:::

研討會總覽

The Dynamic Effect of Prospective Payment on Patient Outcomes and the Length of Stay: With Evidence from Thyroidectomy Hospital Panel Data in Taiwan

2012/02/03
  • 研討會日期 : 2012-02-03
  • 時間 : 14:30
  • 主講人 : Professor Lilin Liang
  • 地點 : Conference Room B110
  • 演講者簡介 : Lilin Liang received her Ph.D. in Health Economics from London School of Economics and Political Science in 2010. She is currently serving as Assistant Professor of China Center for Health Development Studies at Peking University. Her research field is in health econometrics, health economics, modelling of provider behaviour, health care financing and payment mechanisms, health services efficiency, data envelopment analysis, longitudinal research, and evaluation of health policies and regulations.
  • 演講摘要 : Previous evidence indicates that prospective payment reduced the length of stay (LOS) and unnecessary care. However, concerns have arisen over poor outcomes associated with reduced LOS. This study proposes two models to assess how the phasing-in of the prospective payment scheme affected the average LOS and complication rates. The models were estimated by using the random-effects poisson and tobit models, in which the adjusted average LOS, case selection and patient, surgeon and hospital characteristics were controlled for. The data were collected for all patients who underwent thyroidectomy between 1998 and 2009 under Taiwan’s National Health Insurance Scheme (NHI). In 1999, the NHI changed the reimbursement for unilateral thyroidectomy from fee-for-service to case-based payment and changed its bilateral counterpart similarly in 2004. The results suggest that the 1999 reform did not reduce the average LOS but worsened the outcome. In contrast, the 2004 reform reduced the average LOS without any deterioration of outcome, which may be partly attributed to quality and efficiency improvement. This study finds that the advantages of prospective payment are more evident when it achieves full coverage. For example, the marginal effect associated with case selection on worsening outcomes was less in 2004 than in 1999.