Skip Navigation


International Journal for Quality in Health Care Advance Access originally published online on June 28, 2005
International Journal for Quality in Health Care 2005 17(5):421-426; doi:10.1093/intqhc/mzi055
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
17/5/421    most recent
mzi055v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Tsai, Y.-W.
Right arrow Articles by Chen, P.-F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tsai, Y.-W.
Right arrow Articles by Chen, P.-F.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

International Journal for Quality in Health Care vol. 17 no. 5 © The Author 2005. Published by Oxford University Press on behalf of International Society for Quality in Health Care; all rights reserved

The effect of changing reimbursement policies on quality of in-patient care, from fee-for-service to prospective payment

Yi-Wen Tsai1,3, Yi-Chou Chuang2, Weng-Foung Huang3, Lai-Chu See2, Chung-Lin Yang1 and Pei-Fen Chen1

1 National Health Research Institutes, Division of Health Policy Research, Miaoli, Taiwan, 2 Chang-Gung University, Tao-Yuan, Taiwan, 3 National Yang-Ming University, Taipei, Taiwan

Objective. Using insurance claims for hemorrhoidectomies, we examined the effect of Taiwan’s Bureau of National Health Insurance’s case payment system, a fixed case payment rate method used to reimburse health care providers for in-patient care.

Design. This observational natural experimental study examined changes in medical care that occurred between two phases: the 9 months before case payment system was implemented on 1 October 1997 and the 9 months afterwards. The changes were analyzed by performing linear regressions with interaction between hospital type and the implementation of case payment system.

Setting. This study was based on total claim data from National Health Insurance.

Study participants. A total of 23 638 hemorrhoidectomy insurance claims.

Main outcome measures. Length of stay, number of medical services, and number of drug prescriptions. Medical services were stratified into those that were considered minimal requirements and those considered optional by the Bureau of National Health Insurance.

Results. Over the 18-month period, the number of patients increased by 23.7%. After the case payment system was implemented, length of stay decreased by 0.59 days (P < 0.0001), the number of minimally required services increased by 2.19 to 4.24 items (P < 0.0001), the number of optional service items decreased by 0.32 items (P < 0.0001), and drug prescription decreased slightly by 0.58 to 0.99 items (P < 0.0001) per hospitalization.

Conclusions. The case payment system successfully shortened length of stay without significantly sacrificing the provision of services.

Keywords: case payment, drug prescription, moral hazard, national health insurance, Taiwan

Address reprint requests to Yi-Wen Tsai, Ph.D. Associate Investigator, Division of Health Policy Research, National Health Research Institutes, no. 35, Keyan Road, Zhunan Town, Miaoli County 350, Taiwan. E-mail: ivytsai{at}nhri.org.tw

Accepted for publication April 1, 2005.


   Introduction
Top
 Introduction
Methods
 Results
 Discussion
 Acknowledgements
 References
 
On 1 March 1995, Taiwan implemented a national health insurance plan that provided its citizens with universal health care coverage. On 1 October 1997, the Bureau of National Health Insurance implemented a case payment method to reimburse hospitals at a fixed rate for specific surgical procedures and related in-patient care. Since that time, the fixed reimbursement rates have been revised twice, 1 July 1998 and 1 June 2001. For instance, the NT$22 100 rate per hemorrhoidectomy rose to NT$22 660 and then to $23 310. The number of surgical procedures covered by the case payment system during the same period, expanded from 28 to more than 50. One study has found that this reimbursement method decreased in-patient length of stay for laparoscopic cholecystectomies, but not total expense [1]. Another study showed that this method encouraged health care providers to change the status of an in-patient to outpatient to control costs of treatment [2]. Because case payment is the system on which the National Health Insurance global budget for in-patient care is based, a systematic review of how effectively it reduces costs and influences the behavior of health care providers is critical.

Taiwan’s case payment method, like the United States’ prospective payment system (PPS)/DRG system, aims to reduce in-patient costs by imposing the financial pressure of cost-sharing on medical service providers. Many studies have identified unexpected outcomes that result from the responses to such fixed-price policies. One such response is ‘gaming’: reclassifying patients into higher paying categories, increasing admissions, or prescribing more profitable procedures [3]. Other problems associated with a cost-sharing policy are risk selection, cost shifting, and moral hazard. Risk selection occurs when hospitals select their patient mix for higher profit [4–8]. Cost shifting occurs when a purchaser extracts a price discount from a health care provider and that health care provider in turn increases the price it charges to other purchasers, one’s gain becoming another’s loss [9–12]. Moral hazard, directly related to care of the patient, occurs when hospitals reduce the range of medical procedure options and the length of stay per discharge [5,13–16].

