Skip to main content
PMC home page
Preventive Medicine Reports logoLink to Preventive Medicine Reports
. 2015 Jan 10;2:57–64. doi: 10.1016/j.pmedr.2015.01.001

Individual, social and environmental correlates of physical activity in overweight and obese African American and Hispanic women: A structural equation model analysis

Scherezade K Mama a,b,c,⁎, Pamela M Diamond b, Sheryl A McCurdy b, Alexandra E Evans d, Lorna H McNeill a, Rebecca E Lee e
PMCID: PMC4327909  NIHMSID: NIHMS654989  PMID: 25692093

Abstract

Objective

Ecologic frameworks account for multilevel factors related to physical activity (PA) and may be used to develop effective interventions for women. The purpose of this study was to examine the influence of individual, social and environmental factors on PA among African American and Hispanic women using structural equation modeling.

Methods

Overweight and obese women (N = 164, 65.9% African American) completed a 7-day accelerometer protocol, a physical assessment, and questionnaires on body image, self-efficacy, motivational readiness, social support, home environment for physical activity and perceived environment. Trained assessors evaluated each participant's neighborhood and collected objective measures of physical activity resources and the pedestrian environment. Assessments were completed between 2006 and 2008.

Results

Structural model fit was acceptable (RMSEA = .030). Body composition and image was negatively associated with PA, and motivational readiness had an indirect effect on PA through body composition and image. PA resources and the pedestrian environment operated through the perceived environment to positively influence neighborhood cohesion, which was positively associated with body composition and image.

Conclusion

PA is more heavily influenced by intrapersonal factors related to weight. Improving intrapersonal factors related to weight and perceptions of the environment may lead to increased PA in African American and Hispanic women.

Keywords: Exercise, Body image, Social environment, Environment design, Minority health, Adults

Highlights

  • •

    Physical inactivity contributes to health disparities among ethnic minority women.

  • •

    Individual factors directly and indirectly influence physical activity.

  • •

    The physical environment influences individual factors via the social environment.

Introduction

Physical inactivity significantly contributes to the U.S. mortality rate and burden of disease (Derby et al., 2011, Sanderson et al., 2010, U.S. Burden of Disease Collaborators, 2013, Zhao et al., 2011), yet only 20.9% of adults do enough exercise to meet physical activity guidelines (Centers for Disease Control and Prevention, 2011). Disparities in physical activity persist among women and ethnic minorities, increasing disease risk (Centers for Disease Control and Prevention, 2007).

Individually-focused programs fail to achieve sustainable increases in physical activity in ethnic minority women (Fleury and Lee, 2006, Granner et al., 2007). Ecologic frameworks account for factors beyond the individual, including intrapersonal factors, interpersonal relationships, and the physical environment, and how they influence health behaviors (Egger et al., 2003, Martinez et al., 2009, McNeill et al., 2006, Sallis and Owen, 2008). Spence and Lee broadly divided variables framed in ecologic models into two categories, intra-individual and extra-individual. Intra-individual factors related to physical activity include an individual's weight, attitude, and exercise self-efficacy (Spence and Lee, 2003). Extra-individual factors include social factors, like neighborhood cohesion and social support (Anderson et al., 2010, Lackey and Kaczynski, 2009), and environmental factors (Lee et al., 2011a, McAlexander et al., 2009), which have been shown to be important for physical activity. Each level of the ecologic model has the ability to influence physical activity directly or indirectly through one or more of the other levels (Spence and Lee, 2003).

It is important to understand the determinants of physical activity that may be unique or central to ethnic minority women and how they are related for effective behavior change. Although several studies have looked at various factors within the social and physical environment and how they relate to physical activity (McNeill et al., 2006, Sallis et al., 1997, Sallis et al., 2002a), few have explored these simultaneously using structural equation modeling (SEM), and almost none have done so exclusively in ethnic minority women. The purpose of this study was to examine the relative influence of individual, social and environmental factors on physical activity among African American and Hispanic women in the Southern U.S.

Methods

Study design

This study used baseline data from Health Is Power (HIP), which aimed to increase physical activity in African American and Hispanic women. HIP study details have been published previously (Lee et al., 2011b, Lee et al., 2011c, Lee et al., 2012b). For the present study, individual questionnaire data were linked to participants' environmental data in Harris County, Houston and Travis County, Austin, Texas. All HIP assessments, measures and procedures were approved by the Committee for the Protection of Human Subjects at the University of Houston, and participants provided written informed consent.

Participants

Women were recruited to the study and participated in HIP from 2006 to 2008. Eligible participants were African American or Hispanic, aged 25–60 years, English or Spanish speakers, Harris or Travis County residents, not planning to move during the study, doing < 90 min of physical activity per week, and free from health conditions that would be aggravated by physical activity (Thomas et al., 1992). Participants with complete individual and environmental data (N = 164) were included in the current study.

Procedures

Eligible participants attended a baseline health assessment, where they completed interviewer-administered questionnaires and a physical assessment and received an accelerometer. Participants' addresses were geocoded, and their neighborhoods were mapped using ArcGIS 9.1 (Esri, Redlands, CA; Parmenter et al., 2008). Neighborhood assessments were conducted by trained research team members in teams of two following established data collection and safety protocols (Heinrich et al., 2007, Lee et al., 2005).

Conceptual model

A conceptual model relating individual, social and environmental factors to physical activity was developed using the existing scientific literature. An extensive literature review was completed, focusing on studies with similar correlates as those in HIP in order to be able to test the model using existing data. The model was revised based on findings from in-depth interviews conducted with HIP participants (Mama et al., in press). Fig. 1 shows the direct effects of individual, social and environmental factors on physical activity and the indirect effects among factors. Latent constructs and the pathways included and excluded in the model were based on ecologic models of health behavior, empirical evidence derived from the literature, and findings from in-depth interviews.

