“…The odds of treatment persistence were significantly higher in rural versus urban patients in one report (adjusted OR 1.28, 95% CI 1.25, 1.32) [46], but was not statistically different in a second study (OR 0.96, 95% CI 0.55, 1.67) [50], Medication adherence data from three studies (4 cohorts) were pooled and showed no statistically significant difference between rural and urban patients (unadjusted OR 0.94, 95% CI 0.39, 2.27, p = 0.89, I 2 = 91%) Table 1[29,37,70]. In two other reports with data not suitable for meta-analysis, medication adherence was significantly higher in one study (92% versus 83% of doses taken, p = 0.01) [73], and significantly lower in the other (10% versus 17% of patients compliant, p < 0.01) [74], for rural versus urban patients, based on unadjusted data. The three studies reporting treatment adherence adjusted for confounders reported adherence to be significantly higher in rural patients (OR medication possession ratio ≥0.8: antihypertensive agents 1.2, 95% CI 1.1, 1.3; ASA 1.14, 95% CI 1.10, 1.18; ACEI/ARB 1.18 95% CI 1.14, 1.23) or rural men (OR adherent 4.0, 95% CI 1.1, 13.9) [29,46,70].…”