Taiwan’s case payment system is based on the type of surgical procedure only and does not stratify payments according to severity of illness or other health-related factors. It reimburses health care providers retroactively based mainly on a hybrid of the fixed-price and fee-for-service methods (see Figure1). Health care providers are reimbursed a fixed sum (FL in Figure1) when the claim is lower than the pre-set threshold (FH in Figure1). Since the cost of care can vary with severity of disease, fee-for-service is allowed for expenditures higher than the pre-set threshold. Fee-for-service is highly regulated: the Bureau of National Health Insurance sets monthly ceilings (% in Figure1). Fee-for-service claims, submitted with more detailed medical records, take longer to process than fixed-payment claims. The values of fixed-rate (FL), pre-set threshold (FH), and percentage ceiling (%) vary according to accreditation of medical facility. In ‘medical centers’, hemorrhoidectomies were reimbursed at a fixed rate of NT$22 100 per case, a threshold of NT$27 500, and percentage ceiling of 10%. In ‘regional hospitals’, those figures were NT$21 600, NT$27 500, and 5%, respectively, and in ‘district hospitals’, NT$21 100, NT$27 500, and 5%.



View larger version (40K):
[in this window]
[in a new window]
  		Figure 1 The case payment reimbursement policy of Taiwan National Health Insurance.

 

Under this payment scheme patients could be classified into three groups. Low-cost patients, considered the most profitable, are patients whose total medical expenses fall below the fixed price (FL) and for whom the provider is reimbursed at the fixed-price rate (FL). Intermediate-cost patients, considered less profitable, are patients whose total medical expenses exceed the fixed price (FL) and for whom the provider is reimbursed at the FL rate. A high-cost patient is a patient whose total medical expenses exceed the threshold value (FH) and for whom the provider is reimbursed at fee-for-service rates.

Taiwan’s case payment system also regulates quality of care by outlining ‘the minimally required service’, which are recommendations of what medical services, excluding drug prescriptions, a hospitalized patient should be provided per day and how long a minimum length of stay would be for each case (see Appendix1). A health provider is reimbursed only if it provides a patient with at least 65% of the minimally required medical service items.


View this table:
[in this window]
[in a new window]
  		Appendix 1 Hemorrhoidectomy or anal fistulectomy (in-patient) procedure requirement table

 
To find out if there are moral hazard problems in health care provider response to Taiwan’s case payment system, we studied changes in length of stay, medical services, and drug prescriptions for patients undergoing hemorrhoidectomies before and after the system was implemented.


   Methods
Top
 Introduction
 Methods
Results
 Discussion
 Acknowledgements
 References
 
Data sources
Our data consisted of several data sets provided by the hospital on claims submitted to the Bureau of National Health Insurance. We selected two study phases: the first phase (fee-for-service payment phase) was the 9 months before the case payment system was implemented (January 1997–September 1997), and the second phase (prospective payment phase) was the 9 months between case payment system was implemented and the first time the case payment system reimbursement rates were revised (October 1997–June 1998).

A hemorrhoidectomy patient is defined by Taiwan’s case payment program as someone who has undergone a hemorrhoidectomy (ICD9CM procedure code = 49.46: Excision of hemorrhoids) as major surgery or someone who has undergone an anal fistulotomy (49.11: Anal fistulotomy) or anal fistulectomy (49.12: Anal fistulectomy) as a major procedure and a hemorrhoidectomy (49.46) as a secondary procedure. This definition of hemorrhoidectomy patients includes MD-DRG 157 (with complication and co-morbidity) and MD-DRG 158 (without complication and co-morbidity) of Major Diagnostic Category 6 (Diseases and Disorders of the Digestive System).

Because we focus on moral hazard as a response to fixed-rate payment, we chose patients (hemorrhoidectomy patients) whose total medical expenses fell below the threshold value. Before the case payment system, health care providers were reimbursed a fee-for-service for such patients. Since the introduction of the case payment system, they were reimbursed at a fixed rate for a hemmorrhoidectomy regardless of the number of associated services, up to the threshold (FH). Our sample consisted of 23 638 cases reimbursed at a fixed rate, drawn from the 24 408 hemorrhidectomies claimed for during the two study phases. There were 563 and 207 patients whose total medical expense exceeded the threshold value and were excluded from this study.

In this study, we explore the extent to which the case payment system reduced length of stay, medical services, and drug prescriptions. To analyze its relationship to moral hazard, we compared changes in the number of medical services and drug prescriptions only. Medical service variables, listed on claims forms, were further categorized into minimally required services and optional services. Drug prescription variables were the total number of drug prescriptions.