Fig. 1.

Fig. 1

Open in a new tab

Conceptual structural model of individual, social and environmental influences on physical activity.

Measures

Physical activity

Objective physical activity data were collected over 7 days using the ActiGraph GT1M accelerometer (ActiGraph, LLC, Pensacola, FL). Accelerometer data were collected as counts per 60 s and translated into minutes spent in moderate-vigorous physical activity (MVPA) per day for a seven day period using an individual cutpoint (Layne et al., 2011). The average number of MVPA per day was used in analyses.

Individual factors

Items assessing household income and education were drawn from the Maternal Infant Health Assessment survey (California Department of Public Health, 2010, Sarnoff and Hughes, 2005). Measures of body mass index (BMI = kg/m2) and body fat were collected by trained personnel using established protocols (Lee et al., 2011b). Pulvers et al.'s (2004) culturally relevant body image questionnaire was used to measure perceived body image, and has shown good validity and reliability in minority populations. Participants chose a silhouette that most closely resembled them currently from a scale of figures representing BMI measures of 16 through 40 kg/m2.

Psychosocial factors related to physical activity included self-efficacy and motivational readiness. Self-efficacy was measured using Bandura's Exercise Self-Efficacy Scale, an 18-item scale with answer choices in 10-unit intervals, ranging from 0 (cannot do) to 100 (certainly can do; Bandura, 1997), which has demonstrated high reliability in women (Shin et al., 2006). The Weight Stages of Change (WSC) short-form was used to measure weight loss intentions and readiness to engage in weight control. The WSC contains 4 items to determine stage of weight change, pre-contemplation, contemplation, action or maintenance (University of Rhode Island Cancer Prevention Research Center, 2008), and has shown similar patterns to movitational readiness for exercise for pros and cons by stage of change (Prochaska et al., 1994).

Social factors

Social support for physical activity was measured using the social support and exercise survey. The survey has five items to measure family support and five items to measure peer support and has shown high internal consistency and test re-test reliability (Sallis et al., 2002b). Neighborhood cohesion was measured using five items, which have shown high reliability as a measure of social cohesion and control at the neighborhood level (Sampson et al., 1997).

Environmental factors

The Physical Activity Resource Assessment (PARA, ©2010) was used as an objective measure of physical activity resources in participants' neighborhoods (Lee et al., 2005). Resources were audited on dimensions of presence and quality of physical activity features and amenities, and higher feature and amenity scores indicated a better quality physical activity resource (Lee et al., in press).

The Pedestrian Environmental Data Scan (PEDS) instrument was used as an objective measure of environmental features and pedestrian facilities related to walking and cycling (Clifton et al., 2007). Variables specifically used in this study were percent of segments with pedestrian facilities and the presence of buffers, path connections and lighting. These variables were selected to reduce multicollinearity with other PEDS variables and are consistent with previously reported relationships (Lee et al., 2012a).

The Physical Activity Neighborhood Environment Survey (PANES) was used to measure participants' self-reported perceptions of their neighborhood environment (Bauman et al., 2009). The PANES contains 17 items assessing perceptions of the neighborhood environment for walking and cycling and has been shown to have high reliability and content and criterion validity with respective environmental attributes for physical activity (Sallis et al., 2009, Sallis et al., 2010).

Home factors related to physical activity were assessed using the Home Environment scale. Participants were asked to self-report supplies or pieces of equipment that can be used for physical activity in their home using a list of 15 items. The scale has demonstrated high validity for different types of exercise (Sallis et al., 1997). Only items used specifically for exercise (unmovable aerobic equipment, running shoes, weights, equipment to tone muscles, aerobic training videos, and step aerobic equipment) were included in analyses.

Statistical analysis

Descriptive statistics were computed for all variables by ethnicity, and t-tests were used to compare differences. To meet assumptions of normality, physical activity data were transformed using an exponential transformation. Objectively-measured environmental factors, including PARA and PEDS data, were aggregated to the neighborhood level. Confirmatory factor analyses were used to confirm the relationships between latent variables and indicator variables; SEM was used to determine the direct and indirect effects of individual, social and environmental factors on physical activity. Although 200 is typically the goal for SEM research, we determined our sample size of 164 was large enough to assess model fit based on the number of degrees of freedom and use of full information maximum likelihood (FIML) estimates for missing data (Little and Rubin, 2002, Loehlin, 2004, Muthén and Muthén, 1998–2012, Tanaka, 1987). All models were tested in Mplus version 7 (Muthén & Muthén, Los Angeles, CA). FIML estimates with the chi-square fit statistic, the Tucker–Lewis index (TLI), the comparative fit index (CFI), and the root mean square error of approximation (RMSEA) were used as indices of goodness of fit (Gefen et al., 2011, McDonald and Ho, 2002, Schreiber et al., 2006).

Results

Participant characteristics

Of the 164 participants included in the current study, 65.9% were African American. Women were in their mid-40s (M = 45.6 ± 9.3 years), 20.9% were overweight, and 71.8% were obese (M BMI = 34.9 ± 7.8 kg/m2). Most participants completed some college (41.1%) or were college graduates (49.7%), and over half (56.9%) reported a household income greater than $82,600 (above 400% of the federal poverty level in 2007). African American women did more physical activity (25.0 versus 7.9 MVPA min/day; t = 7.1 p < .001) than Hispanic women. Therefore, we controlled for race/ethnicity in all subsequent models. Table 1 presents descriptive statistics for all variables included in the model.

Table 1.

Descriptive statistics for indicator and outcome variables.

