Effectiveness of a Walking Group Intervention to Promote Physical Activity and Cardiovascular Health in Predominantly Non-Hispanic Black and Hispanic Urban Neighborhoods

Abstract

Objectives. The purpose of this study was to evaluate the effectiveness of the Walk Your Heart to Health (WYHH) intervention, one component of the multilevel Community Approaches to Cardiovascular Health: Pathways to Heart Health (CATCH:PATH) intervention designed to promote physical activity and reduce cardiovascular risk among non-Hispanic Black and Hispanic residents of Detroit, Michigan. The study was designed and implemented using a community-based participatory research approach that actively engaged community residents, health service providers and academic researchers. It was implemented between 2009 and 2012. Method. WYHH was a 32-week community health promoter–facilitated walking group intervention. Groups met three times per week at community-based or faith-based organizations, and walked for 45 to 90 minutes (increasing over time). The study used a cluster randomized control design to evaluate effectiveness of WYHH, with participants randomized into intervention or lagged intervention (control) groups. Psychosocial, clinical, and anthropometric data were collected at baseline, 8, and 32 weeks, and pedometer step data tracked using uploadable peisoelectric pedometers. Results. Participants in the intervention group increased steps significantly more during the initial 8-week intervention period, compared with the control group (β = 2004.5, p = .000). Increases in physical activity were associated with reductions in systolic blood pressure, fasting blood glucose, total cholesterol, waist circumference and body mass index at 8 weeks, and maintained at 32 weeks. Conclusion. The WYHH community health promoter–facilitated walking group intervention was associated with significant reductions in multiple indicators of cardiovascular risk among predominantly Hispanic and non-Hispanic Black participants in a low-to-moderate income urban community. Such interventions can contribute to reductions in racial, ethnic, and socioeconomic inequities in cardiovascular mortality.

Keywords

cardiovascular disease community-based participatory research community health promotion health disparities health promotion physical activity/exercise

The purpose of this study was to evaluate the effectiveness of the Walk Your Heart to Health (WYHH) walking group intervention designed to promote physical activity and reduce cardiovascular risk (CVR) among predominantly non-Hispanic Black (NHB) and Hispanic residents of urban communities. Sedentary lifestyles influence multiple risk factors for cardiovascular disease (CVD), the largest contributor to all-cause mortality in the U.S. accounting for one third of the excess mortality experienced by NHBs compared with non-Hispanic Whites (NHWs; Wong, Shapiro, Boscardin, & Ettner, 2002). Nationally, just 48% of U.S. adults meet recommended guidelines for adequate physical activity (U.S. Department of Health and Human Services, 2008a, 2008b). Leisure time physical activity is positively associated with socioeconomic status (SES; Schiller, Lucas, Ward, & Peregoy, 2012), and NHBs and Hispanics report lower levels of leisure time physical activity compared with NHWs (Marshall et al., 2007). NHBs and Hispanics remain disproportionately likely to live in racially segregated neighborhoods with high rates of poverty, and neighborhood characteristics associated with poverty have been linked to heightened behavioral and clinical risk factors for CVD (Moore, Diez Roux, Nettleton, & Jacobs, 2008; Mujahid et al., 2008; Sallis, Floyd, Rodriguez, & Saelens, 2012), including lower levels of physical activity (Echeverria, Diez-Roux, Shea, Borrell, & Jackson, 2008; Sallis et al., 2012). Thus, interventions effective in increasing leisure time physical activity among racial, ethnic, and socioeconomic groups with high CVR stand to make important contributions toward the elimination of persistent CV inequities.

Walking, a widely accessible form of physical activity across SES and age groups (Tudor-Locke, Johnson, & Katzmarzyk, 2010; U.S. Department of Health and Human Services, 2008a, 2008b), poses low risk of injury, does not require formal training or special equipment and can be sustained into old age (U.S. Department of Health and Human Services, 2008a, 2008b) Walking has been demonstrated to reduce multiple indicators of CVR among previously sedentary adults (Ainsworth et al., 2011; Kassavou, Turner, & French, 2013; Murphy, Murtagh, Boreham, Hare, & Nevill, 2006; Murphy, Nevill, Murtagh, & Holder, 2007; Swartz et al., 2003; Tully, Cupples, Chan, McGlade, & Young, 2005) and with substantial reductions in the incidence of cardiovascular events and mortality (Manson et al., 2002; Williams & Thompson, 2013). Interventions to promote walking have been effective in a variety of community contexts (Kassavou et al., 2013; Ogilvie, Craig, Griffin, Macintyre, & Wareham, 2009), and have been shown to be well accepted by previously sedentary participants (Fisher & Li, 2004). Walking group interventions emphasizing strengthening social networks and support for behavior change have been particularly effective (Bjaras, Harberg, Sydhoff, & Ostenson, 2001; Kahn et al., 2002). A recent systematic review of the literature focused on interventions to promote walking in groups concluded that there are no difference in effectiveness between groups facilitated by lay versus professional leaders, with larger effects among groups that include both genders, all ages, and longer duration (Kassavou et al., 2013). In this same review, (Kassavou et al., 2013) noted low homogeneity of results across studies, and identified a number of limitations, including the small number of studies that used objective measures of physical activity, the predominance of NHW participants with limited representation from other racial or ethnic groups, and an inability to compare effectiveness across racial, ethnic, or SES groups (Kassavou et al., 2013). Thus, while the current evidence base supports the benefits of group interventions to promote walking in community settings, there is a need for further evidence regarding the effectiveness of walking group interventions among racial, ethnic, and SES groups with disproportionate CVR.

WYHH Intervention Context

WYHH walking group intervention, described in detail below, was designed to test the effectiveness of a lay, community health promoter (CHP)–led walking group intervention in Detroit, Michigan, a city whose predominantly NHB and Hispanic residents experience substantial excess age-adjusted mortality due to heart disease (318.4 per 100,000), compared with national rates (192.9 per 100,000; Michigan Department of Community Health, 2012; Murphy, Xu, & Kochanek, 2012). A thriving and prosperous community with a strong middle class for much of the 20th century, Detroit, like many similar urban areas, experienced population out-migration and economic disinvestment beginning in the 1950s continuing to the present (Farley, Danziger, & Holzer, 2000; Southeast Michigan Council of Governments, 2006; Sugrue, 1996; U.S. Census Bureau, 2010). As NHW residents left the city, Detroit’s population shifted from 16% NHB in 1950 to 83% NHB and 7% Hispanic in 2010 (U.S. Census Bureau, 2010). Simultaneously, economic divestment resulted in exponential growth of poverty in the city (Farley, Danziger, & Holzer, 2000; Rossi, Freeman, & Lipsey, 1999; U.S. Census Bureau, 2010). In the face of these challenges, Detroit also has a number of unique strengths, including partnerships working together to understand and address cardiovascular health disparities, efforts of community groups and organizations to increase physical activity and improve activity-friendliness of neighborhoods; and longstanding collaborations among these entities.

