- Email: [email protected]
Social and Cultural Environment Factors Influencing Physical Activity Among African-American Adolescents
Journal of Adolescent Health 56 (2015) 536e542
www.jahonline.org Original article
Social and Cultural Environment Factors Influencing Physical Activity Among African-American Adolescents Monica L. Baskin, Ph.D. a, *, Akilah Dulin-Keita, Ph.D. b, Herpreet Thind, Ph.D., M.P.H. c, and Emily Godsey d a
Department of Medicine, Division of Preventive Medicine, UAB School of Medicine, Birmingham, Alabama Department of Behavioral and Social Sciences, Institute for Community Health Promotion, Brown University, Providence, Rhode Island c Department of Psychiatry and Human Behavior, Centers for Behavioral and Preventive Medicine, Brown Alpert Medical School and the Miriam Hospital, Providence, Rhode Island d Department of Sociology, University of Alabama at Birmingham, Birmingham, Alabama b
Article history: Received July 24, 2014; Accepted January 20, 2015 Keywords: Mixed methods; African-American; Youth; Physical activity
A B S T R A C T
Purpose: African-American youth are at high risk for physical inactivity. This study explored social and cultural environment facilitators of physical activity among 12- to 14-year-old AfricanAmerican adolescents living in a metropolitan area in the Southeast. Methods: Youth (n ¼ 51; 45% male) participated in brainstorming focus groups responding to the prompt, “What about your family, friends, and community, encourages you to be physically active?” In a second meeting, participants (n ¼ 56; 37.5% male) sorted statements (n ¼ 84) based on similarity in meaning and rated statements on relative importance. Statement groups and ratings were entered into Concept Systems software where multidimensional scaling and hierarchical cluster analysis were used to create graphical representation of ideas. Finally, researchers named clusters according to the gestalt of grouped statements. Results: The total sample included 28.9% of youth with household incomes $30,000 (area median income ¼ $30,701), 29% who perceived themselves as overweight, and 14.5% who reported being active for 60þ minutes everyday. Nine clusters, in rank order, emerged as follows: access/ availability of physical activity resources; family and friend support; physical activity with friends; physical activity with family members; inspiration to/from others; parental reinforcement; opportunities in daily routine; pressure from social networks; and seeing consequences of activity/ inactivity. Themes analyzed by gender were very similar (r ¼ .90); however, “pressure from social networks” was more important for girls than boys (r ¼ .10). Conclusions: Clear patterns of social and cultural facilitators of physical activity are perceived by African-American adolescents. Interventions targeting this group may benefit by incorporating these themes. Ó 2015 Society for Adolescent Health and Medicine. All rights reserved.
Conflicts of Interest: This study was funded by Active Living Research/Robert Wood Johnson Foundation grant 65659 to M.L.B. Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the Robert Wood Johnson Foundation. * Address correspondence to: Monica L. Baskin, Ph.D., Division of Preventive Medicine, UAB School of Medicine, 1717 11th Avenue South, MT 618, Birmingham, AL 35205. E-mail address: [email protected] (M.L. Baskin). 1054-139X/Ó 2015 Society for Adolescent Health and Medicine. All rights reserved. http://dx.doi.org/10.1016/j.jadohealth.2015.01.012
IMPLICATIONS AND CONTRIBUTION
African-American adolescents are at greater risk of being physically inactive and, concomitantly, of being obese. It is critical to identify facilitators of physical activity for this population. Study findings identify access/availability and family and peer social supports that can be used to develop ecologically valid physical activity frameworks for AfricanAmerican adolescents.
Adolescent obesity quadrupled over the past three decades [1]. Youth with high body mass index are more likely to become obese as adults [2] and are at risk for chronic conditions including hypertension, stroke, heart disease, diabetes, and certain cancers [3]. Although obesity rates increased across populations, low-income and minority youth are at greatest risk [4]. An inverse relationship between socioeconomic status (SES)
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542
and obesity has been noted by some studies [5,6]; but others suggest this relationship is more complex, particularly among racial/ethnic minorities [4,7]. For white adolescents, higher SES appears to have a protective effect; however, black youth obesity rates do not differ by SES [7]. With respect to geographic differences, southern U.S. states have the highest rates regardless of race/ethnicity [8]. Regular physical activity can help reduce risks for obesity and related comorbid conditions [9]. Despite potential benefits [10], most youth do not meet recommended guidelines. Nationwide, only 28.7% meet recommendations for 60 minutes of daily physical activity, with African-American adolescents having lower levels of physical activity and higher levels of inactivity relative to whites [11]. Moreover, physical activity declines with age [12], with peak decline from 13 to 18 years [13]. This decline is particularly steep among African-American girls compared with that of white counterparts [14]. Adolescents from southern U.S. states report the lowest physical activity and the highest sedentary behavior levels, likely contributing to increased rates of overweight and obesity [15]. Taken together, more research on facilitators of physical activity among racial ethnic minority adolescents and those residing in the southern United States is warranted. An ecological model can be used to explain determinants of physical activity. The model asserts that an interactive relationship between characteristics of a person (e.g., gender, selfefficacy), intrapersonal factors (e.g., family and peers), social (e.g., cultural norms, social support/reinforcement), and physical environment (e.g., built environment) factors can influence behavior, including physical activity [16e18]. With regard to social factors, parents’ social support and reinforcement have been associated with minority adolescents’ physical activity in some studies [18,19] but not all [20]. A recent systematic review suggests parental role modeling and social support do not significantly influence physical activity outcomes for African-American youth [21]. Other social factors such as cultural beliefs may partially explain physical inactivity among some African-Americans. Culture, which includes unique shared values, beliefs, and practices of a group, may impact the acceptance and adoption of health promoting messages and influence health-related behaviors [22]. For example, valuing automobiles as a sign of economic attainment and/or purchasing multiple televisions (including for children’s bedrooms) may support sedentary behaviors [23]. Furthermore, physical activity may be seen as “work” and thus competing with desires for rest and relaxation [24]. Such values may be linked to historical vestiges of slavery in which African-Americans worked tirelessly six or more days per week in agriculture and other labor-intensive positions, particularly in the southern United States [25]. Some interventions for African-American youth have integrated cultural beliefs and values (e.g., religiosity, collectivism) and culturally tailored intervention materials and physical activities [21]; however, intervention effects have been inconclusive or nonsignificant [21]. Contradictory results have also been noted regarding the influence of the physical environment on African-American adolescent physical activity. Access to safe parks is inversely related to physical inactivity among whites but not AfricanAmerican youth [26]. Furthermore, our prior work suggests that perceptions of neighborhood social disorder do not significantly influence physical activity among African-American adolescents, although neighborhood perceptions do increase risks
537
for obesity [27]. These findings point to other factors (e.g., cultural influences) as potential drivers of physical inactivity in this population. This further suggests that there may be other aspects of the physical environment (e.g., perceptions of safety) worth consideration. Given the gaps in our understanding of the influences of intrapersonal, social, and physical environment factors on physical activity among African-American adolescents, insights from youth are needed. The present study used a communityengaged approach to identify the social and cultural factors influencing physical activity among African-American adolescents residing in an urban metropolitan area in the southeastern United States. Young adolescents (ages, 12e14 years) were targeted to better understand and potentially intervene before significant declines in physical activity. Methods Study participants Youth who self-reported as African-American, Englishspeaking, aged 12e14 years, and in generally good health (i.e., without physical condition limiting mobility) were included. Participants were recruited using snowball sampling and flyers placed at local recreation centers, churches, schools, and newspaper advertisements. The study protocol was approved by the University of Alabama at Birmingham (UAB) Institutional Review Board for human subjects. Adolescent participants and their parents provided assent and consent respectively. A total sample of 69 African-American girls (n ¼ 42) and boys (n ¼ 27) from the metropolitan area participated. Data collection Concept mapping, a structured approach that produces maps of ideas or concepts developed by individuals or groups using mixed methodology for data collection and analysis [28], was used in this study. The qualitative approach includes generation of ideas through focus groups and unstructured sorting of ideas. The quantitative portion involves multidimensional scaling and hierarchical cluster analysis of the statements based on the rating metric [28]. With this methodology, there is a collaborative process whereby participants are actively involved in generating ideas, structuring the statements, and identifying factors that are most relevant to address the research question [29]. Concept mapping involves five phases: preparation, generation, structuring, map analysis, and interpretation [29]. For the preparation phase, our research team brainstormed potential focus group prompts. Potential prompts were pilot tested with a convenience sample of African-American adolescents (n ¼ 7; 57% female) to select and refine the prompt for age-appropriate language and clarity. During the generation phase, male and female adolescents (gender-specific groupings) participated in a 1-hour focus group. There were four separate sessions held with a total of 51 adolescents (55% female) with approximately 8e13 adolescents per session. Youth responded to the focus prompt, “What about your family, friends, and community, encourages you to be physically active?” A definition of physical activity [15] was provided to ensure a common understanding when generating ideas. Youth were instructed, “by physically active, we mean do you play or
538
