vegetable intake was significantly associated with vegetable intake at post-intervention (b=0.43, SE=0.11, t=3.91, p < 0.01) (Table 5). No other control variables were significantly associated with parents’ fruit or vegetable intake at post-intervention.

Table 5.

Multilevel Models Predicting Parent Fruit and Vegetable Intake

Parameter Estimate (SE) df p FMI Lower CI Upper CI
Parent Fruit Intake
Intercept 0.60 0.43 89 0.16 0.41 −0.25 1.44
Female 0.03 0.48 89 0.95 0.10 −0.91 0.97
Age 0.00 0.02 89 0.93 0.18 −0.03 0.03
Income −0.01 0.08 89 0.95 0.28 −0.17 0.16
Parent BMI 0.01 0.02 89 0.45 0.35 −0.02 0.04
Cohort1.con −0.47 0.59 89 0.43 0.43 −1.65 0.70
Cohort2.con −0.78 0.60 89 0.19 0.47 −1.97 0.41
Cohort3.con −0.73 0.82 89 0.38 0.66 −2.38 0.93
Cohort4.con −1.17 0.63 89 0.07 0.45 −2.41 0.07
Baseline Fruit Intake 0.45 0.26 89 0.09 0.56 −0.07 0.97
Intervention −0.14 0.36 89 0.69 0.35 −0.85 0.57
Parent Vegetable Intake
Intercept 0.92** 0.29 89 0.00 0.21 0.35 1.49
Female −0.29 0.36 89 0.42 0.15 −1.00 0.41
Age 0.00 0.01 89 0.99 0.26 −0.02 0.02
Income −0.00 0.06 89 0.95 0.20 −0.12 0.11
Parent BMI 0.00 0.01 89 0.87 0.26 −0.02 0.02
Cohort1.con 0.14 0.41 89 0.74 0.23 −0.66 0.94
Cohort2.con 0.28 0.39 89 0.48 0.20 −0.48 1.04
Cohort3.con 0.24 0.46 89 0.60 0.34 −0.68 1.16
Cohort4.con 0.29 0.40 89 0.47 0.20 −0.50 1.09
Baseline Vegetable Intake 0.43 0.11 89 0.00 0.44 0.20 0.66
Intervention −0.02 0.25 89 0.94 0.15 −0.51 0.47
Open in a new tab

Note. BMI = Body mass index; SE = Standard error of the parameter estimate adjusted for the use of multiple imputations; df = Estimated degrees of freedom adjusted for use of multiple imputations and capped at sample size; FMI = Fraction of missing information; CI = 95% confidence intervals

**

p < .01,

*

p < .05

Discussion

The present study tested the effects of a family-based behavioral skills plus positive parenting and peer monitoring intervention on the MVPA, LPA, sedentary behavior, and fruit and vegetable intake of African American parents. The intervention resulted in improvements in parent MVPA at post-intervention after controlling for baseline MVPA, such that those in the intervention condition reported approximately nine more minutes per day of MVPA at post-intervention than did those in the comparison condition. No other effects were significant. This study provides preliminary support of a family-based intervention for improving parent accelerometer-assessed MVPA in African American families.

Findings from this study highlight the potentially beneficial effects of family-based health promotion interventions on African American adults and the importance of assessing changes in both youth and adult health behaviors. For adults in the intervention, 76% of whom were in the obese weight range, MVPA increased by nearly a 10-minute bout per day relative to those in the comparison condition. A pooled analysis of data from >650,000 adults in the National Cancer Institute Cohort Consortium showed that compared to no physical activity, a physical activity level equivalent to brisk walking for approximately 10 minutes/day was associated with 1.8 additional years of life expectancy (Moore et al., 2012). In the present study, the mean minutes per week of MVPA for parents in the intervention condition at post-intervention (27.3 minutes/day x 7 days/week = 191.1 minutes/week) was also consistent with meeting the recommended 150 weekly minutes of moderate intensity PA for adults as outlined by the 2008 Physical Activity Guidelines for Americans (Physical Activity Guidelines Committee, 2008).

In light of parent increases in MVPA, findings from this study have important implications for chronic disease prevention in African American adults, particularly in African American women who are 2.10 times more likely than non-Hispanic white women to be obese (Flegal, Kruszon-Moran, Carroll, Fryar, & Ogden, 2016). Increasing PA has been strongly associated with preventing chronic diseases including cardiovascular disease (Bell, Lutsey, Windham, & Folsom, 2013) and type 2 diabetes (Aune, Norat, Leitzmann, Tonstad, & Vatten, 2015). For example, in a large sample of adults from the Atherosclerosis Risk in Communities cohort study that included over 3,000 African Americans, PA had a strong inverse association with cardiovascular disease, heart failure, coronary heart disease, and stroke incidence in African Americans (Bell et al., 2013). Although data indicated that some PA was better than none, hazard ratios for incident cardiovascular disease were lower as levels of PA increased (i.e., 0.59 for recommended levels of PA, 0.65 for intermediate physical activity, 1.00 for poor physical activity). Authors noted that these associations were present regardless of participant obesity, diabetes, hypertension, and dyslipidemia. Similarly, for participants followed for ten years as part of the Black Women’s Health Study (n=45,668), both vigorous activity and brisk walking were inversely associated with type 2 diabetes risk (Krishnan, Rosenberg, & Palmer, 2008).