African American (N = 108)
M (SD)		 Hispanic or Latina (N = 56)
M (SD)		 Total (N = 164)
M (SD)
Intrapersonal factors related to physical activity
 Motivational readiness for weight loss
 Categorical (range: 1–4) 2.9 (0.7) 3.0 (0.7) 3.0 (0.7)
 Body composition and image
 BMI (kg/m2) 34.8 (7.8) 35.1 (7.8) 34.9 (7.8)
 Percent body fat 42.5 (7.0) 43.4 (6.5) 42.8 (6.8)
 Perceived BMI (kg/m2)⁎⁎ 28.7 (6.9) 32.6 (5.7) 30.0 (6.8)
 Exercise self-efficacy⁎
 Scale (range: 0–100) 48.3 (20.9) 42.2 (19.1) 46.2 (20.4)
Interpersonal relationships
 Social support for physical activity
 From family (range: 0–20) 8.7 (4.3) 9.6 (4.5) 9.0 (4.3)
 From peers (range: 0–20) 13.1 (3.8) 13.2 (3.3) 13.1 (3.6)
 Neighborhood cohesion
 Neighbors help each other (range: 1–5) 3.7 (1.0) 3.7 (0.9) 3.7 (1.0)
 Close knit neighborhood (range: 1–5) 3.3 (1.0) 3.1 (1.1) 3.3 (1.0)
 Neighbors can be trusted (range: 1–5) 3.4 (0.8) 3.5 (1.0) 3.4 (0.9)
 Neighbors get along (range: 1–5) 4.0 (0.9) 3.9 (0.7) 3.9 (0.8)
 Neighbors share values (range: 1–5) 3.5 (1.0) 3.3 (1.1) 3.4 (1.0)
Neighborhood and home environment
 Physical activity resources
 Features (range: 0–39) 6.8 (3.7) 7.4 (4.4) 7.0 (4.0)
 Amenities (range: 0–36) 11.9 (5.9) 11.0 (4.8) 11.6 (5.6)
 Pedestrian environment
 Percent of segments with pedestrian facilities 72.7 (27.8) 73.3 (25.1) 72.9 (26.8)
 Number of sidewalk buffers⁎⁎ 1.0 (0.5) 0.7 (0.4) 0.9 (0.5)
 Number of path connections 3.3 (1.0) 3.0 (0.9) 3.2 (1.0)
 Lighting⁎⁎ 1.0 (0.1) 0.8 (0.2) 0.9 (0.2)
 Perceived neighborhood environment
 Scale (range: 1–5) 2.7 (0.5) 2.8 (0.4) 2.7 (0.4)
 Home environment
 Percent with unmovable aerobic equipment 40.7 (44) 37.5 (21) 39.6 (65)
 Percent with running shoes 92.6 (100) 87.5 (49) 90.9 (149)
 Percent with weights 64.8 (70) 67.9 (38) 65.9 (108)
 Percent with equipment to tone muscles 56.5 (61) 41.8 (23) 51.5 (84)
 Percent with aerobic training videos⁎⁎ 80.6 (87) 55.4 (31) 72.0 (118)
 Percent with step aerobic equipment⁎⁎ 36.1 (39) 12.5 (7) 28.0 (46)
 Physical activity
 Accelerometer-measured (min/day)⁎⁎ 25.0 (21.4) 7.9 (7.8) 19.1 (19.7)
Open in a new tab
⁎

p < .05.

⁎⁎

p ≤ .001

Measurement model

The measurement model included nine latent constructs measured by 66 indicator variables. Correlations among indicators across constructs (N = 201; results available upon request) ranged from − .375 to .495, with the greatest number of correlations occurring between neighborhood cohesion and perceived environment (n = 52) and exercise self-efficacy and home environment (n = 31). The overall fit of the measurement model was acceptable based on fit indices (Χ2(1994) = 2491.6, RMSEA = .027, RMSEA 90% CI = [.024,.031], TLI = .939, CFI = .944, SRMR = .051). Final standardized and unstandardized factor loadings for the measurement model are shown in Table 2.

Table 2.

Unstandardized (b) and standardized (β) and coefficients for measurement model.

