Gosden 2003.

Study characteristics
Methods	Controlled before‐after study
Participants	General practitioners
Interventions	England Intervention:salary contract. GPs can be employed on a salaried basis to provide ‘personal medical services’ (PMS) alongside other services needed in the locality. These salaried schemes were intended to improve GP recruitment and retention and quality of care, particularly in underserved areas. Control: capitation and FFS. Standard contract GPs (GMS) are financially rewarded for increasing patient list size (through capitation payments) and for providing specific types of services (through target payments). Years 1998 to 2001, intervention happened in 1999 May
Outcomes	Healthcare provider outcomes: Working times: control group longer 2.22 (−8.69, 13.14), and times on 5 specific activities were analysed Quantity of health care outcome: Number of surgery consultations per GP per practice: intervention group more 4.47 (−36.55, 45.5) Number of patients seen out‐of‐hours per GP per practice: intervention group more 4.5 (−1300.46, 1309.46) % consultations in which prescription given per GP per practice: intervention group less −0.07 (−0.20, 0.05) % consultations in which referral made per GP per practice: intervention group less −0.01 (−0.07, 0.05) Total list size practice‐based: intervention group less −213.4 (−942.11, 515.31) List size per whole time practice‐based: intervention group more increase 20.20 (−294.60, 334.99) Cervical cytology (%) practice‐based: intervention group higher 0.38 (−14.90, 15.67) Childhood immunisation (%) practice‐based: intervention group lower −1.08 (−17.95, 15.80) Pre‐school booster (%) practice‐based: intervention group lower −3.08 (−9.63, 3.47) Quality of health care outcome (patients' assessment of quality of care) Access: −1.07 (−4.84, 2.69) Technical care: 1.28 (−4.62, 7.18) Communication: −0.09 (−6.84, 6.66) Interpersonal care: 0.18 (−6.26, 6.62) Overall satisfaction: 2.36 (−6.12, 1.40) Receptionists: 0.64 (−4.81, 6.09) Continuity of care: 0.29 (−5.37, 5.95) Trust in doctor: 0.57 (−3.88, 5.03) Doctors’ knowledge of patient: 1.18 (−3.88, 6.23) Practice nursing: −0.23 (−4.3, 3.83) Co‐ordination of care: 0.63 (−13.43, 14.69) Appropriate referral: 0.91 (−5.31, 7.12) Recommend to a friend: 1.71 (−5.79, 9.21)
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	This is a controlled before‐after study (years 1998 to 2001, intervention happened in 1999 May).
Allocation concealment (selection bias)	High risk	This is a controlled before‐after study (years 1998 to 2001, intervention happened in 1999 May).
Baseline outcome measurements similar	Unclear risk	No description
Baseline characteristics similar	High risk	Doctors in GMS and PMS practices differed in their age and gender distributions.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	No description
Knowledge of the allocated interventions adequately prevented	High risk	Not all the outcome indicators are objective.
Study adequately protected against contamination	Low risk	The allocation was by practice, and it is unlikely that the control group received the intervention.
Selective reporting (reporting bias)	Low risk	All relevant outcomes in the methods section are reported in the results section.
Other bias	Unclear risk	No description