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542
exercise enough to make you sweat or breathe hard.” Using a modified nominal group technique [30], participants responded to the prompt in a round-robin fashion to control for group dynamics. Ideas were elicited (until saturation) from each participant and recorded on a flip chart. Next, participants edited statements for redundancy and relevance and were allowed to contribute additional relevant statements if desired. Youth did not prioritize ideas. Adolescents and their parents completed a demographic survey for the purpose of describing the sample. Parents reported total annual household income. Adolescents reported their age, gender, and perceived weight status. Adolescents also reported how many days they were physically active at least 60 minutes per day during the past 7 days. Physical activity was defined as, “any activity that increases your heart rate and makes you breathe hard some of the time.” Each focus group session generated, on average, 40 statements for a total of 165 statements across all groups. Using an iterative process, research staff refined the statements to delete duplications, revise gender-specific statements to be gender neutral, and edit statements for clarity. The final set included 84 unique statements. For the structuring phase, 56 adolescents (including of 38 teens from the generation phase) participated. During this meeting, participants sorted and rated the 84 brainstorming statements. Each participant received a stack of cards (in random order) with one statement printed per card. Participants were instructed to read each statement and then sort the statements into piles based on similarity in meaning that made sense to him/ her. Follow-up instructions indicated that a statement could be sorted into a pile by itself if it did not fit with the other statements and that statements could not be sorted into more than one pile. After sorting the statements, participants provided a descriptive label for each pile. Staff reviewed each participant’s sort piles to ensure all statements were sorted, piles were labeled, and no statement had been sorted into multiple piles. Participants then rated each statement based on its importance relative to the other statements in the set. The response ratings used a Likert-type scale ranging from 5 (most important) to 1 (least important). Participants completed demographic and health behavior questionnaires as in the generation phase. Data analysis and interpretation Concept Systems software (Concept Systems Incorporated, Ithaca, NY) was used to analyze the clustering and rating of statements for the map analysis phase. The software uses multidimensional scaling and hierarchical cluster analysis techniques to create graphical mapped representations of the data. The multidimensional scaling method creates a point map where each point represents an individual statement and distance between points reflect the distance between statements based on the average of the group sorting. The software provides a stress value to indicate the goodness of fit of the mapped points to the individual sort data. Stress values can range from 0 (perfect fit) to 1 (poor fit). Acceptable stress values range from .205 to .365 and are indicative of internal representational validity [29]. A recent meta-analysis that pooled concept mapping studies found that the average stress value was .28 (standard deviation, .04) [31]. After the point map is created, boundaries are drawn around clusters of points to represent how statements are clustered together. The clusters are created using Ward’s algorithm in which each statement is its own cluster and successive merges
are completed to minimize the sum of squares of the distance between statements until all statements are in one single cluster [29]. Clusters closer together represent more similarity in meaning, whereas clusters farther apart are less similar. Additionally, clusters of broader concepts (more diverse set of statements) are represented as larger than other clusters, whereas smaller clusters indicate more narrowly focused concepts (less diverse set of statements). Rating data are used to evaluate the average relative importance of each cluster, to compare the average relative importance of statements within and across the clusters, and to develop pattern matches that allowed for comparison of relative importance across important demographic variables. The importance of the cluster is determined by the number of layers, where more layers indicate higher average relative importance compared with other clusters. The pattern match provides a correlation coefficient that represents the strength of the correlation of cluster ratings across variables of interest and ranges from 0 (no correlation) to 1 (strong correlation). The research team completed the interpretation phase. The number of clusters derived by the software is not driven by theory and there is not a fixed limit to clusters derived. To develop the appropriate number of clusters, the ecological model was considered while also keeping in mind parsimony and interpretability of the clusters. Solutions ranging from 8 to 20 clusters were analyzed, but the research team agreed on the nine-cluster solution as the most meaningful and in line with the conceptual framework. Final cluster names were based on a review of statements within each cluster and the list of pile names provided by participants. The final map resulted in an overall stress value of .256 after 15 iterations, suggesting a good fit of the mapped points to the individual sort data and internal representational validity [31]. Results Participant demographic characteristics for the brainstorming and structuring sessions and total sample are described in Table 1. The mean participant age was 13 years. Less than half of participants were male (39.1%) and 28.9% were from households with incomes $30,000. None of the participants perceived themselves to be “very overweight” but 29% endorsed being “somewhat overweight.” Only 14.5% reported meeting recommended guidelines of being active for 60þ minutes everyday. Figure 1 presents the final concept map for all participants. Cluster names, individual statements within clusters, and statement ratings are presented in Table 2. The cluster with the highest average importance rating (3.93) was “access/availability of physical activity resources.” This cluster included 13 statements primarily related to resources at home, school, and the larger community to support active living. The cluster with the lowest average rating (3.34) was “seeing consequences of activity/inactivity.” This cluster included 10 statements primarily focusing on exposure to information linking physical activity and/or weight to health/health outcomes. Individual statement ratings for importance ranged from a high value of 4.47 (“family wants you to stay in shape to live longer and be healthy” and “having year-round sports activities [e.g., swim team]”) to a low value of 2.68 (“peer pressure”). To examine whether there were gender differences in facilitators of physical activity, concept maps, and importance ratings, we computed male and female data separately (data not
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542 Table 1 Demographic characteristics of adolescents participating in generating and structuring sessions (N ¼ 69) Total sample (n ¼ 69); count (%) Age (years) Mean (standard deviation) 13.0 (.88) Gender Male 27 (39.1) Female 42 (60.9) Annual household incomea ($) 0e15,000 11 (15.9) 15,001e30,000 9 (13.0) 30,001e45,000 7 (10.1) 45,001e60,000 5 (7.2) 60,001e75,000 15 (21.7) >75,000 20 (29.0) Perceived weight status Very underweight 3 (4.3) Somewhat underweight 8 (11.6) Normal weight 37 (53.6) Somewhat overweight 20 (29.0) 10 (14.5) Physically active for at least 60 minutes on all 7 days in a week a
Generating (focus groups; n ¼ 51); count (%)
Structuring (n ¼ 56); count (%)
13.0 (.89)
13.04 (.88)
23 (45.1) 28 (54.9)
21 (37.5) 35 (62.5)
7 7 4 4 11 17
(13.7) (13.7) (7.8) (7.8) (21.6) (33.3)
8 6 7 4 11 18
(14.3) (10.7) (12.5) (7.1) (19.6) (32.1)
3 6 27 14 7
(5.9) (11.8) (52.9) (27.5) (13.7)
3 7 31 15 7
(5.4) (12.5) (55.4) (26.8) (12.5)
Percentages do not add to 100 because of missing values.
presented). The final concept maps for boys and girls also included nine clusters. The same clusters from the full sample were generated in the gender-specific map for girls. However, the map for boys did not include the cluster “seeing consequences of activity/inactivity” but rather generated a different cluster “competition.” Cluster pattern matching was used to identify the relative agreement between boys and girls in relation to the overall nine cluster themes (see Table 3). The overall agreement was strong (r ¼ .9); however, there was weak agreement between boys and girls on the importance of “pressure from
539
social network” on influencing physical activity (r ¼ .10). On average, girls rated this cluster higher on importance than boys. Gender differences by rated statements were also noted. For girls, the most important rated statement was “family wants you to stay in shape to live longer and be healthy” (4.51), whereas boys rated “being a good role model” as the most important statement (4.75). The least important statement for girls was “walk to the store” (2.57) and for boys, “watching workout tapes with friends” was the least important statement (2.47). Discussion This study used a community-engaged, mixed-methods approach and incorporated an ecological framework to identify interpersonal (e.g., family, peer), social (e.g., social and cultural norms, social support/reinforcement), and physical environmental (e.g., neighborhoods) factors that promote physical activity among African-American adolescents living in the southern United States. Our findings provide new insights into facilitators of physical activity among a population of youth who are at higher risk of physical inactivity and obesity [11,14]. Adolescents identified several domains that encourage physical activity such as access and availability of physical activity resources, opportunities to be physically active, engaging in physical activity with family and friends as well as family and friend support. Incorporating adolescents’ perspectives into obesity prevention/treatment interventions may improve efficacy, because to date, results of interventions with African-American youth provide inconclusive or weak evidence for increasing physical activity and reducing obesity [21,32]. In the present study, adolescents’ perceived resources at school and within the community as the most important determinants of physical activity. These findings align with prior work suggesting that these factors are important determinants of physical activity [33]. In addition, findings related to community support and physical activity-related norms (e.g., seeing other
Figure 1. Cluster rating map, all participants. Clusters that are closer together represent more similarity in meaning. Broad clusters indicate broader concepts; smaller clusters indicate narrowly focused concepts. Importance of the cluster is determined by the number of layers, where more layers indicate higher average relative importance compared to other clusters in the map.
