Davidson 1992.

Study characteristics
Methods	Randomised trial
Participants	Primary care physicians (PCPs)
Interventions	Suffolk County, New York, USA Intervention (1): capitation. Capitation PCPs were paid USD 6 for 13‐ to 18‐year‐olds; USD 8.50 for 6‐ to 12‐year‐olds; USD 13.50 for 3‐ to 5‐year‐olds; and USD 18.50 for children 2 years or younger. Each month USD 25 was set aside for each child. Intervention (2): Fee‐for‐Service (high rates). PCPs were paid a fee for comprehensive exams (including treatment), routine office visits, initial hospital visits, and follow‐up hospital visits. Control:Fee‐for‐service (low rates). PCPs were paid a fee for the same services as the high‐rate group, but the fee was approximately half the size. Years 1983 to 1985, intervention happened in 1983
Outcomes	Quantity of health services provided Utilisation Primary care physician visits Non‐primary care physician visits Hospitalisations Costs Expenditures
Notes	No details on the characteristics or number of PCPs were reported or compared across the 2 groups. The authors did not compare participating PCPs with those who did not participate.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No specific random component is described.
Allocation concealment (selection bias)	Low risk	The primary care physicians signed up for the demonstration without knowing whether they would be prepaid or paid fee‐for‐service at market‐level rates.
Baseline outcome measurements similar	High risk	Utilisation data for the 6 months prior to the programme showed baseline differences amongst groups.
Baseline characteristics similar	High risk	There is no report of characteristics of providers in the text or tables.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing data
Knowledge of the allocated interventions adequately prevented	Low risk	The outcomes are objective.
Study adequately protected against contamination	Low risk	The comparison group consisted of a sample of children in the demonstration area who had not been asked to join the programme.
Selective reporting (reporting bias)	Low risk	The outcomes mentioned in the methods were reported.
Other bias	Unclear risk	No more information