The 130th Annual Meeting of APHA

Abstract #41512

Melinda S. Forthofer, PhD¹, Scott Alan Johnson, JD, MBA, MSHS², and Gary Walby, MS, MSPH². (1) College of Public Health, University of South Florida, 13201 Bruce B. Downs Blvd, MDC 56, Tampa, FL 33612-3805, 813-974-6681, mforthof@hsc.usf.edu, (2) Florida Health Literacy Study, University of South Florida, 4809 E. Busch Blvd., Suite 104, Tampa, FL 33617

Amid a growing emphasis on developing evidence-based public health programs and interventions with documented accountability for improving health outcomes, public health researchers increasingly turn their attention to maximizing the validity and feasibility of community-based investigations. One of the greatest challenges associated with designing community-based investigations is the identification of matched sites for randomization or the selection of appropriately equivalent comparison sites when the intervention site is predetermined. Communities vary dramatically in their social, political, and programmatic contexts, and the salience of these unique landscapes for assessing program impact is appreciated more than ever by public health practitioners and researchers. To advance the rigor of community-based investigations, methodological approaches that are sensitive to the unique applied contexts in which the research is conducted, while also maintaining the validity of comparisons, is needed. This presentation will illustrate how Euclidian Distance, an underutilized statistic, has been applied to guide the identification of intervention and comparison sites in the Florida Health Literacy Study, an evaluation of a health literacy program promoting management of diabetes and hypertension among patients in Florida’s community health centers. A comprehensive health center profiling process was conducted, followed by estimation of Euclidian Distances between each pair of community health centers. These estimates then guided assignment of community health centers to “matched” pairs for subsequent random assignment to intervention or comparison groups. The presentation will review the methods used to determine matches and the implications of estimated Euclidian Distances for equivalence of matched health centers on study baseline data.

Learning Objectives: At the conclusion of this session, participants will be able to

• Articulate the challenges associated with planning community-based randomized trials.
• Assess the applicability of Euclidian Distance as a measure of community/site matchability.
• List the factors appropriate for inclusion in a site matching analysis.

Keywords: Methodology, Community Research

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.