540
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542
Table 2 Items influencing physical activity by clusters and their average importance ratings Items/statements by clusters Cluster 1: Access/availability of physical activity resources Schools require physical activity Having football, basketball, and baseball leagues available Having a safe place to ride your bike Community has places to be active (i.e., parks, recreation center, and basketball courts) Having teams at school (i.e., Newcomb, basketball, and football) Communities encourage sports School has gym equipment (e.g., treadmills, elliptical machine, and weights) Community service/volunteering (e.g., cleaning up neighborhood, helping senior citizens) Having a community recreation center in neighborhood A lot of people in neighborhood jog, walk, walk dog, and run Community recreation center has swimming pool Having music playing at home or in neighborhood Having to walk your dog Average cluster rating Cluster 2: Family and friends support Family wants you to stay in shape to live longer and be healthy Family and friends encourage me to be active Family encourage you (“you can do anything”) to do sports Friends encourage each other with positive statements (e.g., Way to go! You can do it!) Family encourages physical activity for health reasons Family and friends sign up for sports asks if she wants to join Brothers/sisters encourage each other by suggesting a physical activity (i.e., walking together) Parents tell you to get involved in physical activities at school Friends encourage each other by challenging to do more (e.g., “I don’t think you can.” “You cannot keep up.”) Doing errands for family members (e.g., parents, older relatives) Seeing friends and cousins playing sports Family member puts you on sports teams he/she coaches Adult family members play college or NFL sports If you are in the house too long dad says to go outside and play If you are in the house too long mom says to go outside and play Average cluster rating Cluster 3: Physical activity with friends Teamwork Play activities that help you to be physically active and learn something new too Participating in a sports competition Workout at school with best friend Being physically active with friends at an activity center (e.g., skating rink, and laser tag) Wants to practice skills to get better at sports (e.g., dribbling) Playing at park with friends (i.e., skating, jumping rope, and running) Being physically active with friends at school (e.g., football, basketball) Play competitive games (e.g., 21 on the basketball court) Girls and boys team up and play against each other Friends starting an exercise group Being physically active with friends in neighborhood (e.g., play soccer, Frisbee, and swim) Average cluster rating Cluster 4: Physical activity with family members Play sports with older brother/sister
Average ratingsa 4.35b 4.25b 4.18 4.16 4.15 4.11 4.07 4.00 3.89 3.63 3.61 3.41 3.31 3.93 4.47b 4.36b 4.32b 4.30b 4.26b 4.00 4.00 3.98 3.77 3.72 3.60 3.49 3.39 3.35 3.22 3.88 b
4.44 4.15 4.13 3.98 3.95 3.89 3.76 3.72 3.68 3.62 3.57 3.50 3.87 4.07
(continued)
Table 2 Continued Items/statements by clusters
Average ratingsa
Physically active with your father (e.g., basketball, running, and walking) Friends and family play sports together Family/community sees physical activity as a way to stay out of trouble Physically active with your mother (e.g., basketball, running, and walking) Brothers/sisters compete with each other Friends have competitions with each other to see who is best Play sports with little brother/sister Play sports at home with family Play active electronic games with friends or family (e.g., Wii, Dance Revolution) Horse playing with family members (e.g., wrestling) Average cluster rating Cluster 5: Inspiration to/from others Being a good role model Doing better than the person you admire Inspired by someone you know to better yourself Be an idol (gain popularity) Peer pressure Average cluster rating Cluster 6: Parental reinforcement Mom says not to quit whatever you are doing Dad says not to quit whatever you are doing Mom always tells you to practice Dad always tells you to practice Parents forcing you to play sports Average cluster rating Cluster 7: Opportunities in daily routine Having year-round sports activities (e.g., swim team) School programs that get you active (e.g., gardening classes, ballet, and dance) Park in walking distance from home and a lot of kids your age there Having play time after finishing work Walk to or in the neighborhood park Walk to the store Average cluster rating Cluster 8: Pressure from social networks Most of your friends are healthy and fit, you want to keep up Friends pushing you to the limit Family history of obesity, diabetes, and heart failure Taking care of younger kids (i.e., babysitting) Getting money for chores including physical activity (e.g., cutting grass, washing car, and raking leaves) Teachers commenting about being active Girlfriend/boyfriend wants you to stay slim Average cluster rating Cluster 9: Seeing consequences of activity/inactivity Needing to keep a certain weight for sports team, job, etc. See obese kids your age (do not want to look like that) Having personal trainer/mentor Seeing people who are in shape Seeing people riding bikes on the street Watching fit TV-cardio and step and muscle endurance Television/magazine commercials about weight loss Seeing people who are not in shape Watching people on television who are fit Watching workout tapes with friends Average cluster rating
4.04
a b
4.04 3.96 3.96 3.82 3.81 3.75 3.45 3.36 2.98 3.75 4.41b 4.05 3.93 3.70 2.68 3.75 4.00 3.95 3.86 3.78 3.02 3.72 4.47b 3.89 3.88 3.16 3.07 2.69 3.53 3.98 3.66 3.65 3.57 3.48 3.23 2.93 3.50 3.98 3.84 3.51 3.34 3.33 3.22 3.18 3.09 2.96 2.91 3.34
On a 1 (least important) to 5 (most important) rating scale. Top 10 items by average importance rating for all participants.
neighbors jogging or walking in their neighborhood) further support suggestions that neighborhood social and environmental features supportive of physical activity may increase physical activity-related outcomes among adolescents [33].
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542 Table 3 Cluster match by gender Cluster
Cluster names
r value
1 2 3 4 5 6 7 8 9
Access/availability of physical activity resources Family and friends support Physical activity with friends Physical activity with family members Inspiration to/from others Parental reinforcement Opportunities in daily routine Pressure from social network Seeing consequences of activity/inactivity
.71 .78 .80 .47 .92 .86 .85 .10 .59
r Value indicates agreement between ratings of boys and girls.