The intervention had no effect on other parent outcomes. Because this was a small study, power to detect significant intervention effects was limited at the outset. Post-hoc power analyses using standard errors found in models with non-significant effects indicated the intervention had 80% power to detect only large effects, which are typically not observed in behavioral interventions (Webber et al., 2008). Furthermore, low reliability of dietary recall measures may have further contributed to low power (St. George, Van Horn, Lawman, & Wilson, 2016).

Limitations of the current study include the short intervention duration, lack of follow up, small sample size, and use of convenience sampling. Generalizability of findings may thus be limited. Follow-up data may have been especially beneficial for monitoring effects over time. These limitations aside, research examining family-based health promotion interventions in ethnic minorities is limited, and even fewer randomized controlled trials have assessed outcomes in youth and parents alike. This study shows promise for improving accelerometry-assessed PA in African American adults through a family-based approach and potentially reducing the risk of chronic diseases such as diabetes. The intervention is novel in that it used a combination of behavioral and positive parenting skills for health behavior change. Another notable strength of this study is the use of rigorous methods (i.e., randomization, multiple imputation, multilevel models, 7-day accelerometry estimates). The intervention has already been adapted in a large-scale randomized trial known as Families Improving Together (FIT) for weight loss (Wilson, Kitzman-Ulrich, et al., 2015) that is testing a positive parenting intervention on reducing zBMI in African American adolescents and their parents. Findings from that trial are forthcoming. Overall, longitudinal trials are needed to replicate the PA-related findings from the present study and to better determine which program components are most effective in promoting MVPA in ethnic minority parents.

Figure 3.

Figure 3.

Open in a new tab

Parent MVPA at baseline and post-intervention by condition

Acknowledgments

Funding: This study was funded by the National Institute of Child Health and Human Development (F31 HD 066944 to Sara M. St. George, Ph.D. and Dawn K. Wilson, Ph.D.; R01 HD 072153 to Dawn K. Wilson, Ph.D.).

Footnotes

Disclosure of potential conflicts of interest

The authors declare that they have no conflict of interest.

Research involving Human Participants

Ethical approval. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This study does not contain any studies with animals performed by any of the authors.

Informed consent

Informed consent was obtained from all individual participants included in the study.