In this article, we present results from tests of the effectiveness of the Healthy Environments Partnership (HEP) WYHH walking group intervention, collaboratively designed, implemented, and evaluated by one such Detroit-based partnership, the Detroit HEP (described below). Specifically, we test the following hypotheses: (a) adherence to WYHH will not differ by race, ethnicity, or SES; (b) participants in the WYHH intervention group will show a larger increase in steps per day (physical activity) from baseline to 8 weeks, compared with the lagged intervention group; (c) increases in physical activity during the intervention period (Weeks 1-8) will be associated with reductions in indicators of CVR; and (d) reductions in CVR attained during the intervention period will be sustained during the maintenance period (Weeks 9-32).

Method

WYHH was one component of the multilevel Community Approaches to Cardiovascular Health: Pathways to Heart Health (CATCH:PATH) intervention designed by HEP (see “Authors’ Note” and “Acknowledgments”), using a community-based participatory research partnership approach that engaged community residents, community-based organizations, health service providers, and academic researchers in a 2-year community planning process (see Schulz et al., 2011). CATCH:PATH was grounded in the community-identified priorities that emerged from this planning process, was informed by a social–ecological model that suggests that individual behaviors and risks occur within the context of organizational, community, and policy environments (Sallis et al., 2012). CATCH:PATH aimed to promote cardiovascular health through walking groups, and through organizational, community, and policy change to support activity-friendly communities. The design incorporated risk and protective factors for heart health across multiple levels of a social–ecological model, and attempted to address barriers and strengthen facilitating factors identified in the planning process (Schulz, 2011).

The WYHH component of this intervention was guided by a conceptual model, and empirical evidence, suggesting that participation in lay-health advisor-led walking groups can be effective in promoting physical activity (Kassavou et al., 2013). The intervention was informed by social support interventions (Heaney & Israel, 1997) and group dynamics theories (Johnson & Johnson, 2012). Empirical literature suggests that groups that offer peer support, and whose leaders promote group cohesion can be effective in increasing intervention adherence (Estabrooks & Carron, 1999), increased physical activity (Estabrooks, Fox, Doerksen, Bradshaw, & King, 2005; Lee et al., 2012; Wilson et al., 2010), and increased cardiorespiratory fitness (Perry, Rosenfeld, Bennett, & Potempa, 2007). Guided by this conceptual model and associated empirical evidence, WYHH was a group-level intervention to promote physical activity and heart health through organized groups with structured activities and an emphasis on social support and group cohesion. (See Figure 1 for the overall conceptual model. Dimensions tested in this article are bolded.) Walking groups were hosted by community- or faith-based organizations in Detroit neighborhoods, who received a rental fee for use of their space, as well as funds to support a part-time (10 hours per week) site community health promoter to assist with facilitation of the walking groups, and a small percent effort for on-site supervision. Groups met three times per week for 1.5 hours each time, and were facilitated by CHPs, project staff members who were community residents with strong social and facilitation skills and who matched the demographic and linguistic characteristics of the community. Approval for research on human subjects was granted by the University of Michigan Institutional Review Board on January 31, 2008. The ClinicalTrials.gov identifier is NCT02036593.

Figure 1.

Conceptual model of the Walk Your Heart to Health (WYHH) intervention.

Study Design

Detroit residents aged older than 18 years were eligible to participate following completion of a modified version of the Physical Activity Readiness Questionnaire (Cardinal, Esters, & Cardinal, 1996). Those currently pregnant, with systolic blood pressure >180 mmHg or diastolic blood pressure >110 mmHg, fasting blood glucose levels >180 mg/dL, type 1 or type 2 diabetes, bone or joint pain, chest pain, dizziness, or currently taking medication for high blood pressure or a heart condition were advised to consult their health care provider before enrolling. Following completion of the informed consent process, participants completed an interviewer-administered baseline Health Risk Assessment including a questionnaire, and anthropometric and clinical indicators of CVR (e.g., blood pressure, lipid panel). Interviewers were community residents who received training in conducting computer-assisted personal interviews, clinical and anthropometric measurements, and who worked under the supervision of the CATCH project evaluator. Intervention staff were present at these sessions to issue participants pedometers, and to explain their use. Prior to the start of the intervention, each participant wore a pedometer (with the step-counter blinded) for 1 week to provide baseline steps data.

A cluster randomized controlled design with a lagged intervention for the control group was used to evaluate the effectiveness of WYHH (see Figure 2). Participants were randomized into one of two groups: intervention and lagged intervention. Participants who indicated that they were enrolling with one or more friends or family members were randomized as clusters (e.g., three family members who joined together were randomized as a group to assure that they were in the same walking group). Participants randomized into the intervention arm began the WYHH intervention immediately following completion of baseline data collection. Those randomized into the lagged intervention arm returned for a second Health Risk Assessment at 8 weeks, along with those in the intervention group, as shown in Figure 2. Following this data collection point, the lagged intervention group began the WYHH intervention.

Figure 2.

Lagged intervention design for evaluation of Walk Your Heart to Health intervention.

Following baseline data collection (T1, Figure 2), the intervention group completed the 8-week WYHH intervention, followed by a second wave of data collection, involving interviewer-administered face-to-face interviews and anthropometric measurements for both the initial and lagged intervention groups (T2, Figure 2). The lagged intervention group then completed the 8-week WYHH intervention, followed by a third wave of data collection (T3). Both groups continued walking for another 24 weeks, with follow-up data collection (T4) conducted 32 weeks after the initiation of WYHH to assess maintenance of change over time. Thirty WYHH groups (15 paired intervention/lagged intervention groups) were held at 11 sites located in Detroit neighborhoods (May 2009 to May 2012). Two Detroit-based project coordinators provided supervision and day-to-day oversight for the intervention activities, following protocols developed by the project manager working in conjunction with the HEP SC and the project director.

Data and Measures

Health Risk Assessment s were completed at baseline, 8, and 32 weeks, using face-to-face computer-assisted personal interviews, in English or Spanish based on the preference of the participant. Interviews were conducted by nonintervention staff who were community members and who received training in interviewing techniques. Height (inches) was recorded at baseline. Weight (pounds), waist circumference (cm), and three measures of resting systolic and diastolic blood pressure (micrograms mercury) were recorded (1 minute between measurements, using an Omron HEM-712C digital blood pressure monitor) at each data collection point. At each data collection point, a fasting blood sample was collected using a finger stick to monitor total and high-density lipids (mg/dL), and glucose (mg/dL) using a Cholestec LDX machine. Participants wore an Omron JH 112 piesoelectric pedometer for 1 week prior to initiating the walking groups to record baseline steps (T1, Figure 2), and for the duration of the study period.