b β SEa pa
Intrapersonal factors related to physical activity
 Motivational readiness
 Percent trying to lose weight 1.000 0.144 0.089 .103
 Percent trying not to gain weight 1.412 0.233 0.120 .052
 Percent trying to reach weight goal 3.012 0.926 0.431 .032
 Exercise self-efficacy (ESE)
 When feeling tired 1.000 0.649 0.034 < .001
 When under pressure from work 1.272 0.720 0.029 < .001
 During bad weather 1.257 0.656 0.034 < .001
 After recovering from an injury 1.112 0.727 0.028 < .001
 During/after experiencing personal problems 1.334 0.768 0.025 < .001
 When feeling depressed 1.355 0.721 0.029 < .001
 When feeling anxious 1.290 0.719 0.029 < .001
 After recovering from an illness 1.120 0.722 0.028 < .001
 When feeling physical discomfort 1.081 0.678 0.032 < .001
 After a vacation 1.396 0.768 0.025 < .001
 When there's too much work at home 1.310 0.766 0.025 < .001
 When visitors are present 1.267 0.700 0.030 < .001
 When there are other interesting things to do 1.320 0.754 0.026 < .001
 Without reaching exercise goals 1.367 0.777 0.024 < .001
 Without support from family or friends 1.302 0.719 0.029 < .001
 During a vacation 1.060 0.574 0.039 < .001
 When there are other time commitments 1.188 0.752 0.025 < .001
 After experiencing family problems 1.418 0.782 0.024 < .001
 Body composition and image
 BMI (kg/m2) 1.000 0.939 0.047 < .001
 Percent body fat 0.782 0.870 0.046 < .001
 Perceived BMI (kg/m2) 0.289 0.352 0.052 < .001
Interpersonal relationships
 Social support
 Family encouraged me to do physical activities or play sports 1.000 0.648 0.037 < .001
 Family did physical activity or played sports with me 0.885 0.782 0.027 < .001
 Family provided transportation to place to do physical activities or play sports 0.844 0.713 0.032 < .001
 Family watched me participate in physical activities or sports 0.836 0.780 0.027 < .001
 Family told me I'm doing well in physical activities or sports 0.989 0.769 0.028 < .001
 Encouraged friends to do physical activities or play sports 0.384 0.282 0.055 < .001
 Friends encouraged me to do physical activities or play sports 0.349 0.264 0.055 < .001
 Friends did physical activities or played sports with me 0.358 0.299 0.054 < .001
 Friends teased me for not being good at physical activity sports − 0.086 − 0.106 0.051 0.069
 Friends told me I'm doing well in physical activities or sports 0.456 0.395 0.059 < .001
 Percent trying to lose weightb − 0.104 − 0.181 0.052 .001
 Neighborhood cohesion
 Neighbors help each other 1.000 0.685 0.042 < .001
 Close knit neighborhood 1.156 0.756 0.038 < .001
 Neighbors can be trusted 1.094 0.803 0.036 < .001
 Neighbors get along 0.596 0.496 0.049 < .001
 Neighbors share values 0.565 0.396 0.059 < .001
 Crime makes it unsafe during dayb − 0.367 − 0.227 0.061 < .001
Neighborhood and home environment related to physical activity
 Physical activity resources
 Features 1.000 1.404 0.451 .002
 Amenities 0.473 0.483 0.160 .003
 Pedestrian environment
 Percent of segments with pedestrian facilities 1.000 1.001 0.032 < .001
 Number of sidewalk buffers 1.356 0.712 0.034 < .001
 Number of path connections 1.644 0.446 0.051 < .001
 Lighting 0.145 0.226 0.067 < .001
 Sidewalks on streetsb 1.600 0.426 0.049 .001
 Sidewalks are well maintainedb 0.707 0.170 0.050 < .001
 Perceived neighborhood environment
 Shops and stores nearby 1.000 0.260 0.061 < .001
 Transit stop nearby − 0.415 − 0.104 0.064 .103
 Sidewalks on streets 0.666 0.209 0.057 < .001
 Bicycle facilities nearby 1.860 0.421 0.056 < .001
 Recreation facilities in neighborhood 1.512 0.466 0.053 < .001
 Crime makes it unsafe at night − 1.785 − 0.465 0.054 < .001
 Traffic makes it difficult to walk − 1.705 − 0.498 0.054 < .001
 See people being physically active 1.929 0.638 0.043 < .001
 Interesting things to look at 1.734 0.536 0.049 < .001
 Many four-way intersections − 0.451 − 0.133 0.063 .035
 Sidewalks are well maintained 1.581 0.447 0.053 < .001
 Bicycle facilities are well maintained 2.072 0.473 0.053 < .001
 Traffic makes it difficult to bike − 1.347 − 0.360 0.058 < .001
 Crime makes it unsafe during dayb − 1.186 − 0.350 0.068 < .001
 Many places within walking distance 1.634 0.471 0.053 < .001
 Home environment
 Unmovable aerobic equipment 1.000 0.209 0.066 .001
 Running shoes 0.795 0.281 0.064 < .001
 Weights 2.448 0.546 0.056 < .001
 Equipment to tone muscles 3.062 0.634 0.053 < .001
 Aerobic training videos 2.343 0.537 0.057 < .001
 Step aerobic equipment 2.233 0.499 0.058 < .001
Physical activity 1.000 1.000 .001
Open in a new tab
a

Standard errors and p-values reported for standardized estimates.

b

Indicates cross-loading on multiple constructs.

Structural model

The proposed structural model fit was acceptable (Χ2(2133) = 2786.4, RMSEA = .030, RMSEA 90% CI = [.027, .033], TLI = .923, CFI = .928). The final model with significant pathways and standardized coefficients is shown in Fig. 2, and standardized and unstandardized coefficients for indirect pathways are shown in Table 3. Body composition and image had a significant direct effect on physical activity (β = − 0.307, p = .001) and mediated the relationship between motivational readiness and physical activity (β = 0.111, p = .038). Perceived environment mediated the relationship between physical activity resources and neighborhood cohesion (β = 0.156, p = .001) and social support for physical activity (β = 0.054, p = .016). Perceived environment also mediated the relationship between the pedestrian environment and neighborhood cohesion (β = 0.157, p < .001) and social support for physical activity (β = 0.054, p = .011). There was also a significant indirect association between perceived environment and body composition and image through neighborhood cohesion (β = 0.082, p = .046).

Fig. 2.

Fig. 2

Open in a new tab

Structural model with significant pathways and standardized (β) path coefficients.

Note: Pathways with positive path coefficients are indicated in bold, and dashed lines indicate pathways with negative path coefficients. Pathways from the conceptual model that were not significant in the structural model are not shown.

Table 3.