References

  1. Acock AC (2005). Working with missing values. Journal of Marriage and Family, 67(4), 1012–1028. [Google Scholar]
  2. American Cancer Society. (2016). Cancer Facts & Figures for African Americans 2016–2018. Atlanta, GA: American Cancer Society. [Google Scholar]
  3. Anderson JD, Newby R, Kehm R, Barland P, & Hearst MO (2015). Taking steps together: a family-and community-based obesity intervention for urban, multiethnic children. Health Education and Behavior, 42(2), 194–201. [DOI] [PubMed] [Google Scholar]
  4. Aune D, Norat T, Leitzmann M, Tonstad S, & Vatten LJ (2015). Physical activity and the risk of type 2 diabetes: a systematic review and dose–response meta-analysis: Springer. [DOI] [PubMed] [Google Scholar]
  5. Bandura A (2004). Health promotion by social cognitive means. Health Education and Behavior, 31(2), 143–164. [DOI] [PubMed] [Google Scholar]
  6. Baranowski T, Baranowski J, Cullen K, Thompson D, Nicklas T, Zakeri I, & Rochon J (2003). The fun, food, and fitness project (FFFP): The Baylor GEMS pilot study. Ethnicity and Disease, 13(1; SUPP/1), 1–30. [PubMed] [Google Scholar]
  7. Barr-Anderson DJ, Adams-Wynn AW, DiSantis KI, & Kumanyika S (2013). Family-focused physical activity, diet and obesity interventions in African-American girls: a systematic review. Obesity Reviews, 14(1), 29–51. doi: 10.1111/j.1467-789X.2012.01043.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bell EJ, Lutsey PL, Windham BG, & Folsom AR (2013). Physical activity and cardiovascular disease in African Americans in ARIC. Medicine and Science in Sports and Exercise, 45(5), 901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Biswas A, Oh PI, Faulkner GE, Bajaj RR, Silver MA, Mitchell MS, & Alter DA (2015). Sedentary time and its association with risk for disease incidence, mortality, and hospitalization in adults: A systematic review and meta-analysis. Annals of Internal Medicine, 162(2), 123–132. [DOI] [PubMed] [Google Scholar]
  10. Buckley J, Cohen JD, Kramer AF, McAuley E, & Mullen SP (2014). Cognitive control in the self-regulation of physical activity and sedentary behavior. Frontiers in Human Neuroscience, 8, 747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Catellier DJ, Hannan PJ, Murray DM, Addy CL, Conway TL, Song Y, & Rice JC (2005). Imputation of missing data when measuring physical activity by accelerometry. Medicine and Science in Sports and Exercise, 37(11 Suppl), S555. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Centers for Disease Control and Prevention. (2017). National Diabetes Statistics Report. Atlanta, GA: Centers for Disease Control and Prevention, U.S. Dept of Health and Human Services. [Google Scholar]
  13. Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, & Ogden CL (2016). Trends in obesity among adults in the United States, 2005 to 2014. JAMA, 315(21), 2284–2291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Haire-Joshu D, Elliott MB, Caito NM, Hessler K, Nanney MS, Hale N, … Brownson RC (2008). High 5 for Kids: the impact of a home visiting program on fruit and vegetable intake of parents and their preschool children. Preventive Medicine, 47(1), 77–82. doi: 10.1016/j.ypmed.2008.03.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hooker SP, Hutto B, Zhu W, Blair SN, Colabianchi N, Vena JE, … Howard VJ (2016). Accelerometer measured sedentary behavior and physical activity in white and black adults: the REGARDS study. Journal of Science and Medicine in Sport, 19(4), 336–341. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kirkpatrick SI, Dodd KW, Reedy J, & Krebs-Smith SM (2012). Income and race/ethnicity are associated with adherence to food-based dietary guidance among US adults and children. Journal of the Academy of Nutrition and Dietetics, 112(5), 624–635. e626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kitzman-Ulrich H, Wilson DK, St. George SM, Lawman H, Segal M, & Fairchild A (2010). The integration of a family systems approach for understanding youth obesity, physical activity, and dietary programs. Clinical Child and Family Psychology Review, 13, 231–253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Klesges RC, Obarzanek E, Kumanyika S, Murray DM, Klesges LM, Relyea GE, … McClanahan BS (2010). The Memphis Girls' health Enrichment Multi-site Studies (GEMS): An evaluation of the efficacy of a 2-year obesity prevention program in African American girls. Archives of Pediatrics and Adolescent Medicine, 164(11), 1007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kreuter M, Lukwago S, Bucholtz D, Clark E, & Sanders-Thompson V (2003). Achieving cultural appropriateness in health promotion programs: targeted and tailored approaches. Health Education and Behavior, 30(2), 133. [DOI] [PubMed] [Google Scholar]
  20. Krishnan S, Rosenberg L, & Palmer JR (2008). Physical activity and television watching in relation to risk of type 2 diabetes: the Black Women's Health Study. American Journal of Epidemiology, 169(4), 428–434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Madsen KA, McCulloch CE, & Crawford PB (2009). Parent modeling: perceptions of parents' physical activity predict girls' activity throughout adolescence. The Journal of Pediatrics, 154(2), 278–283. e272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. McClain AD, Chappuis C, Nguyen-Rodriguez ST, Yaroch AL, & Spruijt-Metz D (2009). Psychosocial correlates of eating behavior in children and adolescents: a review. International Journal of Behavioral Nutrition and Physical Activity, 6(1), 54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Moore SC, Patel AV, Matthews CE, de Gonzalez AB, Park Y, Katki HA, … Helzlsouer KJ (2012). Leisure time physical activity of moderate to vigorous intensity and mortality: A large pooled cohort analysis. PLoS Medicine, 9(11), e1001335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, … Fullerton HJ (2016). Heart disease and stroke statistics—2016 update. Circulation, 133(4), e38–e360. [DOI] [PubMed] [Google Scholar]