Measures

Dependent variables included indicators of physical activity and CVR. Physical activity included continuous measures of WYHH Steps (mean steps on days participants walked with WYHH groups), non-WYHH Steps (mean steps on days they walked without the groups), and Overall Steps (mean daily steps on all days), constructed using data from the Omron peisoelectric pedometer. Indicators of CVR included high blood pressure, measured as systolic blood pressure >140 mmHg, diastolic blood pressure >90 mmHg, or currently taking blood pressure medication; total cholesterol (mg/dL); high-density lipoproteins (HDLs; mg/dL); glucose (mg/dL); waist circumference (cm); and body mass index (BMI). We also assessed cumulative biologic risk (CBR), calculated as the sum of the following indicators: systolic blood pressure ≥ 140 mmHg; diastolic blood pressure ≥ 90 mmHg; waist circumference ≥ 102 cm (males) or ≥88 cm (females); glucose ≥ 110 mg/dL; total cholesterol >240 mg/dL or total cholesterol ≤ 240 mg/dL and low-density lipoprotein (LDL) ≥130 mg/dL; and HDL <40 for men or <50 mg/dL for women). Following Geronimus, Hicken, Keene, and Bound (2006), the index included points for individuals whose systolic and diastolic blood pressure levels were below the high blood pressure cut points and who were taking hypertension medication, those with glucose levels below the high risk cut point who were taking medication, and those whose lipid levels were within the normal range and who were taking medication for dyslipidemia.

Independent variables include two measures of adherence to the WYHH protocol, number of sessions attended overall; and consistency of participation, assessed as the number of weeks the participant attended at least one WYHH session. Adherence was entered into models using two sets of dummy variables: The first set reflected adherence during the 8-week intervention period (Adherence I), and the second reflected adherence during the 24-week maintenance period (Adherence M). Adherence I took the value of 0 at baseline and 32 weeks, and at 8 weeks reflected the number of sessions attended during the 8-week intervention period (up to three sessions per week, range 0-24), or the consistency of attendance in the intervention period (range 0-8). Adherence M took the value of 0 at baseline and 8 weeks, and at 32 weeks reflected the value of the number of sessions attended during the maintenance period (3 sessions per week, Weeks 9-32, range 0-72) or the number of weeks the participant had attended at least one session (consistency; 0-24). In models with CVR as the dependent variable, measures of physical activity were used as independent variables. For these models, each measure of physical activity (WYHH Steps, Overall Steps, and non-WYHH Steps, described above) was entered as a dummy variable. Physical Activity I represented the change in mean steps per day from baseline to the end of the 8-week intervention period: It took the value of 0 at baseline and 32 weeks. Physical Activity M represented the change in mean steps per day during the maintenance period, Weeks 9 to 32: It took the value of 0 at baseline and 8 weeks. Demographic controls included age (in years), gender, and self-reported race or ethnicity (NHB, Hispanic, NHW).

Data Analysis

Preliminary analyses included frequencies, means, and variances for key variables, followed by bivariate analysis of crude associations, using odds ratios for categorical and correlation coefficients for continuous variables. Multivariate models examined effects of WYHH participation on physical activity, and on key indicators of CVR. We used generalized estimating equations(GEEs) to deal with intraclass correlations due to observations obtained from the same participant over time, clustering of individuals within walking groups, and clustering due to randomization of groups of friends or family members together into the same walking group.

To assess the first research question, the extent to which the WYHH intervention was successful in recruiting members of the focal population (low-to-moderate income Hispanic and NHB residents), we examined the racial, ethnic, and educational characteristics of the WYHH participants, as well as their self-reported household income. Next, we calculated retention rates at 8 and 32 weeks overall, and by education, income, and race and ethnicity.

To test the hypothesis that participants in the WYHH intervention increased physical activity more than those in the lagged intervention (control) group (effectiveness), we ran GEE models comparing change in steps from baseline to 8 weeks for intervention with lagged intervention groups.

Next, we ran GEE models to assess more specifically the effectiveness of participation in the WYHH intervention in (a) increasing physical activity and (b) reducing CVR with increased physical activity. There were no significant differences between the intervention and lagged intervention groups in demographic characteristics, baseline steps, or changes in physical activity over time during the 8-week intervention period. Hence, data from the intervention and lagged intervention groups were pooled for the tests of these hypotheses. In addition, because there were no significant differences in steps or cardiovascular health indicators in the lagged intervention group between T1 and T2, T1 data were used as baseline data for both the intervention and the lagged intervention groups. This allows for chronologic concordance of baseline data across groups. Model 1 tested the hypothesis that adherence (number of sessions attended, consistency) to the WYHH intervention protocol was positively associated with steps:

\begin{matrix} Physical activity = β 0 + β 1 * Adherence I (Intervention) \\ + β 2 * Adherence M (Maintenance) \\ + β 3 * covariate 1 + β 4 * covariate 2 \dots + ε \end{matrix}

Next, to test the hypothesis that change in steps over time will reduce CVR, we ran regression models of the form shown in Model 2:

\begin{matrix} CVR = β 0 + β 1 * Physical Activity (Baseline) \\ + β 2 * Physical Activity I (Intervention) + \\ β 3 * Physical Activity M (Maintenance) + \\ β 4 * covariate 1 + β 5 * covariate 2 \dots + ε \end{matrix}

CVR is the longitudinal vector of the relevant indicator of CVR risk (e.g., high blood pressure).

Results

As shown in Table 1, 695 individuals enrolled in WYHH. At baseline, 61% identified as NHB, 3% as NHW, 36% as Hispanic. The majority (90%) were female, 42% had high school education or less, 71% reported annual household incomes of less than $35,000, and mean age was 46.7. Of the 695 who enrolled, 603 (87%) attended any WYHH sessions. There were no differences by race, ethnicity, household income, education, or waist circumference between those who enrolled and never attended, and those who enrolled and attended any session (results not shown). Those who enrolled and never attended were younger (40.1 vs. 47.5 years, p = .07), and were less likely to have high blood pressure (28.3% vs. 43.8%, p < .001) than those who enrolled and attended.

Table 1.

Characteristics of WYHH Study Participants at Baseline, 8, and 32 Weeks.

Characteristic	(A) Baseline (n = 695)	(B) 8 Weeks (n = 541)	(C) 32 Weeks (n = 388)
Age in years, mean (standard deviation)	46.6 (13.5)	48.6 (13.6)	49.6 (13.4)
Race/ethnicity, %
Hispanic	36.1	34.2*	33.3*
Non-Hispanic Black	62.0	62.1	62.8
Non-Hispanic White	1.9	0.7	0.8
Other	2.7	3.0	3.1
Baseline education, %
<12 years	21.5	21.7	21.5
12 years	20.7	20.9	20.9
>12 years	55.2	54.9	55.2
Other	2.6	2.4	2.4
Annual household income, $
≤9,999 or less	17.7	13.1	13.9
10,000-19,999	23.9	24.2	25.8
20,000-34,999	23.3	26.3	25.3
≥35,000	28.8	29.6	29.2
Refused	6.3	6.8	5.8

Indicates significantly different from baseline (p < .05).

Also shown in Table 1, 8-week retention was 78% (540/695) of all who enrolled, and 90% (540/603) among those exposed (ever attended WYHH sessions). Thirty-two week retention was 56% of all who enrolled (388/695), and 64% (388/603) of those exposed (ever attended any session). There were no differences in retention by race, ethnicity, education, or income. Those active at 8 and 32 weeks were older (p < .05; 48.6 and 49.6, respectively, compared with 46.7 at baseline).