Statistical analysis
Descriptive statistics including mean, standard deviation, frequency, and percentage were used in this study. The chi-squared test was used to compare changes in categorical data such as sex, complications/co-morbidity, and secondary procedures between the two phases. Analysis of variance (ANOVA) was conducted to compare continuous variables. OLS linear regressions were used to examine differences between the two phases with regard to the changes in medical services and drug prescriptions, adjusting for the effects of hospital type, sex, co-morbidity/complications, and secondary procedures. We added interaction terms of hospital type and phase to examine the response of different types of hospital to the implementation of the case payment system. A linear function of the coefficient was estimated and tested in the model with the interaction terms. In order to simplify interpretation, we show the effect of implementing the case payment system on the three types of hospital instead of the interaction terms. The dummy variable of secondary procedure was defined as 1 if the patient received any other procedure beside hemorrhoidectomy while in the hospital. The dummy variable co-morbidity/complication was defined as 1 if any of the four secondary diagnostic codes was defined as a co-morbidity/complication of a disease in Major Diagnostic Category 6 (Diseases and Disorders of the Digestive System), as defined in MD-DRG Version 16.0.


   Results
Top
 Introduction
 Methods
 Results
Discussion
 Acknowledgements
 References
 
Over the 18-month study period, the number of hemorrhoidectomies increased by 23.74%, from 10 565 cases in fee-for-service payment phase to 13 073 cases in the prospective payment phase (Table1). After the system was implemented, a significant proportion of intermediate-cost patients were shifted into low-cost patients. While low-cost patients had previously made up 86.60% of the total, they now made up 98.98%. The patients were more likely to be younger, male, have complications and co-morbidities, and require secondary procedures. At ‘medical centers’, the proportion of patients receiving hemorrhoidectomies decreased from 39.78% in the fee-for-service payment phase to 33.12% in the prospective payment phase; at ‘district hospitals’ the proportion increased dramatically. The lengths of stay decreased from 3.81 days to 3.20 days. The number of minimally required medical services and optional services increased by 3.59 and 0.09 items per case, respectively; drug prescriptions decreased slightly by 0.69 items.


View this table:
[in this window]
[in a new window]
  		Table 1 Descriptive information of hemorrhoidectomy patients in Taiwan from 1997 to 1998

 

Multivariate linear regressions were performed on length of stay, minimally required medical services, optional medical services, and drug prescriptions (Table2). The length of stay was significantly shortened by 0.59 days at ‘medical centers’, 0.83 days at ‘regional hospitals’, and 0.67 days at ‘district hospitals.’ Minimal medical services increased by 2.19 items at ‘medical centers’, 3.32 items at ‘regional hospitals’, and 4.24 items at ‘district hospitals.’ While optional services were reduced by 0.32 items at ‘medical centers’, they were insignificantly increased by 0.10 (P = 0.21) in ‘regional hospitals’ and by 0.14 (P > 0.05)‘district hospitals.’ Drug prescriptions were significantly reduced by 0.99 (P < 0.0001) prescriptions in ‘medical centers’, 0.58 (P < 0.0001) in ‘regional hospitals’, and 0.85 (P < 0.0001) in ‘district hospitals.’


View this table:
[in this window]
[in a new window]
  		Table 2 Multivariate linear regressions on length of stay, medical service items, and drug prescription items for hemorrhoidectomy patients in Taiwan, 1997–1998

 


   Discussion
Top
 Introduction
 Methods
 Results
 Discussion
Acknowledgements
 References
 
In this study, we evaluate changes in length of stay and number of health services and drug prescriptions provided to hemorrhoidectomy in-patients before and after a new method of reimbursing health care providers was introduced by a national insurance program. Because we were looking for evidence of moral hazard, we excluded patients whose medical care costs exceeded the case payment threshold and their care was reimbursed based on fee-for-service. By excluding these patients in the study of moral hazard, it is understood that certain cost-shifting information (from case payment to fee-for-service) will be lost, which might bring about the underestimation of the overall effect of the new reimbursement method on medical care expense, length of stay, and medical services. Our study is subject to claim data limitations. The in-patient claims data did not provide us with information on follow-up services and outpatient care provided after discharge. Therefore, moral hazard could not be examined comprehensively.

