Statistical Design Features of the Healthy Communities Study

Abstract

The Healthy Communities Study is designed to assess relationships between characteristics of community programs and policies targeting childhood obesity and children’s BMI, diet, and physical activity. The study involved a complex data collection protocol implemented over a 2-year period (2013–2015) across a diverse sample of up to 125 communities, defined as public high school catchment areas. The protocol involved baseline assessment within each community that included in-person or telephone interviews regarding community programs and policies and in-home collection of BMI, nutritional, and physical activity outcomes from a sample of up to 81 children enrolled in kindergarten through eighth grade in public schools. The protocol also involved medical record reviews to establish a longitudinal trajectory of BMI for an estimated 70% of participating children. Staged sampling was used to collect less detailed measures of physical activity and nutrition across the entire sample of children, with a subset assessed using more costly, burdensome, and detailed measures. Data from the Healthy Community Study will be analyzed using both cross-sectional and longitudinal models that account for the complex design and correct for measurement error and bias using a likelihood-based Markov chain Monte Carlo methodology. This methods paper provides insights into the complex design features of the Healthy Communities Study and may serve as an example for future large-scale studies that assess the relationship between community-based programs and policies and health outcomes of community residents.

Community Policies and Programs to Prevent Obesity and Child Adiposity

“…14 This design maximized identification of CPPs that might be associated with child obesity, while providing information about the distribution of CPPs across communities. Strata were defined from combinations of census tracts based on race, ethnicity, income, region, and a preselection ranking of perceived CPP intensity.…”

“…Strata were defined from combinations of census tracts based on race, ethnicity, income, region, and a preselection ranking of perceived CPP intensity. 14 Census tracts were selected randomly from each stratum with probability proportional to the population of children aged 4–15 years; children were in grades Kindergarten–eighth where they could be exposed to CPPs delivered through schools and other means. The community was defined by the area served by the public high school closest to the center of each selected census tract; Kindergarten–eighth grade schools within that high school catchment area were used for household recruitment.…”

“…14 A minimally adjusted model had as covariates child height, gender, age, and age squared; a fully adjusted model also included the other covariates described above. There were some missing data due to non-response; multiple imputation was used to create 20 imputed data sets that were then combined for inference.…”

The Healthy Communities Study

“…Thus, they were randomly selected with weights proportional to size. A more detailed description of the sampling approach for communities, schools, and households can be found in Strauss et al 35 in this issue. One benefit of the hybrid approach is that it ensured the inclusion of communities that had implemented promising programs and policies targeting childhood obesity, rather than leaving inclusion of those types of communities to chance.…”

“…The purpose of this paper is to describe the HCS's rationale, aims, and logic model, and to provide a brief overview of the study design. The other papers in this supplement will present more detailed information on the statistical design (Strauss et al), 35 operational elements (John and colleagues), 36 weight status measures (Sroka et al), 37 dietary measures (Ritchie and colleagues), 38 physical activity measures (Pate et al), 39 and measures of community programs and policies (Fawcett and colleagues). 40 …”