Effectiveness in Promoting Physical Activity

Mean number of steps per day adjusting for age, gender, and race, did not differ between the intervention and lagged intervention groups at baseline (6524.4 and 6694.3, respectively, p = .222). At Week 8, intervention group participants recorded significantly more steps per day compared with the lagged intervention control group (β = 2004.5, p < .001).

Figure 3 shows unadjusted mean steps per person per day for each of the three measures of physical activity included in this analysis: non-WYHH Steps (mean steps per day on days not walking with the WYHH groups), WYHH Steps (mean steps per day on days walking with the WYHH groups), and Overall Steps (mean steps per day on all days of the week). Overall, in these unadjusted models, participants recorded significantly more steps compared with baseline at 8 (β = 2264, p < .001) and 32 weeks (β = 2110, p < .001), respectively. As shown, the unadjusted mean number of steps per day is significantly greater on days that participants walked with the WYHH groups, compared either with baseline steps or with the mean number of steps per day on days they did not walk with the WYHH group. These unadjusted results suggest that adherence or participation in the WYHH groups may be associated with increased physical activity.

Figure 3.

Unadjusted mean steps per day on days walked with Walk Your Heart to Health (WYHH Steps), days not walked with WYHH (non-WYHH), and overall (Overall).

To test this hypothesis more specifically, we ran models examining whether adherence (number of sessions attended, consistency of attendance) was positively associated with physical activity (steps), adjusting for demographic characteristics, using each of the three measures of physical activity described above. Results for Overall Steps are shown in Table 2.

Table 2.

Overall Mean Steps Regressed on Participation (Number of Sessions and Consistency) of Intervention Period (Weeks 1-8), and Maintenance Period (Weeks 9-32) Controlling for Age, Gender, and Race.

	Maintenance period				Intervention period
Parameter	Estimate	SE	95% CI	p Value	Estimate	SE	95% CI	p Value
Intercept	7545.6	555.8	[6456.3, 8634.9]	.000	7216.4	564.9	[6109.1, 8323.6]	.000
Age	−25.2	7.7	[−40.3, −10.1]	.001	−19.4	7.8	[−34.7, −4.0]	.013
Female	−1346.3	449.3	[−2226.9, −465.7]	.003	−1354.1	454.4	[−2244.7, −463.5]	.003
Non-Hispanic Black	−680.1	216.4	[−1104.3, −256]	.002	−693.3	217.3	[−1119.3, −267.4]	.001
Number of sessions intervention (Weeks 0-8)	178.9	7.6	[164, 193.7]	.000
Number of sessions maintenance (Weeks 9-32)^a	64.1	3.8	[56.6, 71.5]	.000
Consistency intervention (Weeks 0-8)					396.0	16.7	[363.3, 428.6]	.000
Consistency maintenance (Weeks 9-32)^a					132.1	8.2	[116, 148.2]	.000

Note. SE = standard error; CI = confidence interval.

Increases in steps during the maintenance period are interpreted as increases over the average number of steps per day at 8 weeks, rather than baseline. That is, continued increases in steps over and above the improvements attained at the 8-week mark.

Overall Steps increased significantly with both the number of sessions (β = 178.9, p < .001) and consistency (number of weeks attended at least one session; β = 396.0, p < .001). Each additional WYHH session (number) attended Weeks 1 to 8, was associated with an additional 178.9 Overall Steps per day. Each additional week of attendance (consistency) was associated with an additional 396.0 Overall Steps per day. Extrapolating these results, a participant who attended 18 WYHH sessions in Weeks 1 to 8 would be expected to increase Overall Steps by 3,364 steps/day. Patterns were similar for models using WYHH Steps and non-WYHH Steps (results not shown).

Similarly, both the number (β = 64.1, p < .001), and consistency (β = 132.1, p < .001) of attendance during the maintenance period, Weeks 9 to 32, were positively associated with Overall Steps, suggesting continued increases in physical activity associated with adherence during the maintenance period. Patterns were similar for models run using WYHH Steps and non-WYHH Steps (results not shown).

Effectiveness in Reducing CVR

Table 3 shows results from tests of the hypothesis that physical activity was inversely associated with anthropometric and clinical indicators of CVR over the intervention and maintenance periods. These models disentangle the associations with physical activity of the mean number of steps per day recorded at baseline (baseline steps), the change in the mean number of steps per day during the 8-week intervention period (Physical Activity Intervention), and the change in steps per day during the 24-week maintenance period (Physical Activity Maintenance). Each increase of 1,000 Overall Steps at 8 weeks compared with baseline (Physical Activity Intervention) was associated with lower odds of high blood pressure (0.9, p = .01), and with reductions in total cholesterol (β = −0.10, p = .02), HDL (β = −0.3, p = .00), blood glucose (β = −0.4, p =.04) and waist circumference (β = −0.2 cm, p=0.01). BMI (β = −0.01, p = .04) and CBR (β = −0.1, p = .00) were also reduced significantly (results not shown) at eight weeks. Similar patterns were found for WYHH Steps and non-WYHH steps (results not shown).

Table 3.

Indicators of Cardiovascular Risk (Systolic BP, Diastolic BP, Total Cholesterol, HDL, Fasting Blood Glucose, Waist Circumference) Regressed on Change in Overall Mean Steps (per 1,000 Steps) During the Intervention Period, and Overall Mean Steps (per 1,000 Steps) During the Maintenance Period, Controlling for Age, Gender, and Racial or Ethnic Status).

	High BP			Total cholesterol			HDLs			Blood glucose levels			Waist circumference			CBR
	OR	95% CI	p Value	Estimate	95% CI	p Value	Estimate	95% CI	p Value	Estimate	95% CI	p Value	Estimate	95% CI	p Value	Estimate	95% CI	p Value
Intercept				152.1	[136.7, 167.4]	.00	17.7	[12, 23.5]	.00	86.9	[79.5, 94.3]	.00	115.6	[107.8, 123.4]	.00	0.9	[0.2, 1.6]	.02
Age	1.1	[1.0, 1.1]	.00	0.4	[0.1, 0.6]	.00	0.3	[0.2, 0.4]	.00	0.4	[0.2, 0.5]	.00	−0.2	[−0.3, 0.0]	.01	0.0	[0, 0.1]	.00
Female	0.8	[0.4, 1.5]	.52	6.0	[−4.9, 16.8]	.28	10.4	[6.9, 13.9]	.00	−0.3	[−4.5, 4.0]	.89	−6.3	[−11.2, −1.5]	.01	0.2	[−0.3, 0.7]	.44
Non-Hispanic Black	2.7	[1.8, 4.1]	.00	−0.5	[−6.5, 5.5]	.87	4.7	[2.2, 7.2]	.00	−0.9	[−4.5, 2.6]	.62	4.8	[1.9, 7.7]	.00	0.1	[−0.2, 0.4]	.41
Physical activity baseline^a	0.9	[0.8, 1.0]	.01	0.4	[−0.6, 1.5]	.38	0.6	[0.2, 1.0]	.01	−0.7	[−1.3, −0.2]	.01	−0.9	[−1.4, −0.4]	.00	−0.1	[−0.1, 0]	.00
Physical activity intervention	0.9	[0.9, 1.0]	0.00	−0.7	[−1.2, −0.1]	.02	−0.3	[−0.5, −0.1]	.00	−0.4	[−0.8, 0.0]	.04	−0.2	[−0.3, 0.0]	.01	−0.1	[−0.1, 0]	.00
Physical activity maintenance^c	0.9	[0.8, 1.1]	0.32	−2.2	[−4, −0.3]	.02	−0.7	[−1.5, 0.0]	.06	0.1	[−1.0, 1.2]	.87	−0.2	[−0.8, 0.5]	.62	−0.1	[−0.2, 0]	.21

