Abstract
Objective: Local health departments (LHDs) have essential roles to play in ensuring the promotion of physical activity (PA) in their communities in order to reduce obesity. Little research exists, however, regarding the existence of these PA interventions across communities and how these interventions may impact community health.
Design: In this exploratory study, we used cluster analysis to identify the structure of co-occurring PA interventions, followed by regression analysis to quantify the association between the patterns of PA interventions and prevalence of PA and obesity at a population level.
Setting: Our study setting included local health jurisdictions in Colorado, Florida, Minnesota, New Jersey, Tennessee, and Washington.
Participants: Participating jurisdictions were those 218 local health jurisdictions (mostly counties) from which LHD leaders had provided data in 2013 for the Multi-Network Practice and Outcome Variation Examination Study.
Main Outcome Measures: We obtained unique public health activities data on PA interventions conducted in 2012 from 218 LHDs in 6 participating states. We categorized jurisdictions using cluster analysis, based on PA intervention approaches indicated by LHD leaders as available in their communities and then examined associations between categories and prevalence of obesity and of residents engaged in PA.
Results: We identified 5 distinct PA intervention categories representing community-wide approaches-Comprehensive Approach, Built Environment, Personal Health, School-Based Interventions, and No Apparent Activities. Prevalence rates of obesity and PA among jurisdictions in the intervention clusters were significantly different from jurisdictions with No Apparent Activities, with more population-level approaches most significantly related to beneficial outcomes.
Conclusion: Our findings suggest the importance of standardized public health services data for generating evidence regarding health-related outcomes. The intervention categories we identified appear to reflect broad, local community-wide prevention approaches and demonstrated that population-level PA interventions can be testable and may have particularly beneficial relationships to community health. Widespread adoption of such standardized data depicting local public health prevention activity could support monitoring practice change, performance improvement, comparisons across communities that could reduce unnecessary variation, and the generation of evidence for public health practice and policy-making.