Unstandardized (b) and standardized (β) coefficients for indirect pathways in structural model.

b β SEa pa
Motivational readiness →
 Body composition and image − 26.192 − 0.360 0.105 .001
Social support for physical activity →
 Motivational readiness − 0.002 − 0.014 0.095 .879
 Exercise self-efficacy 0.011 0.006 0.061 .927
 Body composition and image − 0.532 − 0.059 0.064 .353
Neighborhood cohesion →
 Exercise self-efficacy − 0.005 − 0.002 0.064 .974
 Body composition and image 1.590 0.138 0.067 .039
Physical activity resources →
 Neighborhood cohesion − 0.032 − 0.182 0.067 .006
 Perceived neighborhood environment 0.022 0.262 0.068 < .001
Pedestrian environment →
 Neighborhood cohesion − 0.389 − 0.158 0.061 .010
 Perceived neighborhood environment 0.307 0.263 0.060 < .001
Perceived neighborhood environment →
 Motivational readiness 0.045 0.135 0.094 .153
 Social support for physical activity 0.558 0.206 0.065 .002
 Neighborhood cohesion 1.258 0.598 0.066 < .001
 Home environment − 0.032 − 0.088 0.079 .268
Home environment →
 Exercise self-efficacy − 4.319 − 0.292 0.069 < .001
 Body composition and image 3.753 0.056 0.073 .443
Open in a new tab
a

Standard errors and p-values reported for standardized estimates.

Discussion

This study explored the influence of multilevel factors on physical activity in African American and Hispanic women using SEM. Body composition and image was the only factor directly associated with physical activity, suggesting that physical activity is most heavily influenced by individual factors related to weight. Although the complete indirect pathway from the environment to physical activity was not significant, indirect associations among individual, social and environmental factors contribute to our understanding of how multilevel correlates of physical activity are related in ethnic minority women.

In addition to the direct association between body composition and image and physical activity, we found an indirect association between motivational readiness and physical activity operating through body composition. Although previous studies have shown direct associations between individual factors and physical activity (Delahanty et al., 2006, Tavares et al., 2009), few have explored the relationship between motivational readiness for weight loss and physical activity, and none have explored body composition and image as a mediator of this relationship. Perceptions of the environment significantly mediated the relationship between neighborhood and social environments, which indirectly influenced physical activity through body composition and image. The relationship seen between the objectively measured neighborhood environment and the perceived environment is consistent with the current literature (Foster et al., 2013, McAlexander et al., 2012). Environmental perceptions have been found to be an important mediator between the built environment and physical activity and a moderator of the relationship between social support and physical activity, leading to increases in transportation-related and leisure-time walking (Van Dyck et al., 2013, Van Dyck et al., 2014). There is also a known strong, persistent association between physical activity and social support from neighbors, family and friends (Barber, 2013, de Farias Junior et al., 2014, Martinez et al., 2013, Poortinga, 2006). These findings, coupled with the current study's results, suggest that intervention at the environmental level may directly influence the social environment, and that the perceived environment and neighborhood cohesion play a pivotal role in the relationship between the measured built environment and body composition and image, which we found to directly influence physical activity. Thus, improving African American and Hispanic women's perceptions of their neighborhood environment and increasing neighborhood cohesion may lead to increased physical activity adoption and maintenance in this at risk population.

Results from this study highlight the ongoing need to gain a better understanding of what encourages ethnic minority women to do physical activity for health promotion at a population level. While previous quantitative and qualitative studies have emphasized the importance of the social and physical environments, the current study confirms the need to continue to address individual factors, particularly those related to weight, in physical activity interventions targeting African American and Hispanic women. Results from a recent study (Johnson et al., 2013) also support the need to incorporate body image into health behavior models, such as the transtheoretical model (Johnson et al., 2008). Further research is needed to increase understanding of how body composition and image and motivational readiness may directly and indirectly promote physical activity. Future interventions should also focus on improvements to pedestrian facilities and physical activity resources in the neighborhood environment as part of an ecologic approach and important for behavior change in ethnic minority women. Policy changes to enhance the neighborhood environment, such as increasing the presence of sidewalks, improving lighting, and increasing access to high quality physical activity resources, may lead to better perceptions of the environment and increased social support and neighborhood cohesion among residents.

Few studies have sought to explore multilevel influences on physical activity in African American and Hispanic women. We used SEM to determine direct and indirect correlates of physical activity and used an objective measure of physical activity. Models fit the data well and helped explain physical activity in a sizeable sample of African American and Hispanic women. Further research is needed to determine how results compare to non-Hispanic white populations and diverse low-income populations. Although innovative, this study is not without limitations. Due to the limited number of Hispanic women with complete individual and environmental data, we were unable to complete multi-group analyses to determine differences by ethnicity. As this is one of numerous possible models that could explain physical activity in women using these same variables, the addition or removal of proposed pathways may tell a different story and expose additional direct and indirect influences on physical activity, warranting further research. Additionally, our ability to operationalize variables in the model was limited to variables collected as part of HIP. Therefore other known barriers to physical activity, such as crime and perceived safety (Evenson et al., 2012), should be considered as part of future studies and models. Last, the women included in the current study were of high socioeconomic status in contrast to previous studies that have focused on ethnic minorities. Thus, findings from this study may not be generalizable to low socioeconomic status, non-overweight/obese African American and Hispanic women, or those residing in rural areas.

Conclusions

This study confirms the value of using an ecologic framework to understand behavior and the importance of multilevel strategies to influence physical activity. We found that body composition and image and an interplay among individual factors directly and indirectly influenced physical activity, and that neighborhood cohesion played a pivotal role in the relationship between individual and environmental variables. Findings suggest that environmental changes may have a greater impact on physical activity in ethnic minority women when coupled with individual strategies, such as improving body image perceptions and increasing motivational readiness for weight loss.

Conflict of interest

The authors declare that there are no conflicts of interest.

Acknowledgments

Funding for this study was provided by a predoctoral fellowship (NIH F31NR013349) awarded to Scherezade K. Mama by the National Institute of Nursing Research and by a research grant (NIH R01CA109403) awarded to Rebecca E. Lee by the National Cancer Institute. The authors would like to thank the many colleagues, staff, and students who assisted in the collection, processing and management of data throughout the Health Is Power study.