Note. OR = odds ratio; CI = confidence interval; CBR = cumulative biologic risk; HDL = high-density lipoprotein; BP = blood pressure.

Mean steps per day (per 1,000 steps) at baseline. ^bChange in mean steps per day (per 1,000 steps) from baseline to 8 weeks (intervention period). ^cChange in mean steps per day (per 1,000 steps) from 9 to 32 weeks (maintenance period).

Coefficients for Physical Activity Maintenance indicate that odds of high blood pressure (0.92, p = .32), and associations with total cholesterol (β = −2.2, p = .02), HDLs (β = −0.7, p = .06), blood glucose (β = 0.1, p = .87), and waist circumference (β = −0.2 cm, p = .62) continue to be inversely associated with physical activity. Trends were similar for BMI (β = −0.0, p = .98), and CBR (β = −0.1, p = .21; not shown). With the exception of total cholesterol and CBR, additional reductions during the maintenance period were not statistically significant, beyond reductions achieved during the intervention period, using Overall Steps as the independent variable. Similar patterns were evident for models using non-WYHH Steps and for WYHH Steps, which showed continued significant reductions in waist circumference (β = −0.00, p = .032), and BMI (β = −0.00, p = .031) for each additional 1,000 steps per day during the maintenance period using the latter measure of physical activity (results not shown).

Discussion

There are four main findings from this study. First, WYHH walking groups, conducted at community-based locations and facilitated by CHPs, were effective in engaging residents of low-to-moderate income, predominantly NHB and Hispanic communities at high risk for CVD, in physical activity. There were no differences in retention by race, ethnicity, or SES. Second, results suggest that participation in the WYHH intervention was associated with increased physical activity, compared with participants in the lagged intervention control group. Third, increases in walking (steps), a moderate form of physical activity, were associated with improvements in multiple indicators of CVR during the 8-week intervention period. Finally, reductions in multiple CVD risk indicators attained during the 8-week intervention period were largely maintained during the 24-week maintenance period, with some evidence of continued reduction in risk for specific outcomes. These results are generally consistent with a growing body of evidence suggesting that walking group interventions are efficacious in promoting physical activity (Kassavou et al., 2013), and a more limited body of evidence suggesting that they may be an effective strategy for reducing CVR among racial, ethnic, and socioeconomic groups who experience excess risk (Banks-Wallace, 2007; Banks-Wallace & Conn, 2005; Hogue, 2007; Lee et al., 2012; Rogers, 1997). We discuss each of these findings below, followed by limitations, and implications for interventions to reduce racial, ethnic, and socioeconomic health disparities.

Engaging Residents of Low-to-Moderate Income Urban Communities in Physical Activity

WYHH was successful in recruiting NHB and Hispanic participants, across a range of educational levels, with low-to-moderate household incomes. Our finding that 86% of those who completed baseline assessments attended at least one walking group session is comparable with 84% reported by Coull, Taylor, Elton, Murdoch, and Hargreaves (2004) for a sample of individuals aged 60 years and older with ischaemic heart disease. Retention rates at 8 weeks (86% of all enrolled and 91% of those who attended one or more session) exceed the Prevention Research Synthesis criteria for 70% retention (Lyles, Crepaz, Herbst, & Kay, 2006). At 32 weeks, retention (56% of all enrolled and 65% of those who attended at least one session) falls below this bar. Loss to follow-up is problematic in most longitudinal studies and may lead to bias, depending on whether data are missing completely at random (MCAR), missing at random (MAR), or missing not at random (MNAR). In a simulation study, Kristman, Manno, and Cote (2004) used 1,000 computer replications of a cohort of 500 observations with differing levels of loss to follow-up generated through three mechanisms: MCAR, MAR, and MNAR. They report no important bias with levels of loss that varied from 5% to 60% when loss to follow-up was related to MCAR or MAR mechanisms. However, when observations were lost to follow-up based on an MNAR mechanism, the authors found seriously biased estimates. Participant loss rarely, if ever, occurs completely at random, rather there is usually a systematic reason why participants are lost from a longitudinal study. When observations are MAR, it is assumed that the probability of a participant remaining in the study depends on the exposure or confounders but not on the outcome (Kristman et al., 2004). In this scenario, loss to follow-up should not differ by any of our dependent variables (e.g., blood pressure, total cholesterol, HDL, blood glucose). In this particular analysis, we found no association of loss to follow-up and relevant health outcomes, nor did we find differences by several key independent variables, including race, ethnicity, income, and education. Our finding of differences in retention by age, with older participants less likely to be lost to follow-up is consistent with an MAR scenario, and may be of less concern than MNAR because previously collected variables can help explain the potential bias by controlling for the covariates that are associated with loss to follow-up in multivariate analysis (Kristman et al., 2004).

Thus, while the level of attrition in this study is higher than has been reported in some studies with different sample and intervention characteristics (76%; Estabrooks, Bradshaw, Dzewaltowski, & Smith-Ray, 2008), and we are unable to rule out the possibility that data are MNAR on some other variable not examined here, our assessment of patterns of loss to follow-up are suggestive of MAR. Furthermore, our rate of loss to follow-up is comparable with 6-month retention rates recently reported for community-based walking group interventions with comparable samples (61%, Warren, Maley, Sugarwala, Wells, & Devine, 2010; 68%, Jancey et al., 2008). Variations in intervention design, sampling strategies (e.g., community sample vs. disease-specific sample), age of participants, and in reporting of participant retention contribute to challenges in comparing these results. Despite these limitations, our findings suggest that walking groups based in community- or faith-based organizations in urban communities, that offer social support and structured opportunities for walking, and with leadership from CHPs, can be successful in recruiting participants from racial, ethnic, and socioeconomic groups that experience disproportionate risk of CVD, with needed attention to retention. Our finding that older participants are more likely to remain engaged in the walking groups, and therefore more likely to realize the health benefits described above, is consistent with results from a recent meta-analysis reporting that walking groups with older participants were more likely to realize greater health improvements (Kassavou et al., 2013). While the evidence is limited by the small number of studies, some with relatively small sample sizes, and variations in reporting that contribute to challenges in comparisons across studies, retention may be a particular concern in communities whose demographics are consistent with excess CVR.

