Rural–urban disparities in primary care and geographic continuity of care for children with medical complexity

Purpose Rural‐residing children with medical complexity (CMC) may receive fragmented care given clinician shortages in rural communities. This study characterized differences in continuity of care between rural‐ and urban‐residing CMC, applying novel measures of geographic care continuity and assessing associations between continuity, neighborhood social disadvantage, and unplanned hospital utilization. Methods This retrospective cohort study analyzed 2012–2017 all‐payer claims data from Colorado, Massachusetts, and New Hampshire. After identifying CMC using validated algorithms, we calculated three continuity measures: (i) primary care continuity using the Bice–Boxerman Continuity of Care Index (CoCi), (ii) geographic continuity applying the CoCi at the county‐level, and (iii) proportion of clinic visits within one's home county. We specified regression models to estimate rural–urban differences and interactions between rurality and neighborhood disadvantage, and to model associations between care continuity and unplanned hospital utilization. Findings Among 93,948 CMC, those who were rural‐residing had higher mean primary care CoCi (50.6 [95% CI: 49.6–51.6] vs. 46.9 [95% CI: 46.6–47.2] for urban‐residing), lower mean county‐level CoCi (66.8 [95% CI: 66.1–67.5] vs. 70.4 [95% CI: 70.2–70.6]) and a lower local care continuity (53.5% [95% CI: 52.5%–54.5% vs. 60.3% [95% CI: 60.0%–60.5%]). Neighborhood social disadvantage was a significant effect modifier of the relationship between rurality and all continuity measures. Higher care continuity was associated with lower risk of unplanned hospitalization and emergency department visits. Conclusion Rural‐residing CMC had higher primary care continuity than their urban‐residing peers but lower geographic continuity. Several associations between rurality and care continuity were moderated by neighborhood social disadvantage, highlighting the importance of considering area‐level characteristics when implementing programs and policies to support this population.