THE ASSOCIATION BETWEEN LENGTH OF STAY, MEDICAL EXPENSES, AND PSYCHIATRIC PATIENT DUMPING – A STUDY OF PROVIDER COMPETITION UNDER GLOBAL BUDGETING IN TAIWAN
Embargo until
Date
2015-04-06
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
Abstract
Background:
Legislation of the Mental Health Law in 1990, along with the amendments of Disabled Citizens Act, and the establishment of the National Health Insurance (NHI) program in 1995 had provided access to care for the mentally ill in Taiwan, divided into six NHI branch areas. The above legislative acts have created a universal coverage system and removed the majority of financial barriers to psychiatric admission. However, one way to minimize financial losses of the mental health service providers is to transfer psychiatric patients to another institution before complete inpatient treatment is received; these economically motivated transfers are defined here as patient dumping. The level of psychiatric patient dumping could be reflected by the practices of NHI utilization and costs review focusing on length of stay (LOS) and medical expenses (MED) per hospital admission. As might be expected, the intensity of managed care competition at each branch area was positively correlated with the severity of psychiatric patient dumping. The incentives of psychiatric patient dumping could have been exacerbated due to the rivalry among hospitals in the local market, produced by the global budget (GB) system implemented since July 1 of 2002.
Aims:
The objectives of this study are to: (1) examine the associations between length of stay, medical expenses and psychiatric patient dumping; (2) identify if there are relationships between psychiatric patient dumping and provider competition through GB impact among hospitals in each NHI branch area.
Methods:
Hospital discharge records of the psychotic patients were obtained from a one million person subset of the National Health Insurance Research Database (NHIRD) for the period from 1999 to 2005. The psychotic cases of re-ER within 3 days and re-hospitalization within 14 days, received by different hospitals, were counted as patient dumping events. Chi-square tests & t-tests were applied to descriptive statistics. Generalized Estimating Equation (GEE) was used to estimate the associations between length of stay, medical expenses and psychiatric patient dumping. The linear regression model examines the trends of psychiatric patient dumping before and after the implementation of GB.
Results:
Findings from the study are that dumping patient/events, male gender, and being treated in regional or district hospitals were factors associated with higher LOS, with the exception of the Kao-Ping area having lower LOS. There were additional factors associated with higher medical expenses than longer LOS. These included dumping patient/events, older age, male gender, being treated in regional or district hospitals, and hospitals located in the Northern area, the Southern area and the Eastern area as compared to other areas. Being treated in clinics or hospitals located in the Kao-Ping area was associated with lower medical expenses. The market provider competition in each area, measured by Herfindahl-Hirschman Index (HHI), remained somewhat stable during this 7-year time period. However, the GB system contributed to the sharp increasing trends of psychiatric patient dumping in all six branch areas, especially the Taipei area.
Conclusions:
The dumped patient/events had significantly higher LOS and medical expenses than the un-dumped patient/events. Because almost 100% of the residents are insured in Taiwan, the main reason for being dumped is more likely about reimbursement incentives from the Bureau of National Health Insurance (BNHI), Department of Health and Welfare. While government decision makers attempt to use GB to control mental health utilization and costs, the hospital‟s response through stable provider competition might have triggered patient dumping. A complex mechanism to avoid psychiatric patient dumping should be explored and established.
Description
Keywords
Length of stay, Medical expenses, Generalized Estimating Equation, Psychiatric patient dumping, Market competition, Global budget