Accounting for morbidity in capitation payments: a person-based workload formula for primary medical care in England

Background: Accurate needs-based capitation is key to effective and equitable primary care funding. Most capitation schemes use only basic demographic and area characteristics.

Objective: We developed capitation weights for general practices in England using morbidity indicators recorded in primary and secondary care.

Methods: We used primary care records from the Clinical Practice Research Datalink (CPRD) linked with Hospital Episode Statistics (HES) for 12,667,755 patients registered with 1,397 general practices on 1 April 2018. Using linear regression models, we estimated the effects on cost-weighted clinical appointments of patient age and gender, ethnicity, area-level deprivation, new registration, and morbidity (four sets of indicators covering 20 to 209 conditions). We included practice fixed-effects to adjust for differences in capacity and productivity. We applied the coefficients on patient characteristics as need-weights to data available nationally and we calculated weighted-patients for all 6,892 practices in England.

Results: Most patients (71%) had at least one appointment per-year. The average annual workload per-patient was £110, with large variations across patients (range £0-£882) and practices (£47-£179). Workload increased with age and with deprivation, but their direct effects halved when including morbidity in the model. Including morbidity widened the range of weighted-patient between practices at the 5th and 95th percentiles (from 0.84-1.14 to 0.79-1.16 ) and in the least and most deprived deciles (from 0.96-1.04 to 0.95-1.06).

Conclusion: Needs-based capitation weights accounting for morbidity and adjusting for unexplained variations in practice capacity and productivity increase workload differentiation and direct resources toward practices in more deprived areas.