Participation in Walking Groups and Physical Activity Levels

A comparison of participants in the WYHH intervention walking groups with the lagged intervention groups shows that those in the intervention group increased average steps per day about 2,000 steps more (p < .0001) than those in the lagged group during the same 8-week period. Mean increases in overall steps per day at 8 weeks and 32 weeks (2,264 and 2,110, respectively) are comparable to results reported elsewhere in the literature, including studies by Banks-Wallace (2007) who reports increases of 203 and 1,425 steps per day at 6 and 12 months, respectively; Zoellner et al. (2010) who report an increase in 2,567 steps per day at 6 months; and Duru, Sarkisian, Leng, and Mangione (2010) who report an average increase in weekly steps of 9,883 (1,411 steps per day) at 6 months.

Adherence, whether assessed as number of sessions attended or consistency (number of weeks attended at least one session), was positively associated with increases in physical activity. Participants walked significantly more steps on the days that they walked with the walking groups compared with either baseline or with the days that they did not walk with WYHH groups. Attending a greater number of sessions, or a greater number of weeks, was associated with greater increases in physical activity during the intervention and the maintenance period.

Reductions in Anthropometric and Clinical Indicators of CVR (Intervention)

Results reported here suggest that walking was associated with significant reductions in multiple indicators of CVR. Reductions in the odds of high blood pressure, total cholesterol, blood glucose levels, waist circumference, BMI, and CBR were realized in just 8 weeks of moderate physical activity. These results are consistent with findings reported elsewhere in the literature, with participation in walking group interventions associated with reduced risk of multiple indicators of CVR (Banks-Wallace, 2007).

The one exception to this pattern was reflected in HDL, where reductions are in a direction generally associated with increased CVR. Meta-analysis of randomized controlled trials conducted by Kodama et al. (2007) found wide variation in the effect of aerobic physical activity on serum HDL levels. Specifically, minimum exercise length and energy expenditure were needed to produce increases in HDL. Participants with high total cholesterol and low BMI experienced greater exercise effects on HDL (Kodama et al., 2007). Dietary factors may also have played a role in participant serum HDL if participants replaced saturated fats with polyunsaturated fats or carbohydrates during the intervention period (Barnard, 1991).

Coefficients reported in Table 3 indicate improvements in multiple-CVR indicators, and are reported for each 1,000 step increase in mean steps taken per day (Overall Steps). As reported above, WYHH participants increased mean steps per day by 2004.5 steps during the 8-week intervention period. As a result, these results are a conservative estimate of reductions in risk achieved by participants in the WYHH intervention.

Maintenance of Reductions in CVR

During the maintenance period (Weeks 9-32), participants experienced continued significant reductions in total cholesterol associated with Overall Steps, in waist circumference and BMI associated with WYHH Steps, and in CBR regardless of the measure of physical activity used. Although other indicators generally trended toward reductions in CVR, these reductions were not significantly beyond the reductions in risk realized during the initial 8-week intervention period.

Limitations

There are a number of limitations associated with this study. Notably, while retention at 8 weeks was strong, at 32 weeks study retention fell below 70%, thus raising concerns about potential retention bias. While we were able to demonstrate that there were no differences in retention by race, ethnicity, income, or household education, we are not able to rule out the possibility that those who did not continue did not differ from those who continued in some way that may have affected our findings. The demographic composition of the study communities, and thus the demographic composition of participants, allowed us to test for differences in retention between NHB and Hispanic participants, but not for comparisons with NHWs. Finally, the inclusion of specific measures of dietary intake may have allowed further exploration of the declines in HDL observed over the course of the intervention, and to better understand what, if any, role changes in dietary intake may have contributed to changes in anthropometric and clinical indicators of CVR observed.

Implications for Practice

Despite the limitations described above, the findings reported here contribute to a growing body of evidence suggesting that interventions to promote walking in groups are effective in increasing physical activity (Kassavou et al., 2013). Perhaps most important, these findings contribute to, and substantially strengthen, a relatively small body of evidence suggesting that such interventions may be effective for NHB and Hispanic residents of low-to-moderate income neighborhoods, as well as NHWs (Banks-Wallace, 2007; Duru, Sarkisian, Leng, & Mangione, 2010; Lee et al., 2012; Zoellner et al., 2010). They expand and strengthen previously reported results in several important ways. Specifically, the large sample size and randomized, controlled study design provide more robust evidence than has been available through several previous studies. The use of uploadable pedometers throughout the intervention and maintenance periods, and clinical and anthropometric measurements taken at baseline, 8, and 32 weeks is an advance over several previous studies that relied on self-report data. The inclusion of both NHB and Hispanic participants in this study, and availability of measures of education and income, allowed testing for differences in adherence on these demographic indicators, with none found.

Thus, the results reported here contribute to an increasingly robust body of evidence suggesting that members of racial, ethnic, and socioeconomic groups that experience excess CVR, and who reside in neighborhoods where built and social environments may pose particular challenges for physical activity, have an interest in participation in walking groups. Furthermore, adherence at 8 weeks was excellent, and although lower than optimal, roughly 60% of those enrolled continued to participate after 32 weeks. Participants realized increases in physical activity and reductions in CVR, with adherence highest and, thus, CVR reductions most likely among older individuals.

Walking group interventions designed in collaboration with community members to address aspects of local social and physical environments described as challenging for regular physical activity (e.g., mapping safe routes for groups to walk in their neighborhoods, identifying community locations where community members could walk indoors when weather conditions were poor). Because residents of such neighborhoods experience disproportionate CVR, designing interventions that take into account the contexts in which people attempt to be physically active are particularly important. The findings reported here provide evidence that such interventions can be effective in promoting physical activity and reducing CVR among groups that experience excess risk.

Dissemination of interventions such as WYHH that have demonstrated effectiveness in increasing physical activity and reducing CVR may contribute to reductions in racial, ethnic, and socioeconomic health disparities. However, such reductions are unlikely, in and of themselves, to eliminate disparities in CVR. Rather, they are most likely to contribute to sustained reductions in CVR disparities if enacted as one component of multilevel efforts that encompass changes in social and economic policies as well as community efforts that simultaneously address socioeconomic, built, and social environmental conditions that contribute to excess CVR (Frohlich & Potvin, 2008; Garcia, Bracho, Cantero, & Glenn, 2009; Sallis et al., 2006).

In sum, NHBs and Hispanics experience excess CVR and are more likely to reside in communities with limited walkability compared with NHWs in the United States. Relatively few studies have examined the acceptability and effectiveness of walking groups as interventions that may help address challenges experienced by residents of low-to-moderate income urban communities in maintaining active lifestyles. The WYHH CHP-led walking group intervention was designed and implemented using a community-based participatory process with particular attention to addressing challenges and building on assets within low-to-moderate income neighborhoods in Detroit. Results reported here suggest that collaboratively designed interventions that engage community and academic partners, build on community strengths and assets (e.g., community-based organizations as host sites, community residents as CHPs), and that address resident-identified community challenges (e.g., safety) can effectively promote physical activity and cardiovascular health among those who experience excess risk. Dissemination of such evidence-based interventions can be one component of multilevel approaches to promote physical activity and reduce excess CVR.

