Hickson 1987.

Study characteristics
Methods	Randomised trial
Participants	Primary care physicians (PCPs)
Interventions	Continuity Clinic, USA Intervention payment:FFS group received USD 2 per visit Control payment: Salary group received USD 20 per month Year 1983 to 1984, intervention happened in 1983
Outcomes	Quantity of health services provided Patients enrolled per PCP Patient visits attended per PCP Percentage visits attended by patient’s primary physician (continuity) Emergency room visits Scheduled visits Completed visits Sick, primary visits Sick follow‐up visit Well‐child visits The number of recommended visits missed The number of visits in excess of the recommended Adverse effects The number of well patients
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Paediatric residents were initially matched for year of training and the day of the week their clinics were held. Each pair was randomised by the flip of a coin.
Allocation concealment (selection bias)	Low risk	Patients are recruited by residents from the health unit or assigned by the secretary. No attempts were made by the faculty or clinic staff to adjust the size of these groups.
Baseline outcome measurements similar	Unclear risk	No comparison of outcome measurements was reported.
Baseline characteristics similar	Low risk	No significant PCP differences in time since graduation, gender, interest in continuity. No significant patient differences in salary group, mother's age, number of children at home, number of fathers in home.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing outcome data
Knowledge of the allocated interventions adequately prevented	Low risk	Outcomes are objective.
Study adequately protected against contamination	Unclear risk	No more information
Selective reporting (reporting bias)	Low risk	No reporting bias
Other bias	Unclear risk	No more information