Does the Pay-For-Performance Program Reduce Emergency Service Utilization for Asthma Patients?

• Main Authors: Chang-Hung Tsai, 蔡昌宏
• Other Authors: Blossom Yen-Ju Lin
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/15368230434229503103

碩士 === 中國醫藥大學 === 醫務管理學系碩士在職專班 === 100 === Purpose: The asthma population in Taiwan was about 2 million people and increasing both in number and in impact in recent years. In this study, we aimed to discuss about the issue of “Does the pay-for-performance (p4p) program reduce emergency service utilization for asthma patients in Taiwan?”. Method: A random sample of a million people from the National Health Insurance Research Database Data from 1997 to 2010 was extracted for analyses. In this study, we defined the newly-diagnosed asthma patients with insurance from 1998 to 2009 as study population. A total of 65,489 people were included in our study. Then we used propensity score to match asthma patients joining p4p program to others in a 1:4 ratio. After matching, 49,195 subjects were left for subsequent analyses. Then we used Cox proportional hazards model to analyze the impact of P4P program on emergency service utilization after adjusting for demographic characteristics, co-morbidity, principle medical facility characteristics, and asthma severity. Result: Overall, the emergency service utilization of asthma patients joining p4p program was obviously higher than others (HR=1.57, 95% CI=1.49-1.67, P<0.001). And we also found the emergency service utilization positively correlated with asthma severity (HR=2.11-6.20, p<0.001) and negatively correlated with age (HR=0.39-0.84, p<0.001). Conclusion and Recommendation: It is important to improve the effectiveness of P4P program in Taiwan. We suggested that outcome indicators such as emergency service utilization more closely reflect P4P effectiveness than process indicators. We still needed to develop a risk adjusted system to evaluate the asthma patients to avoid physician selecting patients. In conclusion, we still need to take more efforts to improve the pay-for-performance program effectiveness.