Abstract
Positive deviance approaches, which have been used to identify and study high performers (bright spots) and translate their successes to poorer performers, offer great potential for chronic disease management. However, there are few examples of applying positive deviance approaches across different geographic contexts. Building on prior research that developed a new measure for appropriate diabetes preventive care (DMPrevCare) and identified priority counties for this strategy, we introduce a geospatial approach for identifying bright spot counties and case matching them to priority counties that need improvement. We used the Local Moran’s I tool to identify DMPrevCare spatial outliers, which are counties with larger percentages of Medicare beneficiaries receiving appropriate diabetes preventive care (DMPrevCare) surrounded by counties with smaller percentages of Medicare beneficiaries receiving DMPrevCare. We define these spatial outliers as bright spots. The Robert Wood Johnson Foundation County Health Rankings Peer Counties tool was used to link bright spot counties to previously identified priority counties. We identified 25 bright spot counties throughout the southern and mountain western United States. Bright spot counties were linked to 45 priority counties, resulting in 23 peer (bright/priority) county groups. A geospatial approach was shown to be effective in identifying peer counties across the United States that had either poor or strong metrics related to DMPrevCare, but were otherwise similar in terms of demographics and socioeconomic characteristics. We describe a framework for the next steps in the positive deviance process, which identifies potential factors in bright spot counties that positively impact diabetes care and how they may be applied to their peer priority counties.