References

  1. Anderson E.S., Winett R.A., Wojcik J.R., Williams D.M. Social cognitive mediators of change in a group randomized nutrition and physical activity intervention: social support, self-efficacy, outcome expectations and self-regulation in the guide-to-health trial. J. Health Psychol. 2010;15:21–32. doi: 10.1177/1359105309342297. [DOI] [PubMed] [Google Scholar]
  2. Bandura A. W. H. Freeman; New York, NY: 1997. Self-efficacy: The Exercise of Control. [Google Scholar]
  3. Barber F.D. Effects of social support on physical activity, self-efficacy, and quality of life in adult cancer survivors and their caregivers. Oncol. Nurs. Forum. 2013;40:481–489. doi: 10.1188/13.ONF.481-489. [DOI] [PubMed] [Google Scholar]
  4. Bauman A., Bull F., Chey T. The International Prevalence Study on Physical Activity: results from 20 countries. Int. J. Behav. Nutr. Phys. Act. 2009;6:21. doi: 10.1186/1479-5868-6-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. California Department of Public Health Maternal and Infant Health Assessment (MIHA) survey. 2010. http://www.cdph.ca.gov/data/surveys/Pages/MaternalandInfantHealthAssessment(MIHA)survey.aspx (Accessed July 27, 2013)
  6. Centers for Disease Control and Prevention Prevalence of regular physical activity among adults—United States, 2001 and 2005. MMWR Morb. Mortal. Wkly Rep. 2007;56:1209–1212. [PubMed] [Google Scholar]
  7. Centers for Disease Control and Prevention . Centers for Disease Control and Prevention; Atlanta, GA: 2011. Behavioral Risk Factor Surveillance System: Prevalence and Trends Data, Nationwide (States and DC) — 2011, Physical Activity, Participated in Enough Aerobic and Muscle Strengthening Exercises to Meet Guidelines. ( http://apps.nccd.cdc.gov/brfss/display.asp?cat=PA&yr=2011&qkey=8291&state=US. Accessed August 27, 2013) [Google Scholar]
  8. Clifton K.J., Livi Smith A., Rodriguez D. The development and testing of an audit for the pedestrian environment. Landsc. Urban Plan. 2007;80:95–110. [Google Scholar]
  9. de Farias Junior J.C., Florindo A.A., Santos M.P., Mota J., Barros M.V. Perceived environmental characteristics and psychosocial factors associated with physical activity levels in adolescents from Northeast Brazil: structural equation modelling analysis. J. Sports Sci. 2014;32:963–973. doi: 10.1080/02640414.2013.873137. [DOI] [PubMed] [Google Scholar]
  10. Delahanty L.M., Conroy M.B., Nathan D.M., Diabetes Prevention Program Research G. Psychological predictors of physical activity in the diabetes prevention program. J. Am. Diet. Assoc. 2006;106:698–705. doi: 10.1016/j.jada.2006.02.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Derby C.A., Wildman R.P., McGinn A.P. Cardiovascular risk factor variation within a Hispanic cohort: SWAN, the Study of Women's Health Across the Nation. Ethn. Dis. 2011;20:396–402. [PMC free article] [PubMed] [Google Scholar]
  12. Egger G., Swinburn B., Rossner S. Dusting off the epidemiological triad: could it work with obesity? Obes. Rev. 2003;4:115–119. doi: 10.1046/j.1467-789x.2003.00100.x. [DOI] [PubMed] [Google Scholar]
  13. Evenson K.R., Block R., Diez Roux A.V., McGinn A.P., Wen F., Rodriguez D.A. Associations of adult physical activity with perceived safety and police-recorded crime: the multi-ethnic study of atherosclerosis. Int. J. Behav. Nutr. Phys. Act. 2012;9:146. doi: 10.1186/1479-5868-9-146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fleury J., Lee S.M. The social ecological model and physical activity in African American women. Am. J. Community Psychol. 2006;37:129–140. doi: 10.1007/s10464-005-9002-7. [DOI] [PubMed] [Google Scholar]
  15. Foster S., Wood L., Christian H., Knuiman M., Giles-Corti B. Planning safer suburbs: do changes in the built environment influence residents' perceptions of crime risk? Soc. Sci. Med. 2013;97:87–94. doi: 10.1016/j.socscimed.2013.08.010. [DOI] [PubMed] [Google Scholar]
  16. Gefen D., Rigdon E.E., Straub D. An update and extension to SEM guidelines for administrative and social science research. MIS Q. 2011;35:iii–xiv. [Google Scholar]
  17. Granner M.L., Sharpe P.A., Hutto B., Wilcox S., Addy C.L. Perceived individual, social, and environmental factors for physical activity and walking. J. Phys. Act. Health. 2007;4:278–293. doi: 10.1123/jpah.4.3.278. [DOI] [PubMed] [Google Scholar]
  18. Heinrich K.M., Lee R.E., Suminski R.R. Associations between the built environment and physical activity in public housing residents. Int. J. Behav. Nutr. Phys. Act. 2007;4:56. doi: 10.1186/1479-5868-4-56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Johnson S.S., Paiva A.L., Cummins C.O. Transtheoretical model-based multiple behavior intervention for weight management: effectiveness on a population basis. Prev. Med. 2008;46:238–246. doi: 10.1016/j.ypmed.2007.09.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Johnson P., Fallon E.A., Harris B.S., Burton B. Body satisfaction is associated with transtheoretical model constructs for physical activity behavior change. Body Image. 2013;10:163–174. doi: 10.1016/j.bodyim.2012.12.002. [DOI] [PubMed] [Google Scholar]