  25. Pandey A, Garg S, Khunger M, Darden D, Ayers C, Kumbhani DJ, … Berry JD (2015). Dose response relationship between physical activity and risk of heart failure: a meta-analysis. Circulation, CIRCULATIONAHA. 115.015853. [DOI] [PubMed] [Google Scholar]
  26. Physical Activity Guidelines Committee. (2008). 2008 Physical Activity Guidelines for Americans. Washington D.C.: U.S. Department of Health and Human Services. [Google Scholar]
  27. Robinson TN, Matheson DM, Kraemer HC, Wilson DM, Obarzanek E, Thompson NS, … Killen JD (2010). A randomized controlled trial of culturally tailored dance and reducing screen time to prevent weight gain in low-income African American girls: Stanford GEMS. Archives of Pediatrics and Adolescent Medicine, 164(11), 995–1004. doi: 10.1001/archpediatrics.2010.197 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Schafer JL, & Olsen MK (1988). Multiple imputation for multivariate missing-data problems: A data analyst’s perspective. Multivariate Behavioral Research, 33, 545–571. [DOI] [PubMed] [Google Scholar]
  29. St. George SM, Van Horn ML, Lawman HG, & Wilson DK (2016). Reliability of 24-hour dietary recalls as a measure of diet in African-American youth. Journal of the Academy of Nutrition and Dietetics, 116(10), 1551–1559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. St. George SM, Wilson DK, McDaniel TM, & Alia KA (2016). Process evaluation of the Project SHINE intervention for African American families: An integrated positive parenting and peer monitoring approach to health promotion. Health Promotion Practice, 17, 557–568. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. St. George SM, Wilson DK, Schneider EM, & Alia KA (2013). Project SHINE: Effects of parent–adolescent communication on sedentary behavior in African American adolescents. Journal of Pediatric Psychology, 38, 997–1009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Stolley MR, & Fitzgibbon ML (1997). Effects of an obesity prevention program on the eating behavior of African American mothers and daughters. Health Education and Behavior, 24(2), 152–164. [DOI] [PubMed] [Google Scholar]
  33. Subar AF, Kirkpatrick SI, Mittl B, Zimmerman TP, Thompson FE, Bingley C, … McNutt S (2012). The Automated Self-Administered 24-hour dietary recall (ASA24): A resource for researchers, clinicians, and educators from the National Cancer Institute. Journal of the Academy of Nutrition and Dietetics, 112(8), 1134–1137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Trumpeter NN, Lawman HG, Wilson DK, Pate RR, Van Horn ML, & Tate AK (2012). Accelerometry cut points for physical activity in underserved African Americans. International Journal of Behavioral Nutrition and Physical Activity, 9(73), 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Tucker JM, Welk GJ, & Beyler NK (2011). Physical activity in US adults: compliance with the physical activity guidelines for Americans. American Journal of Preventive Medicine, 40(4), 454–461. [DOI] [PubMed] [Google Scholar]
  36. Ulmer M, Robinaugh D, Friedberg JP, Lipsitz SR, & Natarajan S (2008). Usefulness of a run-in period to reduce drop-outs in a randomized controlled trial of a behavioral intervention. Contemporary Clinical Trials, 29(5), 705–710. [DOI] [PubMed] [Google Scholar]
  37. van Buuren S, & Groothuis-Oudshoorn K (2011). MICE: Multivariate imputation by chained equations in R. Journal of Statistical Software, 45(3), 1–67. [Google Scholar]
  38. Webber L, Catellier D, Lytle L, Murray D, Pratt C, Young D, … Jobe J (2008). Promoting physical activity in middle school girls: Trial of Activity for Adolescent Girls. American Journal of Preventive Medicine, 34(3), 173–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Wilcox S, Parrott A, Baruth M, Laken M, Condrasky M, Saunders R, … Warren TY (2013). The Faith, Activity, and Nutrition Program: A randomized controlled trial in African American churches. American Journal of Preventive Medicine, 44(2), 122–131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wilson DK, Kitzman-Ulrich H, Resnicow K, Van Horn ML, George SMS, Siceloff ER, … Huffman L (2015). An overview of the Families Improving Together (FIT) for weight loss randomized controlled trial in African American families. Contemporary Clinical Trials, 42, 145–157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wilson DK, Lawman HG, Segal M, & Chappell S (2011). Neighborhood and parental supports for physical activity in minority adolescents. American Journal of Preventive Medicine, 41(4), 399–406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wilson DK, Sweeney AM, Kitzman-Ulrich H, Gause H, & St. George SM (2017). Promoting social nurturance and positive social environments to reduce obesity in high-risk youth. Clinical Child and Family Psychology Review, 20, 64–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Wilson DK, Van Horn ML, Kitzman-Ulrich H, Saunders R, Pate R, Lawman HG, … Addy (2011). Results of the “Active by Choice Today”(ACT) randomized trial for increasing physical activity in low-income and minority adolescents. Health Psychology, 30(4), 463–471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wilson DK, Van Horn ML, Siceloff ER, Alia KA, George SMS, Lawman HG, … Wandersman A (2015). The results of the “Positive Action for Today’s Health”(PATH) trial for increasing walking and physical activity in underserved African-American communities. Annals of Behavioral Medicine, 49(3), 398–410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Xu SY, Nelson S, Kerr J, Godbole S, Patterson R, Merchant G, … Natarajan L (2016). Statistical approaches to account for missing values in accelerometer data: Applications to modeling physical activity. Statistical Methods in Medical Research, 0962280216657119. [DOI] [PubMed] [Google Scholar]

RESOURCES