Although physical activity-related encouragement and motivational support from friends are generally reported as significant predictors of physical activity, this relationship may be complex. Evidence from a systematic review on social networks and physical activity suggests adolescents are more likely to participate in physical activity when a high proportion of their friends are physically active [34]. Peer modeling is another facilitator of physical activity among adolescents [34,35]. This suggests that mere advice or encouragement to be active may not be as effective as actual engagement in physical activity by members of social network. Adolescents reported seeing the consequences of physical activity/inactivity as a facilitator of physical activity but it was rated least important relative to other factors. This cluster included media influence such as watching people on television who are fit and commercials/ads about weight loss. This finding is consistent with prior research with African-American female adolescents suggesting celebrities are not influential role models for physical activity, and body size preferences are less determined by people they do not know [36]. As such, intrasocial network factors may be better motivators of physical activity, especially for African-American girls. Although there is strong agreement between boys and girls on the major clusters of influence and the relative importance of clusters, some differences are noted. One area of divergence relates to competition (e.g., contests among friends) as a motivator of physical activity for boys but not girls. This finding is supported by previous research [37]. Additionally, boys suggest that reinforcement and encouragement for physical activity from both peer and family influences are motivators. In contrast, girls report that parental reinforcement and encouragement are more salient than that received from friends. This finding also contradicts prior work [37] suggesting both boys and girls seek support from parents and friends. However, findings from the present study and others [33,38,39] point to parental support increasing physical activity more so for African-American girls. Top rated statements for boys to be physically active rely mostly on internal motivations (e.g., being a role model), whereas among girls, the most important statements relate to external motivations for physical activity (e.g., family wants you to stay in shape to live longer). These findings may suggest that to increase physical activity, girls require more encouragement from their social networks, whereas boys may need to be equipped with the skills to increase their internal motivators. In addition, girls perceive safety to be an important issue, which is not rated highly for boys. Similar findings are reported by Ries et al. [40] where girls but not boys report safety as a determinant for physical activity. However, for the present study results, it is
541
unclear what specific elements of safety are important. Although some qualitative research indicates that elements of safety are important for physical activity [37], our empirical research [27] with a larger cohort of African-American adolescents from the same community as the present study, did not find support for perceived neighborhood disorder as a barrier to physical activity levels. The current findings suggest that future interventions for African-American adolescents should address gender differences in facilitators of physical activity. For instance, potential intervention strategies for boys might focus on the benefits of role modeling physical activity for others, identifying competitiverelated activities, identifying strategies to involve older siblings in active play with their younger siblings, and providing information about community-specific resources for physical activity. In contrast, girls may benefit from intervention strategies focused on enhancing supportive social networks, identifying physical activities that involve teamwork, providing messaging for parents to encourage physical activity and relate it to health, and mapping areas in neighborhoods and communities that girls (and their parents) feel safe to engage in physical activity. Future physical activity and obesity prevention/treatment interventions that address these gender-specific motivations may improve intervention outcomes. Limitations Although informative, this study is not without limitations. The sample for this study was small and may not be representative of youth in this region.Most of the study participants were from households with incomes greater than the median income. As such, influencing factors related to youth from lower SES families may not be well represented in our findings. Similarly, more than half of the participants perceived themselves as normal weight, and we did not objectively measure weight and height to classify obesity status. On the other hand, given the noted lack of consistent association between SES and obesity among minority adolescents [4,7], findings from this study may still capture salient influential factors on physical activity for many African-American youth. Strengths of the study This study is unique in its use of a participatory methodology to explore social and cultural environment influencing factors on physical activity among African-American youth. Given the higher risk of physical inactivity, obesity, and related comorbidities among this population, having the perspectives of this group are key to future public health interventions. As used in this study, concept mapping increases the robustness of qualitative findings by combining with quantitative analysis to produce outputs (e.g., concept maps) that show relationships between individual or group ideas and present them in a way that is easy to convey to stakeholders [29]. In the present study, this approach helped to generate potential physical activity intervention strategies that are ecologically valid and work within the context of adolescents’ everyday lives and environments. Additionally, because the ideas originated from the target audience, future interventions with this population may be more acceptable. This study incorporates a mixed-methods approach to identify and examine facilitators of physical activity among
542
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542
African-American adolescents. Our results suggest that factors related to access and availability, the importance of family and friend support for physical activity and physical activity with friends are key determinants of physical activity among African-American adolescents. Furthermore, these results are strengthened by the high level of agreement between boys and girls, suggesting that the development of interventions that address these similarly rated domains will be of benefit to both gender groups. Future interventions to increase physical activity among African-American adolescents should include strategies to strengthen the social and cultural environment, particularly related to the factors identified by this study. References [1] Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA 2014;311:806e14. [2] Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in overweight among US children and adolescents, 1999-2000. J Am Med Assoc 2002;288:1728e32. [3] Malnick SD, Knobler H. The medical complications of obesity. QJM 2006;99: 565e79. [4] Rossen LM. Neighbourhood economic deprivation explains racial/ethnic disparities in overweight and obesity among children and adolescents in the USA. J Epidemiol Community Health 2014;68:123e9. [5] Shrewsbury V, Wardle J. Socioeconomic status and adiposity in childhood: A systematic review of cross-sectional studies 1990e2005. Obesity (Silver Spring) 2008;16:275e84. [6] Janssen I, Boyce WF, Simpson K, Pickett W. Influence of individual- and arealevel measures of socioeconomic status on obesity, unhealthy eating, and physical inactivity in Canadian adolescents. Am J Clin Nutr 2006;83:139e45. [7] Gordon-Larsen P, Adair LS, Popkin BM. The relationship of ethnicity, socioeconomic factors, and overweight in US adolescents. Obes Res 2003; 11:121e9. [8] Bethell C, Read D, Goodman E, et al. Consistently inconsistent: A snapshot of across- and within-state disparities in the prevalence of childhood overweight and obesity. Pediatrics 2009;123(Suppl 5):S277e86. [9] Ross R, Bradshaw AJ. The future of obesity reduction: Beyond weight loss. Nat Rev Endocrinol 2009;5:319e25. [10] Rowland TW. Promoting physical activity for Children’s Health: Rationale and strategies. Sports Med 2007;37:929e36. [11] Eaton D, Kann L, Kinchen S, et al. Youth risk behavior surveillanceeUnited States, 2011. MMWR Surveill Summ 2012;61:1e162. [12] Troiano RP, Berrigan D, Dodd KW, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc 2008;40:181e8. [13] Sallis J. Age-related decline in physical activity: A synthesis of human and animal studies. Med Sci Sports Exerc 2000;32:1598e600. [14] Kimm SYS, Glynn NW, Kriska AM, et al. Decline in physical activity in black girls and white girls during adolescence. N Engl J Med 2002;347:709e15. [15] Springer AE, Hoelscher DM, Kelder SH. Prevalence of physical activity and sedentary behaviors in US high school students by metropolitan status and geographic region. J Phys Activity Health 2006;3:365. [16] Sallis JF, Cervero RB, Ascher W, et al. An ecological approach to creating active living communities. Annu Rev Public Health 2006;27:297e322. [17] Fitzpatrick KM, LaGory M. “Placing” health in an urban sociology: Cities as mosaics of risk and protections. City & Community 2003;2:33e46. [18] Lawman H, Wilson D. Associations of social and environmental supports with sedentary behavior, light and moderate-to-vigorous physical
[19]
[20]
[21]
[22]
[23] [24]
[25] [26] [27]
[28]
[29] [30] [31] [32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
activity in obese underserved adolescents. Int J Behav Nutr Phys Act 2014;11:92. Edwardson CL, Gorely T. Parental influences on different types and intensities of physical activity in youth: A systematic review. Psychol Sport Exerc 2010;11:522e35. Van Der Horst K, Paw M, Twisk J, Van Mechelen W. A brief review on correlates of physical activity and sedentariness in youth. Med Sci Sports Exerc 2007;39:1241e50. Barr-Anderson DJ, Adams-Wynn AW, DiSantis KI, Kumanyika S. Family-focused physical activity, diet and obesity interventions in African-American girls: A systematic review. Obes Rev 2013;14:29e51. Pasick R, D’Onofrio C, Otero-Sabogal R. Similarities and differences across cultures: Questions to inform a third generation for health promotion research. Health Educ Q 1996;23:S142e61. Kumanyika SK. Environmental influences on childhood obesity: Ethnic and cultural influences in context. Physiol Behav 2008;94:61. Airhihenbuwa CO, Kumanyika S, Agurs TD, Lowe A. Perceptions and beliefs about exercise, rest, and health among African-Americans. Am J Health Promot 1995;9:426e9. Ruef M, Fletcher B. Legacies of American slavery: Status attainment among southern blacks after emancipation. Soc Forces 2003;82:445e80. Babey SH, Hastert TA, Yu H, Brown ER. Physical activity among adolescents: When do parks matter? Am J Prev Med 2008;34:345. Dulin-Keita A, Kaur Thind H, Affuso O, Baskin M. The associations of perceived neighborhood disorder and physical activity with obesity among African American adolescents. BMC Public Health 2013;13:440. Trochim W, Kane M. Concept mapping: An introduction to structured conceptualization in health care. Int J Qual Health Care 2005;17: 187e91. Kane M, Trochim WMK. Concept mapping for planning and evaluation, Vol. 50. Thousand Oaks, CA: Sage Publications; 2007. Gallagher M, Hares T, Spencer J, et al. The nominal group technique: A research tool for general practice? Fam Pract 1993;10:76e81. Rosas SR, Kane M. Quality and rigor of the concept mapping methodology: A pooled study analysis. Eval Program Plann 2012;35:236e45. Barr-Anderson DJ, Singleton C, Cotwright CJ, et al. Outside-of-school time obesity prevention and treatment interventions in African American youth. Obes Rev 2014;15:26e45. Wilson D, Lawman H, Segal M, Chappell S. Neighborhood and parental supports for physical activity in minority adolescents. Am J Prev Med 2011; 41:399e406. Macdonald-Wallis K, Jago R, Sterne JAC. Social network analysis of childhood and youth physical activity: A systematic review. Am J Prev Med 2012;43:636e42. Salvy SJ, de la Haye K, Bowker JC, Hermans RCJ. Influence of peers and friends on children’s and adolescents’ eating and activity behaviors. Physiol Behav 2012;106:369e78. Boyington JEA, Carter-Edwards L, Piehl M, et al. Cultural attitudes toward weight, diet, and physical activity among overweight African American girls. Prev Chronic Dis 2008;5:A36. St George SM, Wilson DK. A qualitative study for understanding family and peer influences on obesity-related health behaviors in low-income African-American adolescents. Child Obes 2012;8:466e76. Adkins S, Sherwood NE, Story M, Davis M. Physical activity among African-American Girls: The role of parents and the home environment. Obes Res 2004;12 suppl:38Se45S. Wright MS, Wilson DK, Griffin S, Evans A. A qualitative study of parental modeling and social support for physical activity in underserved adolescents. Health Educ Res 2010;25:224e32. Ries AV, Voorhees CC, Gittelsohn J, et al. Adolescents’ perceptions of environmental influences on physical activity. Am J Health Behav 2008;32: 26e39.