In spite of these limitations, taking advantage of National Health Insurance’s detailed claim data, we found that the case payment system shortened length of stay without significantly sacrificing the provision of services. Length-of-stay decreased by 0.59–0.83 days, depending on the type of hospital accreditation. The provision of minimally-required medical service items increased by 2.19–4.24 items depending on hospital accreditation, but optional items decreased by only 0.32 items at ‘medical centers’. Based on these observations, we believe that the new reimbursement system successfully shortens the length of stay per case, and the minimal-requirement services stipulated by the Bureau of National Health Insurance has effectively standardized health care for patients having hemorrhoidectomy by reducing the risk of extreme cases of provider moral hazard in which a hospital might reduce or distort services for higher profits. Although the fee-for-service option mitigates the incentive for health care providers to cut costs and reduce expenditure, our study probably did not reflect this incentive because fee-for-service patients comprised such a small proportion of our study population.

However, we found that the case payment system slightly increased the possibility that provider response might involve moral hazard, though in-patient medical services for hemorrhoidectomy did not decrease. The number of drug prescriptions sheds light on financial incentives and provider behavior. Although the proportion of patients with complications and co-morbidity was found to have increased, the number of prescriptions decreased by 0.58 to 0.99. It might be important to find out whether the decrease in drug prescriptions was a result of shifting costs to outpatient care or changing to cheaper generic drugs and whether the decrease affected patient recovery.

Another new area of this study is whether the new reimbursement method brought about a supply-induced increase in the number of hemorrhoidectomies (23.74%). Our study examines the first term response to the case payment system, so future studies may want to extend the study period to discover whether the increase was a one-time response or a response that will continue over time. Finally, similar analyses could be performed on other medical conditions to more clearly understand the overall effect of the case payment system on health care provider behavior.


   Acknowledgements
Top
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
References
 
This research was supported by a National Health Research Institutes Intramural Grant (Grant number: HP-091-CB03, HP-092-PP-09, HP-093-PP-09).


   References
Top
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
 References
 

  1. Lang HC, Chi C, Liu CM. Impact of the case payment reimbursement method on the utilization and costs of laparoscopic cholecystectomy. Health Policy 2004; 67: 195–206.[Medline]

  2. Lin HC, Xirasagar S, Kao SY. Association of hospital ownership with patient transfers to outpatient care under a prospective payment system in Taiwan. Health Policy 2004; 69: 11–19.[Medline]

  3. Ellis RP, McGuire TG. Provider behavior under prospective reimbursement: cost sharing and supply. J Health Econ 1986; 5: 129–151.[CrossRef][Web of Science][Medline]

  4. Ellis RP. Creaming, skimping and dumping: provider competition on the intensive and extensive margins. J Health Econ 1998; 17: 537–555.[CrossRef][Web of Science][Medline]

  5. Ellis RP, McGuire TG. Hospital response to prospective payment: moral hazard, selection, and practice style. J Health Econ 1996; 15: 257–277.[CrossRef][Web of Science][Medline]

  6. Feldman R, Dowd B. Is there a competitive market for hospital services? J Health Econ 1986; 5: 277–292.[CrossRef][Web of Science][Medline]

  7. Glazer J, McGuire TG. Optimal risk adjustment in markets with adverse selection: An application to managed care. Am Econ Rev 2000; 90: 1055–1071.

  8. Shen Y, Ellis RP. Cost-minimizing risk adjustment. Boston University working paper. Boston University, 1998.

  9. Cutler DM. Cost shifting or cost cutting? The incidence of reductions in Medicare payments. NBER/Tax Policy and the Economy 1998; 12: 1–27.

  10. Dranove D. Pricing by non-profit institutions. The case of hospital cost-shifting. J Health Econ 1998; 7: 47–57.[CrossRef]

  11. Ginsburg PB, Sloan FA. Hospital cost shifting. N Engl J Med 1984; 310: 893–898.[Medline]

  12. Sloan FA, Becker ER. Cross-subsidies and payment for hospital care. J Health Polit Policy Law 1984; 8: 660–685.

  13. Coulam RF, Gaumer GL. Medicare’s prospective payment system: a critical appraisal. Health Care Financ Rev 1991; Annual Supplement: 45-77.

  14. Gilman BH. Hospital response to Drg refinements: The impact of multiple reimbursement incentives on inpatient length of stay. Health Econ 2000; 9: 277–294.[Medline]

  15. Newhouse JP, Byrne DJ. Did Medicare’s prospective payment system cause length stay to fall? J Health Econ 1988; 7: 413–416.[CrossRef][Web of Science][Medline]

  16. Norton EC, Houtven CV, Lindrooth RC, Normand SL, Dickey B. Does prospective payment reduce inpatient length of stay? Health Econ 2002; 11: 377–387.[Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
17/5/421    most recent
mzi055v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Tsai, Y.-W.
Right arrow Articles by Chen, P.-F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tsai, Y.-W.
Right arrow Articles by Chen, P.-F.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?