  21. Lackey K.J., Kaczynski A.T. Correspondence of perceived vs. objective proximity to parks and their relationship to park-based physical activity. Int. J. Behav. Nutr. Phys. Act. 2009;6(53) doi: 10.1186/1479-5868-6-53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Layne C.S., Mama S.K., Banda J.A., Lee R.E. Development of an ecologically valid approach to assess moderate physical activity using accelerometry in community dwelling women of color: a cross-sectional study. Int. J. Behav. Nutr. Phys. Act. 2011;8:21. doi: 10.1186/1479-5868-8-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lee R.E., Booth K.M., Reese-Smith J.Y., Regan G., Howard H.H. The Physical Activity Resource Assessment (PARA) instrument: evaluating features, amenities and incivilities of physical activity resources in urban neighborhoods. Int. J. Behav. Nutr. Phys. Act. 2005;2:13. doi: 10.1186/1479-5868-2-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lee R.E., Mama S.K., McAlexander K.P., Adamus H., Medina A.V. Neighborhood and PA: neighborhood factors and physical activity in African American public housing residents. J. Phys. Act. Health. 2011;8(Suppl. 1):S83–S90. doi: 10.1123/jpah.8.s1.s83. [DOI] [PubMed] [Google Scholar]
  25. Lee R.E., Mama S.K., Medina A.V. Multiple measures of physical activity, dietary habits and weight status in African American and Hispanic or Latina women. J. Community Health. 2011;36:1011–1023. doi: 10.1007/s10900-011-9403-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lee R.E., Medina A.V., Mama S.K. Health is power: an ecological theory-based health intervention for women of color. Contemp. Clin. Trials. 2011;32:916–923. doi: 10.1016/j.cct.2011.07.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lee R.E., Mama S.K., Medina A.V., Ho A., Adamus H.J. Neighborhood factors influence physical activity among African American and Hispanic or Latina women. Health Place. 2012;18:63–70. doi: 10.1016/j.healthplace.2011.08.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lee R.E., O'Connor D.P., Smith-Ray R. Mediating effects of group cohesion on physical activity and diet in women of color: health is power. Am. J. Health Promot. 2012;26:e116–e125. doi: 10.4278/ajhp.101215-QUAN-400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lee R.E., Mama S.K., Adamus-Leach H.J., Soltero E.G. Contribution of neighborhood income and access to quality physical activity resources to physical activity in ethnic minority women over time. Am. J. Health Promot. 2014 doi: 10.4278/ajhp.130403-QUAN-148. (in press) [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Little R.J., Rubin D.B. Second ed. John Wiley & Sons; New York, NY: 2002. Statistical Analysis With Missing Data. [Google Scholar]
  31. Loehlin J.C. 4th ed. Psychology Press; New York, NY: 2004. Latent Variable Models: An Introduction to Factor, Path, and Structural Equation Analysis. [Google Scholar]
  32. Mama S.K., McCurdy S.A., Evans A.E., Thompson D.I., Diamond P.M., Lee R.E. Using community insight to understand physical activity adoption in overweight and obese African American and Hispanic women: a qualitative study. Health Educ. Behav. 2014 doi: 10.1177/1090198114557128. (in press), (pii: 1090198114557128) [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Martinez S.M., Arredondo E.M., Perez G., Baquero B. Individual, social, and environmental barriers to and facilitators of physical activity among Latinas living in San Diego County: focus group results. Fam. Community Health. 2009;32:22–33. doi: 10.1097/01.FCH.0000342814.42025.6d. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Martinez S.M., Arredondo E.M., Roesch S. Physical activity promotion among churchgoing Latinas in San Diego, California: does neighborhood cohesion matter? J. Health Psychol. 2013;18:1319–1329. doi: 10.1177/1359105312462433. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. McAlexander K.M., Banda J.A., McAlexander J.W., Lee R.E. Physical activity resource attributes and obesity in low-income African Americans. J. Urban Health. 2009;86:696–707. doi: 10.1007/s11524-009-9385-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. McAlexander K.M., Mama S.K., Medina A.V., O'Connor D.P., Lee R.E. Concordance and correlates of direct and indirect built environment measurement among minority women. Am. J. Health Promot. 2012;26:239–244. doi: 10.4278/ajhp.100715-QUAN-241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. McDonald R.P., Ho M.H. Principles and practice in reporting structural equation analyses. Psychol. Methods. 2002;7:64–82. doi: 10.1037/1082-989x.7.1.64. [DOI] [PubMed] [Google Scholar]
  38. McNeill L.H., Wyrwich K.W., Brownson R.C., Clark E.M., Kreuter M.W. Individual, social environmental, and physical environmental influences on physical activity among black and white adults: a structural equation analysis. Ann. Behav. Med. 2006;31:36–44. doi: 10.1207/s15324796abm3101_7. [DOI] [PubMed] [Google Scholar]
  39. Muthén L.K., Muthén B.O. Seventh ed. Muthén & Muthén; Los Angeles, CA: 1998–2012. Mplus User's Guide. [Google Scholar]
  40. Parmenter B.M., McMillan T., Cubbin C., Lee R.E. Developing geospatial data management, recruitment, and analysis techniques for physical activity research. J. Urban Reg. Inf. Syst. Assoc. 2008;20 [Google Scholar]
  41. Poortinga W. Perceptions of the environment, physical activity, and obesity. Soc. Sci. Med. 2006;63:2835–2846. doi: 10.1016/j.socscimed.2006.07.018. [DOI] [PubMed] [Google Scholar]