www.jahonline.org Original article
Social and Cultural Environment Factors Influencing Physical Activity Among African-American Adolescents Monica L. Baskin, Ph.D. a, *, Akilah Dulin-Keita, Ph.D. b, Herpreet Thind, Ph.D., M.P.H. c, and Emily Godsey d a
Department of Medicine, Division of Preventive Medicine, UAB School of Medicine, Birmingham, Alabama Department of Behavioral and Social Sciences, Institute for Community Health Promotion, Brown University, Providence, Rhode Island c Department of Psychiatry and Human Behavior, Centers for Behavioral and Preventive Medicine, Brown Alpert Medical School and the Miriam Hospital, Providence, Rhode Island d Department of Sociology, University of Alabama at Birmingham, Birmingham, Alabama b
Article history: Received July 24, 2014; Accepted January 20, 2015 Keywords: Mixed methods; African-American; Youth; Physical activity
A B S T R A C T
Purpose: African-American youth are at high risk for physical inactivity. This study explored social and cultural environment facilitators of physical activity among 12- to 14-year-old AfricanAmerican adolescents living in a metropolitan area in the Southeast. Methods: Youth (n ¼ 51; 45% male) participated in brainstorming focus groups responding to the prompt, “What about your family, friends, and community, encourages you to be physically active?” In a second meeting, participants (n ¼ 56; 37.5% male) sorted statements (n ¼ 84) based on similarity in meaning and rated statements on relative importance. Statement groups and ratings were entered into Concept Systems software where multidimensional scaling and hierarchical cluster analysis were used to create graphical representation of ideas. Finally, researchers named clusters according to the gestalt of grouped statements. Results: The total sample included 28.9% of youth with household incomes $30,000 (area median income ¼ $30,701), 29% who perceived themselves as overweight, and 14.5% who reported being active for 60þ minutes everyday. Nine clusters, in rank order, emerged as follows: access/ availability of physical activity resources; family and friend support; physical activity with friends; physical activity with family members; inspiration to/from others; parental reinforcement; opportunities in daily routine; pressure from social networks; and seeing consequences of activity/ inactivity. Themes analyzed by gender were very similar (r ¼ .90); however, “pressure from social networks” was more important for girls than boys (r ¼ .10). Conclusions: Clear patterns of social and cultural facilitators of physical activity are perceived by African-American adolescents. Interventions targeting this group may benefit by incorporating these themes. Ó 2015 Society for Adolescent Health and Medicine. All rights reserved.
Conflicts of Interest: This study was funded by Active Living Research/Robert Wood Johnson Foundation grant 65659 to M.L.B. Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the Robert Wood Johnson Foundation. * Address correspondence to: Monica L. Baskin, Ph.D., Division of Preventive Medicine, UAB School of Medicine, 1717 11th Avenue South, MT 618, Birmingham, AL 35205. E-mail address: [email protected] (M.L. Baskin). 1054-139X/Ó 2015 Society for Adolescent Health and Medicine. All rights reserved. http://dx.doi.org/10.1016/j.jadohealth.2015.01.012
IMPLICATIONS AND CONTRIBUTION
African-American adolescents are at greater risk of being physically inactive and, concomitantly, of being obese. It is critical to identify facilitators of physical activity for this population. Study findings identify access/availability and family and peer social supports that can be used to develop ecologically valid physical activity frameworks for AfricanAmerican adolescents.
Adolescent obesity quadrupled over the past three decades [1]. Youth with high body mass index are more likely to become obese as adults [2] and are at risk for chronic conditions including hypertension, stroke, heart disease, diabetes, and certain cancers [3]. Although obesity rates increased across populations, low-income and minority youth are at greatest risk [4]. An inverse relationship between socioeconomic status (SES)
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542
and obesity has been noted by some studies [5,6]; but others suggest this relationship is more complex, particularly among racial/ethnic minorities [4,7]. For white adolescents, higher SES appears to have a protective effect; however, black youth obesity rates do not differ by SES [7]. With respect to geographic differences, southern U.S. states have the highest rates regardless of race/ethnicity [8]. Regular physical activity can help reduce risks for obesity and related comorbid conditions [9]. Despite potential benefits [10], most youth do not meet recommended guidelines. Nationwide, only 28.7% meet recommendations for 60 minutes of daily physical activity, with African-American adolescents having lower levels of physical activity and higher levels of inactivity relative to whites [11]. Moreover, physical activity declines with age [12], with peak decline from 13 to 18 years [13]. This decline is particularly steep among African-American girls compared with that of white counterparts [14]. Adolescents from southern U.S. states report the lowest physical activity and the highest sedentary behavior levels, likely contributing to increased rates of overweight and obesity [15]. Taken together, more research on facilitators of physical activity among racial ethnic minority adolescents and those residing in the southern United States is warranted. An ecological model can be used to explain determinants of physical activity. The model asserts that an interactive relationship between characteristics of a person (e.g., gender, selfefficacy), intrapersonal factors (e.g., family and peers), social (e.g., cultural norms, social support/reinforcement), and physical environment (e.g., built environment) factors can influence behavior, including physical activity [16e18]. With regard to social factors, parents’ social support and reinforcement have been associated with minority adolescents’ physical activity in some studies [18,19] but not all [20]. A recent systematic review suggests parental role modeling and social support do not significantly influence physical activity outcomes for African-American youth [21]. Other social factors such as cultural beliefs may partially explain physical inactivity among some African-Americans. Culture, which includes unique shared values, beliefs, and practices of a group, may impact the acceptance and adoption of health promoting messages and influence health-related behaviors [22]. For example, valuing automobiles as a sign of economic attainment and/or purchasing multiple televisions (including for children’s bedrooms) may support sedentary behaviors [23]. Furthermore, physical activity may be seen as “work” and thus competing with desires for rest and relaxation [24]. Such values may be linked to historical vestiges of slavery in which African-Americans worked tirelessly six or more days per week in agriculture and other labor-intensive positions, particularly in the southern United States [25]. Some interventions for African-American youth have integrated cultural beliefs and values (e.g., religiosity, collectivism) and culturally tailored intervention materials and physical activities [21]; however, intervention effects have been inconclusive or nonsignificant [21]. Contradictory results have also been noted regarding the influence of the physical environment on African-American adolescent physical activity. Access to safe parks is inversely related to physical inactivity among whites but not AfricanAmerican youth [26]. Furthermore, our prior work suggests that perceptions of neighborhood social disorder do not significantly influence physical activity among African-American adolescents, although neighborhood perceptions do increase risks
537
for obesity [27]. These findings point to other factors (e.g., cultural influences) as potential drivers of physical inactivity in this population. This further suggests that there may be other aspects of the physical environment (e.g., perceptions of safety) worth consideration. Given the gaps in our understanding of the influences of intrapersonal, social, and physical environment factors on physical activity among African-American adolescents, insights from youth are needed. The present study used a communityengaged approach to identify the social and cultural factors influencing physical activity among African-American adolescents residing in an urban metropolitan area in the southeastern United States. Young adolescents (ages, 12e14 years) were targeted to better understand and potentially intervene before significant declines in physical activity. Methods Study participants Youth who self-reported as African-American, Englishspeaking, aged 12e14 years, and in generally good health (i.e., without physical condition limiting mobility) were included. Participants were recruited using snowball sampling and flyers placed at local recreation centers, churches, schools, and newspaper advertisements. The study protocol was approved by the University of Alabama at Birmingham (UAB) Institutional Review Board for human subjects. Adolescent participants and their parents provided assent and consent respectively. A total sample of 69 African-American girls (n ¼ 42) and boys (n ¼ 27) from the metropolitan area participated. Data collection Concept mapping, a structured approach that produces maps of ideas or concepts developed by individuals or groups using mixed methodology for data collection and analysis [28], was used in this study. The qualitative approach includes generation of ideas through focus groups and unstructured sorting of ideas. The quantitative portion involves multidimensional scaling and hierarchical cluster analysis of the statements based on the rating metric [28]. With this methodology, there is a collaborative process whereby participants are actively involved in generating ideas, structuring the statements, and identifying factors that are most relevant to address the research question [29]. Concept mapping involves five phases: preparation, generation, structuring, map analysis, and interpretation [29]. For the preparation phase, our research team brainstormed potential focus group prompts. Potential prompts were pilot tested with a convenience sample of African-American adolescents (n ¼ 7; 57% female) to select and refine the prompt for age-appropriate language and clarity. During the generation phase, male and female adolescents (gender-specific groupings) participated in a 1-hour focus group. There were four separate sessions held with a total of 51 adolescents (55% female) with approximately 8e13 adolescents per session. Youth responded to the focus prompt, “What about your family, friends, and community, encourages you to be physically active?” A definition of physical activity [15] was provided to ensure a common understanding when generating ideas. Youth were instructed, “by physically active, we mean do you play or
538
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542
