Geo-spatial, sociodemographic, and community risk factors associated with access to pharmacies, over-the-counter syringe sales, and hotspot infectious disease clusters in Massachusetts

background: In Massachusetts, over-the-counter syringe sales and syringe possession legislation was adopted in 2006, requiring that all pharmacies make syringes available for sale without a prescription. While reported hepatitis C virus (HCV) infections are an increasing concern in subpopulations in the Commonwealth, people who inject drugs in many regions still face challenges accessing sterile syringes. We used statistical models, a geographic information system (GIS), and spatial epidemiological methods to identify “regions of risk”, where sterile syringe access was limited, and “regions of disease”, where spatial disease clusters were present, and factors associated with these regions.

methods:We compiled pharmacy, sociodemographic, and disease surveillance data for Massachusetts census tracts (n=1467) from 2002-2013. We conducted logistic regression analyses to determine factors associated with pharmacy access. We used GIS to identify locations throughout the state where geographic access to pharmacies was limited according to calculated five-, ten-, and fifteen-minute walk-time and drive-time buffers. Finally, we conducted spatial Getis Ord Gi* analyses to identify statistically significant hotspot clusters of pharmacies and infectious disease.

results: In bivariate analyses, total population, population density, percent of population that was Hispanic, percent high school graduates, percent living in poverty, median household income, and median age were associated with pharmacy access. In multivariable regressions, adjusted for the aforementioned variables, population density (adjusted odds ratio [AOR]: 1.52; 95% confidence interval [CI]: 1.15, 2.03), median age (AOR: 2.26; 95% CI: 1.05, 4.85), and median household income (AOR: 0.39; 95% CI: 0.26, 0.59) were independently associated with pharmacy access. We identified nine HCV hotspot clusters across Massachusetts, with the largest clusters in Boston, New Bedford/Fall River, Worcester, and Springfield (p<0.05). According to walk-time and drive-time buffer analyses, pharmacy syringe sales coverage surrounding HCV clusters appeared adequate in most parts of the state, but greater access to 24-hour pharmacies and over-the-counter syringe sales may be merited in southeastern and western regions of the Commonwealth.

conclusions: Combined geo-spatial and statistical analyses can help identify regions of risk and regions of disease, and inform public health policy decisions. We identified several locations where infectious disease clusters were a concern, and where enhanced access to pharmacies and sterile syringes could strengthen harm reduction efforts in Massachusetts.