Primary care providers' preferences for pay-for-performance programs: a discrete choice experiment study in Shandong China.

BACKGROUND: Pay-for-performance (P4P) schemes are commonly used to incentivize primary healthcare (PHC) providers to improve the quality of care they deliver. However, the effectiveness of P4P schemes can vary depending on their design. In this study, we aimed to investigate the preferences of PHC providers for participating in P4P programs in a city in Shandong province, China. METHOD: We conducted a discrete choice experiment (DCE) with 882 PHC providers, using six attributes: type of incentive, whom to incentivize, frequency of incentive, size of incentive, the domain of performance measurement, and release of performance results. Mixed logit models and latent class models were used for the statistical analyses. RESULTS: Our results showed that PHC providers had a strong negative preference for fines compared to bonuses (- 1.91; 95%CI - 2.13 to - 1.69) and for annual incentive payments compared to monthly (- 1.37; 95%CI - 1.59 to - 1.14). Providers also showed negative preferences for incentive size of 60% of monthly income, group incentives, and non-release of performance results. On the other hand, an incentive size of 20% of monthly income and including quality of care in performance measures were preferred. We identified four distinct classes of providers with different preferences for P4P schemes. Class 2 and Class 3 valued most of the attributes differently, while Class 1 and Class 4 had a relatively small influence from most attributes. CONCLUSION: P4P schemes that offer bonuses rather than fines, monthly rather than annual payments, incentive size of 20% of monthly income, paid to individuals, including quality of care in performance measures, and release of performance results are likely to be more effective in improving PHC performance. Our findings also highlight the importance of considering preference heterogeneity when designing P4P schemes.

Longitudinal Association of Salaries for Medical Staff With Medical Service Utilization and Expenditure in China, 2007-2016.

Objectives: To test the hypothesis that higher salary levels of the medical staff are associated with lower medical service utilization and expenditure. Methods: Using longitudinal data from 31 Chinese provinces for the period 2007-2016, we constructed fixed effects models to analyze the association between the salary of medical staff and medical service utilization, medical expenditure, medication expenditure, and medication proportion. Results: A 10,000 CNY increase in medical staff's salaries was associated with a 0.89% decrease in the average number of annual inpatient admissions per person; 1.88 and 1.59% decreases in average expenditures per outpatient visit and inpatient admission, respectively; 3.05 and 2.66% decreases in drug expenditures per outpatient visit and inpatient admission, respectively; 0.58 percent point and 0.39 percent point decreases in the share of drug expenditure in outpatient and inpatient, respectively. When medical staff's salaries increased by 450,000 CNY, the turning point was reached when the maximum medical expenditure savings offset the medical staff salary increases, yielding a 634 billion CNY surplus from medical expenditure. Conclusions: Our results supported the hypothesis that higher salary levels of the medical staff are associated with lower medical service utilization and expenditure. Further studies are requested to test whether higher medical staff's salaries will attenuate over-treatment and that savings from reduced prescriptions and service charges will offset the increased salaries of medical staff.