exercise enough to make you sweat or breathe hard.” Using a modified nominal group technique [30], participants responded to the prompt in a round-robin fashion to control for group dynamics. Ideas were elicited (until saturation) from each participant and recorded on a flip chart. Next, participants edited statements for redundancy and relevance and were allowed to contribute additional relevant statements if desired. Youth did not prioritize ideas. Adolescents and their parents completed a demographic survey for the purpose of describing the sample. Parents reported total annual household income. Adolescents reported their age, gender, and perceived weight status. Adolescents also reported how many days they were physically active at least 60 minutes per day during the past 7 days. Physical activity was defined as, “any activity that increases your heart rate and makes you breathe hard some of the time.” Each focus group session generated, on average, 40 statements for a total of 165 statements across all groups. Using an iterative process, research staff refined the statements to delete duplications, revise gender-specific statements to be gender neutral, and edit statements for clarity. The final set included 84 unique statements. For the structuring phase, 56 adolescents (including of 38 teens from the generation phase) participated. During this meeting, participants sorted and rated the 84 brainstorming statements. Each participant received a stack of cards (in random order) with one statement printed per card. Participants were instructed to read each statement and then sort the statements into piles based on similarity in meaning that made sense to him/ her. Follow-up instructions indicated that a statement could be sorted into a pile by itself if it did not fit with the other statements and that statements could not be sorted into more than one pile. After sorting the statements, participants provided a descriptive label for each pile. Staff reviewed each participant’s sort piles to ensure all statements were sorted, piles were labeled, and no statement had been sorted into multiple piles. Participants then rated each statement based on its importance relative to the other statements in the set. The response ratings used a Likert-type scale ranging from 5 (most important) to 1 (least important). Participants completed demographic and health behavior questionnaires as in the generation phase. Data analysis and interpretation Concept Systems software (Concept Systems Incorporated, Ithaca, NY) was used to analyze the clustering and rating of statements for the map analysis phase. The software uses multidimensional scaling and hierarchical cluster analysis techniques to create graphical mapped representations of the data. The multidimensional scaling method creates a point map where each point represents an individual statement and distance between points reflect the distance between statements based on the average of the group sorting. The software provides a stress value to indicate the goodness of fit of the mapped points to the individual sort data. Stress values can range from 0 (perfect fit) to 1 (poor fit). Acceptable stress values range from .205 to .365 and are indicative of internal representational validity [29]. A recent meta-analysis that pooled concept mapping studies found that the average stress value was .28 (standard deviation, .04) [31]. After the point map is created, boundaries are drawn around clusters of points to represent how statements are clustered together. The clusters are created using Ward’s algorithm in which each statement is its own cluster and successive merges
are completed to minimize the sum of squares of the distance between statements until all statements are in one single cluster [29]. Clusters closer together represent more similarity in meaning, whereas clusters farther apart are less similar. Additionally, clusters of broader concepts (more diverse set of statements) are represented as larger than other clusters, whereas smaller clusters indicate more narrowly focused concepts (less diverse set of statements). Rating data are used to evaluate the average relative importance of each cluster, to compare the average relative importance of statements within and across the clusters, and to develop pattern matches that allowed for comparison of relative importance across important demographic variables. The importance of the cluster is determined by the number of layers, where more layers indicate higher average relative importance compared with other clusters. The pattern match provides a correlation coefficient that represents the strength of the correlation of cluster ratings across variables of interest and ranges from 0 (no correlation) to 1 (strong correlation). The research team completed the interpretation phase. The number of clusters derived by the software is not driven by theory and there is not a fixed limit to clusters derived. To develop the appropriate number of clusters, the ecological model was considered while also keeping in mind parsimony and interpretability of the clusters. Solutions ranging from 8 to 20 clusters were analyzed, but the research team agreed on the nine-cluster solution as the most meaningful and in line with the conceptual framework. Final cluster names were based on a review of statements within each cluster and the list of pile names provided by participants. The final map resulted in an overall stress value of .256 after 15 iterations, suggesting a good fit of the mapped points to the individual sort data and internal representational validity [31]. Results Participant demographic characteristics for the brainstorming and structuring sessions and total sample are described in Table 1. The mean participant age was 13 years. Less than half of participants were male (39.1%) and 28.9% were from households with incomes $30,000. None of the participants perceived themselves to be “very overweight” but 29% endorsed being “somewhat overweight.” Only 14.5% reported meeting recommended guidelines of being active for 60þ minutes everyday. Figure 1 presents the final concept map for all participants. Cluster names, individual statements within clusters, and statement ratings are presented in Table 2. The cluster with the highest average importance rating (3.93) was “access/availability of physical activity resources.” This cluster included 13 statements primarily related to resources at home, school, and the larger community to support active living. The cluster with the lowest average rating (3.34) was “seeing consequences of activity/inactivity.” This cluster included 10 statements primarily focusing on exposure to information linking physical activity and/or weight to health/health outcomes. Individual statement ratings for importance ranged from a high value of 4.47 (“family wants you to stay in shape to live longer and be healthy” and “having year-round sports activities [e.g., swim team]”) to a low value of 2.68 (“peer pressure”). To examine whether there were gender differences in facilitators of physical activity, concept maps, and importance ratings, we computed male and female data separately (data not
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542 Table 1 Demographic characteristics of adolescents participating in generating and structuring sessions (N ¼ 69) Total sample (n ¼ 69); count (%) Age (years) Mean (standard deviation) 13.0 (.88) Gender Male 27 (39.1) Female 42 (60.9) Annual household incomea ($) 0e15,000 11 (15.9) 15,001e30,000 9 (13.0) 30,001e45,000 7 (10.1) 45,001e60,000 5 (7.2) 60,001e75,000 15 (21.7) >75,000 20 (29.0) Perceived weight status Very underweight 3 (4.3) Somewhat underweight 8 (11.6) Normal weight 37 (53.6) Somewhat overweight 20 (29.0) 10 (14.5) Physically active for at least 60 minutes on all 7 days in a week a
Generating (focus groups; n ¼ 51); count (%)
Structuring (n ¼ 56); count (%)
13.0 (.89)
13.04 (.88)
23 (45.1) 28 (54.9)
21 (37.5) 35 (62.5)
7 7 4 4 11 17
(13.7) (13.7) (7.8) (7.8) (21.6) (33.3)
8 6 7 4 11 18
(14.3) (10.7) (12.5) (7.1) (19.6) (32.1)
3 6 27 14 7
(5.9) (11.8) (52.9) (27.5) (13.7)
3 7 31 15 7
(5.4) (12.5) (55.4) (26.8) (12.5)
Percentages do not add to 100 because of missing values.
presented). The final concept maps for boys and girls also included nine clusters. The same clusters from the full sample were generated in the gender-specific map for girls. However, the map for boys did not include the cluster “seeing consequences of activity/inactivity” but rather generated a different cluster “competition.” Cluster pattern matching was used to identify the relative agreement between boys and girls in relation to the overall nine cluster themes (see Table 3). The overall agreement was strong (r ¼ .9); however, there was weak agreement between boys and girls on the importance of “pressure from
539
social network” on influencing physical activity (r ¼ .10). On average, girls rated this cluster higher on importance than boys. Gender differences by rated statements were also noted. For girls, the most important rated statement was “family wants you to stay in shape to live longer and be healthy” (4.51), whereas boys rated “being a good role model” as the most important statement (4.75). The least important statement for girls was “walk to the store” (2.57) and for boys, “watching workout tapes with friends” was the least important statement (2.47). Discussion This study used a community-engaged, mixed-methods approach and incorporated an ecological framework to identify interpersonal (e.g., family, peer), social (e.g., social and cultural norms, social support/reinforcement), and physical environmental (e.g., neighborhoods) factors that promote physical activity among African-American adolescents living in the southern United States. Our findings provide new insights into facilitators of physical activity among a population of youth who are at higher risk of physical inactivity and obesity [11,14]. Adolescents identified several domains that encourage physical activity such as access and availability of physical activity resources, opportunities to be physically active, engaging in physical activity with family and friends as well as family and friend support. Incorporating adolescents’ perspectives into obesity prevention/treatment interventions may improve efficacy, because to date, results of interventions with African-American youth provide inconclusive or weak evidence for increasing physical activity and reducing obesity [21,32]. In the present study, adolescents’ perceived resources at school and within the community as the most important determinants of physical activity. These findings align with prior work suggesting that these factors are important determinants of physical activity [33]. In addition, findings related to community support and physical activity-related norms (e.g., seeing other
Figure 1. Cluster rating map, all participants. Clusters that are closer together represent more similarity in meaning. Broad clusters indicate broader concepts; smaller clusters indicate narrowly focused concepts. Importance of the cluster is determined by the number of layers, where more layers indicate higher average relative importance compared to other clusters in the map.
