OBJECTIVE: To develop and evaluate alternative methods of adjusting primary medical care capitation payments for variations in relative need for health care among enrolled practice populations. METHODS: We developed alternative needs-based capitation formulae and applied them to a sample of capitation-funded primary care practices to assess each formula's performance against a reference standard of capitation payments based on age, sex and self-assessed health status of the enrolled populations. The alternative formulae were based on: (1) age and sex; (2) age, sex and individually-measured socioeconomic characteristics; (3) age, sex and socioeconomic characteristics imputed from census data for enrollees' neighbourhood of residence; (4) age, sex and standardized mortality ratio for enrollees' neighbourhood of residence. RESULTS: Age/sex-adjusted capitation payments for the six practices studied ranged from 10% higher to 18% lower than the reference standard payments. Capitation formulae based on socioeconomic and mortality data did not perform consistently better than the current age/sex-based formula. CONCLUSIONS: Primary medical care capitation payments adjusted only for age and sex do not reflect the relative health care needs of enrolled practice populations. Our alternative formulae based on socioeconomic and mortality data also failed to reflect relative needs. Methods that use other approaches to adjusting for differences in relative need among enrolled populations should be investigated.
OBJECTIVE: To develop and evaluate alternative methods of adjusting primary medical care capitation payments for variations in relative need for health care among enrolled practice populations. METHODS: We developed alternative needs-based capitation formulae and applied them to a sample of capitation-funded primary care practices to assess each formula's performance against a reference standard of capitation payments based on age, sex and self-assessed health status of the enrolled populations. The alternative formulae were based on: (1) age and sex; (2) age, sex and individually-measured socioeconomic characteristics; (3) age, sex and socioeconomic characteristics imputed from census data for enrollees' neighbourhood of residence; (4) age, sex and standardized mortality ratio for enrollees' neighbourhood of residence. RESULTS: Age/sex-adjusted capitation payments for the six practices studied ranged from 10% higher to 18% lower than the reference standard payments. Capitation formulae based on socioeconomic and mortality data did not perform consistently better than the current age/sex-based formula. CONCLUSIONS: Primary medical care capitation payments adjusted only for age and sex do not reflect the relative health care needs of enrolled practice populations. Our alternative formulae based on socioeconomic and mortality data also failed to reflect relative needs. Methods that use other approaches to adjusting for differences in relative need among enrolled populations should be investigated.