Footnotes

Acknowledgements

We thank the members of the HEP Steering Committee for their contributions to the work presented here, including representatives from Brightmoor Community Center, Detroit Department of Health and Wellness Promotion and Institute for Population Health, Detroit Hispanic Development Corporation, Friends of Parkside, Henry Ford Health System, Warren Conner Development Coalition, and University of Michigan School of Public Health. We thank our funders for support for this article and also thank Sue Andersen and Cassandra Parks for critical support in its preparation.

Authors’ Note

The Healthy Environments Partnership (www.hepdetroit.org) is a community-based participatory research partnership affiliated with the Detroit Community–Academic Urban Research Center (). The results presented here are solely the responsibility of the authors and do not necessarily represent the views of the National Institute on Minority Health and Health Disparities, its directors, officers, or staff.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study and analysis were supported by the National Institute for Minority Health and Health Disparities (R24 MD001619).

References

Ainsworth

B. E.

Haskell

W. L.

Herrmann

S. D.

Meckes

Bassett

D. R.

Jr. Tudor-Locke

. . . Leon

A. S.

(2011). 2011 Compendium of physical activities: A second update of codes and MET values. Medicine & Science in Sports & Exercise, 43, 1575-1581.

Banks-Wallace

(2007). Outcomes from Walk the Talk: A nursing intervention for Black women. ABNF Journal, 18, 19-24.

Banks-Wallace

Conn

(2005). Changes in steps per day over the course of a pilot walking intervention. ABNF Journal, 16, 28-32.

Barnard

R. J.

(1991). Effects of life-style modification on serum lipids. Archives of Internal Medicine, 151, 1389-1394.

Bjaras

Harberg

L. K.

Sydhoff

Ostenson

C. G.

(2001). Walking campaign: A model for developing participation in physical activity? Experiences from three campaign periods of the Stockholm Diabetes Prevention Program (SDPP). Patient Education and Counseling, 42, 9-14.

Cardinal

B. J.

Esters

Cardinal

M. K.

(1996). Evaluation of the revised physical activity readiness questionnaire in older adults. Medicine & Science in Sports & Exercise, 28, 468-472.

Coull

A. J.

Taylor

V. H.

Elton

Murdoch

P. S.

Hargreaves

A. D.

(2004). A randomised controlled trial of senior Lay Health Mentoring in older people with ischaemic heart disease: The Braveheart Project. Age and Ageing, 33, 348-354.

Duru

O. K.

Sarkisian

C. A.

Leng

Mangione

C. M.

(2010). Sisters in motion: A randomized controlled trial of a faith-based physical activity intervention. Journal of the American Geriatrics Society, 58, 1863-1869.

Echeverria

Diez-Roux

A. V.

Shea

Borrell

L. N.

Jackson

(2008). Associations of neighborhood problems and neighborhood social cohesion with mental health and health behaviors: The multi-ethnic study of atherosclerosis. Health & Place, 14, 853-865.

10.

Estabrooks

P. A.

Bradshaw

Dzewaltowski

D. A.

Smith-Ray

R. L.

(2008). Determining the impact of Walk Kansas: Applying a team-building approach to community physical activity promotion. Annals of Behavioral Medicine, 36, 1-12.

11.

Estabrooks

P. A.

Carron

A. V.

(1999). Group cohesion in older adult exercisers: Prediction and intervention effects. Journal of Behavioral Medicine, 22, 575-588.

12.

Estabrooks

P. A.

Fox

E. H.

Doerksen

S. E.

Bradshaw

M. H.

King

A. C.

(2005). Participatory research to promote physical activity at congregate-meal sites. Journal of Aging and Physical Activity, 13, 121-144.

13.

Farley

Danziger

Holzer

H. J.

(2000). Detroit divided. New York, NY: Russell Sage Foundation.

14.

Fisher

K. J.

(2004). A community-based walking trial to improve neighborhood quality of life in older adults: A multilevel analysis. Annals of Behavioral Medicine, 28, 186-194.

15.

Frohlich

K. L.

Potvin

(2008). Transcending the known in public health practice: The inequality paradox: The population approach and vulnerable populations. American Journal of Public Health, 98, 216-221.

16.

Garcia

Bracho

Cantero

Glenn

B. A.

(2009). “Pushing” physical activity, and justice. Preventive Medicine, 49, 330-333.

17.

Geronimus

A. T.

Hicken

Keene

Bound

(2006). “Weathering” and age patterns of allostatic load scores among blacks and whites in the United States. American Journal of Public Health, 96, 826-833.

18.

Heaney

C. A.

Israel

B. A.

(1997). Social networks and social support in health education. In Glanz

Lewis

F. M.

Rimer

B. K.

(Eds.), Health behavior and health education (pp. 179-205). San Francisco, CA: Jossey-Bass.

19.

Hogue

(2007). The effects of buddy support on physical activity in African American women (Unpubished doctoral dissertation). University of Toledo, Toledo, OH.

20.

Jancey

J. M.

Lee

A. H.

Howat

P. A.

Clarke

Wang

Shilton

(2008). The effectiveness of a physical activity intervention for seniors. American Journal of Health Promotion, 22, 318-321.

21.

Johnson

D. W.

Johnson

F. P.

(2012). Joining together: Group theory and group skills. Upper Saddle River, NJ: Pearson.

22.

Kahn

E. B.

Ramsey

L. T.

Brownson

R. C.

Heath

G. W.

Howze

Powell

K. E.

. . . Corso

(2002). The effectiveness of interventions to increase physical activity: A systematic review. American Journal of Preventive Medicine, 22(4 Suppl.), 73-107.

23.

Kassavou

Turner

French

D. P.

(2013). Do interventions to promote walking in groups increase physical activity? A meta-analysis. International Journal of Behavioral Nutrition and Physical Activity, 10, 18.

24.

Kodama

Tanaka

Saito

Shu

Sone

Onitake

. . . Sone

(2007). Effect of aerobic exercise training on serum levels of high-density lipoprotein cholesterol: A meta-analysis. Archives of Internal Medicine, 167, 999-1008.

25.

Kristman

Manno

Cote

(2004). Loss to follow-up in cohort studies: How much is too much? European Journal of Epidemiology, 19, 751-760.

26.

Lee

R. E.

O’Connor

D. P.

Smith-Ray

Mama

S. K.

Medina

A. V.

Reese-Smith

J. Y.

. . . Estabrooks

P. A.

(2012). Mediating effects of group cohesion on physical activity and diet in women of color: Health is power. American Journal of Health Promotion, 26, e116-e125.

27.

Lyles

C. M.

Crepaz

Herbst

J. H.

Kay

L. S.

(2006). Evidence-based HIV behavioral prevention from the perspective of the CDC’s HIV/AIDS Prevention Research Synthesis Team. AIDS Education and Prevention, 18(4 Suppl. A), 21-31.