540
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542
Table 2 Items influencing physical activity by clusters and their average importance ratings Items/statements by clusters Cluster 1: Access/availability of physical activity resources Schools require physical activity Having football, basketball, and baseball leagues available Having a safe place to ride your bike Community has places to be active (i.e., parks, recreation center, and basketball courts) Having teams at school (i.e., Newcomb, basketball, and football) Communities encourage sports School has gym equipment (e.g., treadmills, elliptical machine, and weights) Community service/volunteering (e.g., cleaning up neighborhood, helping senior citizens) Having a community recreation center in neighborhood A lot of people in neighborhood jog, walk, walk dog, and run Community recreation center has swimming pool Having music playing at home or in neighborhood Having to walk your dog Average cluster rating Cluster 2: Family and friends support Family wants you to stay in shape to live longer and be healthy Family and friends encourage me to be active Family encourage you (“you can do anything”) to do sports Friends encourage each other with positive statements (e.g., Way to go! You can do it!) Family encourages physical activity for health reasons Family and friends sign up for sports asks if she wants to join Brothers/sisters encourage each other by suggesting a physical activity (i.e., walking together) Parents tell you to get involved in physical activities at school Friends encourage each other by challenging to do more (e.g., “I don’t think you can.” “You cannot keep up.”) Doing errands for family members (e.g., parents, older relatives) Seeing friends and cousins playing sports Family member puts you on sports teams he/she coaches Adult family members play college or NFL sports If you are in the house too long dad says to go outside and play If you are in the house too long mom says to go outside and play Average cluster rating Cluster 3: Physical activity with friends Teamwork Play activities that help you to be physically active and learn something new too Participating in a sports competition Workout at school with best friend Being physically active with friends at an activity center (e.g., skating rink, and laser tag) Wants to practice skills to get better at sports (e.g., dribbling) Playing at park with friends (i.e., skating, jumping rope, and running) Being physically active with friends at school (e.g., football, basketball) Play competitive games (e.g., 21 on the basketball court) Girls and boys team up and play against each other Friends starting an exercise group Being physically active with friends in neighborhood (e.g., play soccer, Frisbee, and swim) Average cluster rating Cluster 4: Physical activity with family members Play sports with older brother/sister
Average ratingsa 4.35b 4.25b 4.18 4.16 4.15 4.11 4.07 4.00 3.89 3.63 3.61 3.41 3.31 3.93 4.47b 4.36b 4.32b 4.30b 4.26b 4.00 4.00 3.98 3.77 3.72 3.60 3.49 3.39 3.35 3.22 3.88 b
4.44 4.15 4.13 3.98 3.95 3.89 3.76 3.72 3.68 3.62 3.57 3.50 3.87 4.07
(continued)
Table 2 Continued Items/statements by clusters
Average ratingsa
Physically active with your father (e.g., basketball, running, and walking) Friends and family play sports together Family/community sees physical activity as a way to stay out of trouble Physically active with your mother (e.g., basketball, running, and walking) Brothers/sisters compete with each other Friends have competitions with each other to see who is best Play sports with little brother/sister Play sports at home with family Play active electronic games with friends or family (e.g., Wii, Dance Revolution) Horse playing with family members (e.g., wrestling) Average cluster rating Cluster 5: Inspiration to/from others Being a good role model Doing better than the person you admire Inspired by someone you know to better yourself Be an idol (gain popularity) Peer pressure Average cluster rating Cluster 6: Parental reinforcement Mom says not to quit whatever you are doing Dad says not to quit whatever you are doing Mom always tells you to practice Dad always tells you to practice Parents forcing you to play sports Average cluster rating Cluster 7: Opportunities in daily routine Having year-round sports activities (e.g., swim team) School programs that get you active (e.g., gardening classes, ballet, and dance) Park in walking distance from home and a lot of kids your age there Having play time after finishing work Walk to or in the neighborhood park Walk to the store Average cluster rating Cluster 8: Pressure from social networks Most of your friends are healthy and fit, you want to keep up Friends pushing you to the limit Family history of obesity, diabetes, and heart failure Taking care of younger kids (i.e., babysitting) Getting money for chores including physical activity (e.g., cutting grass, washing car, and raking leaves) Teachers commenting about being active Girlfriend/boyfriend wants you to stay slim Average cluster rating Cluster 9: Seeing consequences of activity/inactivity Needing to keep a certain weight for sports team, job, etc. See obese kids your age (do not want to look like that) Having personal trainer/mentor Seeing people who are in shape Seeing people riding bikes on the street Watching fit TV-cardio and step and muscle endurance Television/magazine commercials about weight loss Seeing people who are not in shape Watching people on television who are fit Watching workout tapes with friends Average cluster rating
4.04
a b
4.04 3.96 3.96 3.82 3.81 3.75 3.45 3.36 2.98 3.75 4.41b 4.05 3.93 3.70 2.68 3.75 4.00 3.95 3.86 3.78 3.02 3.72 4.47b 3.89 3.88 3.16 3.07 2.69 3.53 3.98 3.66 3.65 3.57 3.48 3.23 2.93 3.50 3.98 3.84 3.51 3.34 3.33 3.22 3.18 3.09 2.96 2.91 3.34
On a 1 (least important) to 5 (most important) rating scale. Top 10 items by average importance rating for all participants.
neighbors jogging or walking in their neighborhood) further support suggestions that neighborhood social and environmental features supportive of physical activity may increase physical activity-related outcomes among adolescents [33].
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542 Table 3 Cluster match by gender Cluster
Cluster names
r value
1 2 3 4 5 6 7 8 9
Access/availability of physical activity resources Family and friends support Physical activity with friends Physical activity with family members Inspiration to/from others Parental reinforcement Opportunities in daily routine Pressure from social network Seeing consequences of activity/inactivity
.71 .78 .80 .47 .92 .86 .85 .10 .59
r Value indicates agreement between ratings of boys and girls.
