The impact of rural classification systems on a comparison of risky drug-related behaviors in Kentucky and Ohio counties

To analyze county-level differences in injection drug use (IDU) behaviors and overdoses in Kentucky and Ohio, and to assess the impact of using different models to classify counties as rural. Existing clinical data were pooled from three regional programs: 1) an opioid overdose prevention program (OOPP) within a residential addiction treatment program (n=163), 2) a regional evaluation of four OOPPs (n=655), and 3) a mobile syringe services program (n=1,379), representing 17 counties each in Kentucky (n=516) and Ohio (n=1,618). Pearson chi-square tests were used for dichotomous variables and an analysis of variance (ANOVA) was used to compare continuous variables. A Tukey HSD posttest was used to identify statistically significant comparisons. Age at first IDU was consistently younger in rural counties. However, findings changed based on the classification scheme used. This was also true for the prevalence of methamphetamine use. The rurality classification applied at the county level may be an important consideration in the analysis of research studies.