28.

Manson

J. E.

Greenland

LaCroix

A. Z.

Stefanick

M. L.

Mouton

C. P.

Oberman

. . . Siscovick

D. S.

(2002). Walking compared with vigorous exercise for the prevention of cardiovascular events in women. New England Journal of Medicine, 347, 716-725.

29.

Marshall

S. J.

Jones

D. A.

Ainsworth

B. E.

Reis

J. P.

Levy

S. S.

Macera

C. A.

(2007). Race/ethnicity, social class, and leisure-time physical inactivity. Medicine & Science in Sports & Exercise, 39, 44-51.

30.

Michigan Department of Community Health. (2012). Mortality statistics. Retrieved from http://www.mdch.state.mi.us/pha/osr/CHI/Deaths/frame.asp

31.

Moore

L. V.

Diez Roux

A. V.

Nettleton

J. A.

Jacobs

D. R.

Jr. (2008). Associations of the local food environment with diet quality—A comparison of assessments based on surveys and geographic information systems: The multi-ethnic study of atherosclerosis. American Journal of Epidemiology, 167, 917-924.

32.

Mujahid

M. S.

Diez Roux

A. V.

Morenoff

J. D.

Raghunathan

T. E.

Cooper

R. S.

Shea

(2008). Neighborhood characteristics and hypertension. Epidemiology, 19, 590-598.

33.

Murphy

M. H.

Murtagh

E. M.

Boreham

C. A.

Hare

L. G.

Nevill

A. M.

(2006). The effect of a worksite based walking programme on cardiovascular risk in previously sedentary civil servants (NCT00284479). BMC Public Health, 6, 136.

34.

Murphy

M. H.

Nevill

A. M.

Murtagh

E. M.

Holder

R. L.

(2007). The effect of walking on fitness, fatness and resting blood pressure: A meta-analysis of randomised, controlled trials. Preventive Medicine, 44, 377-385.

35.

Murphy

S. L.

J. Q.

Kochanek

K. D.

(2012, January 11). Deaths: Preliminary data for 2010. National Vital Statistics Report, 60(4), 9-16.

36.

Ogilvie

Craig

Griffin

Macintyre

Wareham

N. J.

(2009). A translational framework for public health research. BMC Public Health, 9, 116.

37.

Perry

C. K.

Rosenfeld

A. G.

Bennett

J. A.

Potempa

(2007). Heart-to-Heart: Promoting walking in rural women through motivational interviewing and group support. Journal of Cardiovascular Nursing, 22, 304-312.

38.

Rogers

(1997). Effectiveness of a walking club and a self-directed physical activity program in increasing moderate intensity physical activity among African-American females (Unpublished doctoral dissertation). University of Oregon, Eugene.

39.

Rossi

P. H.

Freeman

H. E.

Lipsey

M. W.

(1999). Evaluation: A systematic approach. Thousand Oaks, CA: Sage.

40.

Sallis

J. F.

Cervero

R. B.

Ascher

Henderson

K. A.

Kraft

M. K.

Kerr

(2006). An ecological approach to creating active living communities. Annual Reviews of Public Health, 27, 297-322.

41.

Sallis

J. F.

Floyd

M. F.

Rodriguez

D. A.

Saelens

B. E.

(2012). Role of built environments in physical activity, obesity, and cardiovascular disease. Circulation, 125, 729-737.

42.

Schiller

J. S.

Lucas

J. W.

Ward

B. W.

Peregoy

J. A.

(2012, January). Summary health statistics for U.S. adults: National Health Interview Survey, 2010. Vital Health Statistics, 10(252), 1-207.

43.

Schulz

A. J.

(2011). Urban environments and health. In Nriagu

(Ed.), Encyclopedia of environmental health (pp. 549-555). Burlington, MA: Elsevier.

44.

Schulz

A. J.

Israel

B. A.

Coombe

Gaines

Reyes

Rowe

. . . Weir

(2011). A community-based participatory planning process and multilevel intervention design: Toward eliminating cardiovascular health inequities. Health Promotion Practice, 12, 900-911.

45.

Southeast Michigan Council of Governments. (2006, September). Population and housing in Southeast Michigan 2000-2006. Retrieved from www.semcog.org

46.

Sugrue

T. J.

(1996). The origins of the urban crisis: Race and inequality in Postwar Detroit. Princeton, NJ: Princeton University Press.

47.

Swartz

A. M.

Strath

S. J.

Bassett

D. R.

Moore

J. B.

Redwine

B. A.

Groer

Thompson

D. L.

(2003). Increasing daily walking improves glucose tolerance in overweight women. Preventive Medicine, 37, 356-362.

48.

Tudor-Locke

Johnson

W. D.

Katzmarzyk

P. T.

(2010). Frequently reported activities by intensity for U.S. adults: The American Time Use Survey. American Journal of Preventive Medicine, 39, e13-e20.

49.

Tully

M. A.

Cupples

M. E.

Chan

W. S.

McGlade

Young

I. S.

(2005). Brisk walking, fitness, and cardiovascular risk: A randomized controlled trial in primary care. Preventive Medicine, 41, 622-628.

50.

U.S. Census Bureau. (2010). 2010 Census, Detroit City quickfacts. Retrieved from http://quickfacts.census.gov/qfd/states/26/2622000.html

51.

U.S. Department of Health and Human Services. (2008a). Physical guidelines for Americans: Be active, healthy, and happy. Washington, DC: Author.

52.

U.S. Department of Health and Human Services. (2008b). Prevalence of self-reported physically active adults—United States, 2007. MMWRMorbidity Mortality Weekly Report, 57, 1297-1300.

53.

Warren

B. S.

Maley

Sugarwala

L. J.

Wells

M. T.

Devine

C. M.

(2010). Small steps are easier together: A goal-based ecological intervention to increase walking by women in rural worksites. Preventive Medicine, 50, 230-234.

54.

Williams

P. T.

Thompson

P. D.

(2013). Walking versus running for hypertension, cholesterol, and diabetes mellitus risk reduction. Artherosclerosis, Thrombosis, and Vascular Biology, 33, 1085-1091.

55.

Wilson

M. G.

Basta

T. B.

Bynum

B. H.

DeJoy

D. M.

Vandenberg

R. J.

Dishman

R. K.

(2010). Do intervention fidelity and dose influence outcomes? Results from the move to improve worksite physical activity program. Health Education Research, 25, 294-305.

56.

Wong

M. D.

Shapiro

M. F.

Boscardin

W. J.

Ettner

S. L.

(2002). Contribution of major diseases to disparities in mortality. New England Journal of Medicine, 347, 1585-1592.

57.

Zoellner

Connell

Powers

Avis-Williams

Yadrick

Bogle

M. L.

(2010). Does a six-month pedometer intervention improve physical activity and health among vulnerable African Americans? A feasibility study. Journal of Physical Activity & Health, 7, 224-231.