Although physical activity-related encouragement and motivational support from friends are generally reported as significant predictors of physical activity, this relationship may be complex. Evidence from a systematic review on social networks and physical activity suggests adolescents are more likely to participate in physical activity when a high proportion of their friends are physically active [34]. Peer modeling is another facilitator of physical activity among adolescents [34,35]. This suggests that mere advice or encouragement to be active may not be as effective as actual engagement in physical activity by members of social network. Adolescents reported seeing the consequences of physical activity/inactivity as a facilitator of physical activity but it was rated least important relative to other factors. This cluster included media influence such as watching people on television who are fit and commercials/ads about weight loss. This finding is consistent with prior research with African-American female adolescents suggesting celebrities are not influential role models for physical activity, and body size preferences are less determined by people they do not know [36]. As such, intrasocial network factors may be better motivators of physical activity, especially for African-American girls. Although there is strong agreement between boys and girls on the major clusters of influence and the relative importance of clusters, some differences are noted. One area of divergence relates to competition (e.g., contests among friends) as a motivator of physical activity for boys but not girls. This finding is supported by previous research [37]. Additionally, boys suggest that reinforcement and encouragement for physical activity from both peer and family influences are motivators. In contrast, girls report that parental reinforcement and encouragement are more salient than that received from friends. This finding also contradicts prior work [37] suggesting both boys and girls seek support from parents and friends. However, findings from the present study and others [33,38,39] point to parental support increasing physical activity more so for African-American girls. Top rated statements for boys to be physically active rely mostly on internal motivations (e.g., being a role model), whereas among girls, the most important statements relate to external motivations for physical activity (e.g., family wants you to stay in shape to live longer). These findings may suggest that to increase physical activity, girls require more encouragement from their social networks, whereas boys may need to be equipped with the skills to increase their internal motivators. In addition, girls perceive safety to be an important issue, which is not rated highly for boys. Similar findings are reported by Ries et al. [40] where girls but not boys report safety as a determinant for physical activity. However, for the present study results, it is
541
unclear what specific elements of safety are important. Although some qualitative research indicates that elements of safety are important for physical activity [37], our empirical research [27] with a larger cohort of African-American adolescents from the same community as the present study, did not find support for perceived neighborhood disorder as a barrier to physical activity levels. The current findings suggest that future interventions for African-American adolescents should address gender differences in facilitators of physical activity. For instance, potential intervention strategies for boys might focus on the benefits of role modeling physical activity for others, identifying competitiverelated activities, identifying strategies to involve older siblings in active play with their younger siblings, and providing information about community-specific resources for physical activity. In contrast, girls may benefit from intervention strategies focused on enhancing supportive social networks, identifying physical activities that involve teamwork, providing messaging for parents to encourage physical activity and relate it to health, and mapping areas in neighborhoods and communities that girls (and their parents) feel safe to engage in physical activity. Future physical activity and obesity prevention/treatment interventions that address these gender-specific motivations may improve intervention outcomes. Limitations Although informative, this study is not without limitations. The sample for this study was small and may not be representative of youth in this region.Most of the study participants were from households with incomes greater than the median income. As such, influencing factors related to youth from lower SES families may not be well represented in our findings. Similarly, more than half of the participants perceived themselves as normal weight, and we did not objectively measure weight and height to classify obesity status. On the other hand, given the noted lack of consistent association between SES and obesity among minority adolescents [4,7], findings from this study may still capture salient influential factors on physical activity for many African-American youth. Strengths of the study This study is unique in its use of a participatory methodology to explore social and cultural environment influencing factors on physical activity among African-American youth. Given the higher risk of physical inactivity, obesity, and related comorbidities among this population, having the perspectives of this group are key to future public health interventions. As used in this study, concept mapping increases the robustness of qualitative findings by combining with quantitative analysis to produce outputs (e.g., concept maps) that show relationships between individual or group ideas and present them in a way that is easy to convey to stakeholders [29]. In the present study, this approach helped to generate potential physical activity intervention strategies that are ecologically valid and work within the context of adolescents’ everyday lives and environments. Additionally, because the ideas originated from the target audience, future interventions with this population may be more acceptable. This study incorporates a mixed-methods approach to identify and examine facilitators of physical activity among
542
M.L. Baskin et al. / Journal of Adolescent Health 56 (2015) 536e542
African-American adolescents. Our results suggest that factors related to access and availability, the importance of family and friend support for physical activity and physical activity with friends are key determinants of physical activity among African-American adolescents. Furthermore, these results are strengthened by the high level of agreement between boys and girls, suggesting that the development of interventions that address these similarly rated domains will be of benefit to both gender groups. Future interventions to increase physical activity among African-American adolescents should include strategies to strengthen the social and cultural environment, particularly related to the factors identified by this study. References [1] Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA 2014;311:806e14. [2] Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in overweight among US children and adolescents, 1999-2000. J Am Med Assoc 2002;288:1728e32. [3] Malnick SD, Knobler H. The medical complications of obesity. QJM 2006;99: 565e79. [4] Rossen LM. Neighbourhood economic deprivation explains racial/ethnic disparities in overweight and obesity among children and adolescents in the USA. J Epidemiol Community Health 2014;68:123e9. [5] Shrewsbury V, Wardle J. Socioeconomic status and adiposity in childhood: A systematic review of cross-sectional studies 1990e2005. Obesity (Silver Spring) 2008;16:275e84. [6] Janssen I, Boyce WF, Simpson K, Pickett W. Influence of individual- and arealevel measures of socioeconomic status on obesity, unhealthy eating, and physical inactivity in Canadian adolescents. Am J Clin Nutr 2006;83:139e45. [7] Gordon-Larsen P, Adair LS, Popkin BM. The relationship of ethnicity, socioeconomic factors, and overweight in US adolescents. Obes Res 2003; 11:121e9. [8] Bethell C, Read D, Goodman E, et al. Consistently inconsistent: A snapshot of across- and within-state disparities in the prevalence of childhood overweight and obesity. Pediatrics 2009;123(Suppl 5):S277e86. [9] Ross R, Bradshaw AJ. The future of obesity reduction: Beyond weight loss. Nat Rev Endocrinol 2009;5:319e25. [10] Rowland TW. Promoting physical activity for Children’s Health: Rationale and strategies. Sports Med 2007;37:929e36. [11] Eaton D, Kann L, Kinchen S, et al. Youth risk behavior surveillanceeUnited States, 2011. MMWR Surveill Summ 2012;61:1e162. [12] Troiano RP, Berrigan D, Dodd KW, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc 2008;40:181e8. [13] Sallis J. Age-related decline in physical activity: A synthesis of human and animal studies. Med Sci Sports Exerc 2000;32:1598e600. [14] Kimm SYS, Glynn NW, Kriska AM, et al. Decline in physical activity in black girls and white girls during adolescence. N Engl J Med 2002;347:709e15. [15] Springer AE, Hoelscher DM, Kelder SH. Prevalence of physical activity and sedentary behaviors in US high school students by metropolitan status and geographic region. J Phys Activity Health 2006;3:365. [16] Sallis JF, Cervero RB, Ascher W, et al. An ecological approach to creating active living communities. Annu Rev Public Health 2006;27:297e322. [17] Fitzpatrick KM, LaGory M. “Placing” health in an urban sociology: Cities as mosaics of risk and protections. City & Community 2003;2:33e46. [18] Lawman H, Wilson D. Associations of social and environmental supports with sedentary behavior, light and moderate-to-vigorous physical
[19]
[20]
[21]
[22]
[23] [24]
[25] [26] [27]
[28]
[29] [30] [31] [32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
activity in obese underserved adolescents. Int J Behav Nutr Phys Act 2014;11:92. Edwardson CL, Gorely T. Parental influences on different types and intensities of physical activity in youth: A systematic review. Psychol Sport Exerc 2010;11:522e35. Van Der Horst K, Paw M, Twisk J, Van Mechelen W. A brief review on correlates of physical activity and sedentariness in youth. Med Sci Sports Exerc 2007;39:1241e50. Barr-Anderson DJ, Adams-Wynn AW, DiSantis KI, Kumanyika S. Family-focused physical activity, diet and obesity interventions in African-American girls: A systematic review. Obes Rev 2013;14:29e51. Pasick R, D’Onofrio C, Otero-Sabogal R. Similarities and differences across cultures: Questions to inform a third generation for health promotion research. Health Educ Q 1996;23:S142e61. Kumanyika SK. Environmental influences on childhood obesity: Ethnic and cultural influences in context. Physiol Behav 2008;94:61. Airhihenbuwa CO, Kumanyika S, Agurs TD, Lowe A. Perceptions and beliefs about exercise, rest, and health among African-Americans. Am J Health Promot 1995;9:426e9. Ruef M, Fletcher B. Legacies of American slavery: Status attainment among southern blacks after emancipation. Soc Forces 2003;82:445e80. Babey SH, Hastert TA, Yu H, Brown ER. Physical activity among adolescents: When do parks matter? Am J Prev Med 2008;34:345. Dulin-Keita A, Kaur Thind H, Affuso O, Baskin M. The associations of perceived neighborhood disorder and physical activity with obesity among African American adolescents. BMC Public Health 2013;13:440. Trochim W, Kane M. Concept mapping: An introduction to structured conceptualization in health care. Int J Qual Health Care 2005;17: 187e91. Kane M, Trochim WMK. Concept mapping for planning and evaluation, Vol. 50. Thousand Oaks, CA: Sage Publications; 2007. Gallagher M, Hares T, Spencer J, et al. The nominal group technique: A research tool for general practice? Fam Pract 1993;10:76e81. Rosas SR, Kane M. Quality and rigor of the concept mapping methodology: A pooled study analysis. Eval Program Plann 2012;35:236e45. Barr-Anderson DJ, Singleton C, Cotwright CJ, et al. Outside-of-school time obesity prevention and treatment interventions in African American youth. Obes Rev 2014;15:26e45. Wilson D, Lawman H, Segal M, Chappell S. Neighborhood and parental supports for physical activity in minority adolescents. Am J Prev Med 2011; 41:399e406. Macdonald-Wallis K, Jago R, Sterne JAC. Social network analysis of childhood and youth physical activity: A systematic review. Am J Prev Med 2012;43:636e42. Salvy SJ, de la Haye K, Bowker JC, Hermans RCJ. Influence of peers and friends on children’s and adolescents’ eating and activity behaviors. Physiol Behav 2012;106:369e78. Boyington JEA, Carter-Edwards L, Piehl M, et al. Cultural attitudes toward weight, diet, and physical activity among overweight African American girls. Prev Chronic Dis 2008;5:A36. St George SM, Wilson DK. A qualitative study for understanding family and peer influences on obesity-related health behaviors in low-income African-American adolescents. Child Obes 2012;8:466e76. Adkins S, Sherwood NE, Story M, Davis M. Physical activity among African-American Girls: The role of parents and the home environment. Obes Res 2004;12 suppl:38Se45S. Wright MS, Wilson DK, Griffin S, Evans A. A qualitative study of parental modeling and social support for physical activity in underserved adolescents. Health Educ Res 2010;25:224e32. Ries AV, Voorhees CC, Gittelsohn J, et al. Adolescents’ perceptions of environmental influences on physical activity. Am J Health Behav 2008;32: 26e39.