  42. Prochaska J.O., Velicer W.F., Rossi J.S. Stages of change and decisional balance for 12 problem behaviors. Health Psychol. 1994;13:39–46. doi: 10.1037//0278-6133.13.1.39. [DOI] [PubMed] [Google Scholar]
  43. Pulvers K.M., Lee R.E., Kaur H. Development of a culturally relevant body image instrument among urban African Americans. Obes. Res. 2004;12:1641–1651. doi: 10.1038/oby.2004.204. [DOI] [PubMed] [Google Scholar]
  44. Sallis J.F., Owen N. Ecologic models of health behavior. In: Glanz K., Lewis F.M., Rimer B.K., editors. Health Behaviour and Health Education: Theory, Research, and Practice. 4th ed. Jossey-Bass; San Francisco, CA: 2008. [Google Scholar]
  45. Sallis J.F., Johnson M.F., Calfas K.J., Caparosa S., Nichols J.F. Assessing perceived physical environmental variables that may influence physical activity. Res. Q. Exerc. Sport. 1997;68:345–351. doi: 10.1080/02701367.1997.10608015. [DOI] [PubMed] [Google Scholar]
  46. Sallis J.F., Kraft K., Linton L.S. How the environment shapes physical activity: a transdisciplinary research agenda. Am. J. Prev. Med. 2002;22:208. doi: 10.1016/s0749-3797(01)00435-4. [DOI] [PubMed] [Google Scholar]
  47. Sallis J.F., Taylor W.C., Dowda M., Freedson P.S., Pate R.R. Correlates of vigorous physical activity for children in grades 1 through 12: comparing parent-reported and objectively measured physical activity. Pediatr. Exerc. Sci. 2002;14:30–44. [Google Scholar]
  48. Sallis J.F., Bowles H.R., Bauman A. Neighborhood environments and physical activity among adults in 11 countries. Am. J. Prev. Med. 2009;36:484–490. doi: 10.1016/j.amepre.2009.01.031. [DOI] [PubMed] [Google Scholar]
  49. Sallis J.F., Kerr J., Carlson J.A. Evaluating a brief self-report measure of neighborhood environments for physical activity research and surveillance: Physical Activity Neighborhood Environment Scale (PANES) J. Phys. Act. Health. 2010;7:533–540. doi: 10.1123/jpah.7.4.533. [DOI] [PubMed] [Google Scholar]
  50. Sampson R.J., Raudenbush S.W., Earls F. Neighborhoods and violent crime: a multilevel study of collective efficacy. Science. 1997;277:918–924. doi: 10.1126/science.277.5328.918. [DOI] [PubMed] [Google Scholar]
  51. Sanderson M., Peltz G., Perez A. Diabetes, physical activity and breast cancer among Hispanic women. Cancer Epidemiol. 2010;34:556–561. doi: 10.1016/j.canep.2010.06.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Sarnoff R., Hughes D. Increasing health insurance coverage in the first year of life. Matern. Child Health J. 2005;9:343–350. doi: 10.1007/s10995-005-0023-5. [DOI] [PubMed] [Google Scholar]
  53. Schreiber J.B., Nora A., Stage F.K., Barlow E.A., King J. Reporting structural equation modeling and confirmatory factor analysis results: a review. J. Educ. Res. 2006;99:323–338. [Google Scholar]
  54. Shin Y.H., Hur H.K., Pender N.J., Jang H.J., Kim M.S. Exercise self-efficacy, exercise benefits and barriers, and commitment to a plan for exercise among Korean women with osteoporosis and osteoarthritis. Int. J. Nurs. Stud. 2006;43:3–10. doi: 10.1016/j.ijnurstu.2004.10.008. [DOI] [PubMed] [Google Scholar]
  55. Spence J.C., Lee R.E. Toward a comprehensive model of physical activity. Psychol. Sport Exerc. 2003;4:7–24. [Google Scholar]
  56. Tanaka J.S. “How big is big enough?”: sample size and goodness of fit in structural equation models with latent variables. Child Dev. 1987;58:134–146. [Google Scholar]
  57. Tavares L.S., Plotnikoff R.C., Loucaides C. Social–cognitive theories for predicting physical activity behaviours of employed women with and without young children. Psychol. Health Med. 2009;14:129–142. doi: 10.1080/13548500802270356. [DOI] [PubMed] [Google Scholar]
  58. Thomas S., Reading J., Shephard R.J. Revision of the Physical Activity Readiness Questionnaire (PAR-Q) Can. J. Sport Sci. 1992;17:338–345. [PubMed] [Google Scholar]
  59. U.S. Burden of Disease Collaborators The state of US health, 1990–2010: burden of diseases, injuries, and risk factors. JAMA. 2013;310:591–608. doi: 10.1001/jama.2013.13805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. University of Rhode Island Cancer Prevention Research Center . Cancer Prevention Research Center; Rhode Island: 2008. 2008. Weight Stages of Change — Short Form, Cancer Prevention Research Center measures. ( http://www.uri.edu/research/cprc/Measures/Weight01.htm. Accessed June 28, 2013) [Google Scholar]
  61. Van Dyck D., Veitch J., De Bourdeaudhuij I., Thornton L., Ball K. Environmental perceptions as mediators of the relationship between the objective built environment and walking among socio-economically disadvantaged women. Int. J. Behav. Nutr. Phys. Act. 2013;10:108. doi: 10.1186/1479-5868-10-108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Van Dyck D., Cerin E., Conway T.L. Interacting psychosocial and environmental correlates of leisure-time physical activity: a three-country study. Health Psychol. 2014;33:699–709. doi: 10.1037/a0033516. [DOI] [PubMed] [Google Scholar]
  63. Zhao G., Ford E.S., Li C., Balluz L.S. Physical activity in U.S. older adults with diabetes mellitus: prevalence and correlates of meeting physical activity recommendations. J. Am. Geriatr. Soc. 2011;59:132–137. doi: 10.1111/j.1532-5415.2010.03236.x. [DOI] [PubMed] [Google Scholar]

Articles from Preventive Medicine Reports are provided here courtesy of Elsevier

RESOURCES