Association of hospital ownership with patient transfers to
outpatient care under a prospective payment system in Taiwan
Herng-Ching Lin
a,∗, Sudha Xirasagar
b, Senyeong Kao
c,1 aTaipei Medical University, School of Health Care Administration, 250 Wu-Hsing St., Taipei, TaiwanbDepartment of Health Services Policy and Management, Arnold School of Public Health, University of South Carolina, Columbia, SC, USA cNational Defense Medical Center, School of Public Health, Taipei, Taiwan
Received 2 July 2003; received in revised form 8 November 2003; accepted 8 November 2003
Abstract
Case payment, a prospective payment system akin to diagnosis-related groups (DRGs) has in-built incentives for hospitals to transfer inpatients to their own ambulatory care units following early discharge. This study used nation-wide inpatient claims data on a total of 100,730 patients treated in 2000 in (Taiwan): cesarean section (59,364 cases), femoral/inguinal hernia operation (18,675 cases), and hemorrhoidectomy (22,691 cases), all reimbursed by case payment, to explore the relationship between hospital ownership and patient transfers to outpatient treatment. For all three diagnoses, for-profit (FP) hospitals not only had lower lengths of stay (LOS) compared to public hospitals, but also showed very high odds of patient transfer to their own outpatient units, after controlling for institutional variables, (hospital level, teaching status, and geographic location), hospital competitive environment (the Herfindal–Hirschman index), and patient variables (gender, age, length of stay, and number of secondary diagnoses, a proxy for severity of illness). Similar, though slightly lower odds were observed with not-for-profit (NFP) hospitals relative to public hospitals. The findings support the property rights theory, suggesting that in Taiwan, institutional profit maximization motives may be driving patient transfers under the case payment diagnoses, rather than medical care needs. In NFP hospitals, their physician compensation mechanism, driven largely by care volumes provided by each physician, appears to be driving the disproportionately greater likelihood of patient transfer to outpatient care.
© 2003 Elsevier Ireland Ltd. All rights reserved.
Keywords: Case payment; Patient transfer; Institutional ownership; Herfindal index
1. Introduction
Progressive introduction of the case payment sys-tem, a form of prospective payment system (PPS)
∗Corresponding author. Tel.:+886-22-7721226;
fax:+886-22-3789788.
E-mail addresses: henry11111@tmu.edu.tw (H.-C. Lin),
kao@ndmctsgh.edu.tw (S.Y. Kao).
1P.O. Box 90048-509, Taipei 114, Taiwan,
Tel.:+886-2-8792-4812; fax: +02-2378-9788.
in Taiwan, has reduced the cost of care for disease items brought under its ambit [1–3]. Case payment is similar to the diagnosis-related groups (DRGs) system in the US, providing for fixed reimbursement rates based on diagnostic codes rather than the actual costs incurred/charged by the provider for diagnosis and treatment. PPS allows hospitals to retain the dif-ference between the cost of care and the reimbursed rate, but inevitably, induces early, and sometimes premature discharge to maximize institutional profits 0168-8510/$ – see front matter © 2003 Elsevier Ireland Ltd. All rights reserved.
[4]. This is reflected in significantly reduced hospital lengths of stay (LOS) under PPS reimbursement as documented by several authors in Taiwan [5–7] and in other countries, chiefly the United States[8,9].
The case payment system in Taiwan has some checks to ensure quality of care and to prevent ad-ditional reimbursement for care provided in other care units. Taiwan’s BNHI specifies the minimal care requirements including standardized procedures for each diagnosis. To obtain full reimbursement, at least 65% of the minimal required items should have been accomplished. Case payment also specifies optional care items to be provided if needed without extra charge, which is also monitored. The patient’s health status at discharge must meet prescribed standards for the hospital to be reimbursed. Hospitals are not paid for unscheduled readmissions for the same disease within 14 days. This is to prevent premature dis-charge, known as the “quicker and sicker” syndrome in the US. Despite these caveats, cost reduction un-der case payment in Taiwan could be more apparent than real, given that the restrictions are limited to inpatient care. There are no restrictions on continu-ing outpatient care, which continue to be reimbursed on a cost-plus basis. Therefore, to accurately assess the impact of case payment on health care costs in Taiwan, it is essential to take into account patient transfers to outpatient care.
Many authors in the United States have documented the incentive to transfer patients to outpatient and other care units outside the ambit of PPS restrictions to com-pensate for diminished inpatient revenues under DRGs
[10–12]. Kominski and Biddle[9]also demonstrated the concurrent reduction in inpatient claims with an in-crease in outpatient claims, following implementation of PPS in the United States. Building on these findings from the United States, we hypothesized that in Tai-wan, for case payment diagnoses, for-profit (FP) hos-pitals would manifest significantly increased patient transfer to their ambulatory care units, compared to not-for-profits (NFP) and public hospitals. Under the assumptions of the property rights theory propounded by Furubotn and Pejovich[13], NFP and public insti-tutions lack the incentive to fulfill the objectives set out by their principals, and therefore, FP hospitals tend to outperform the NFP and public institutions. Since FP ownership allows managers and owners to bene-fit from the probene-fits, there is an incentive to maximize
profits. In public institutions, the benefits of efficiency and increased profits accrue to the public that owns the public institution, and therefore, the managers of these institutions have no incentive to exert unduly to maximize efficiency and profits [14]. Many authors have documented the differences in profit-maximizing behaviors of FP and NFP hospitals under PPS. Fried-man and Shortell [15] reported that investor-owned hospitals were more responsive to cost containment pressures under PPS in the United States. Sorrentino
[16] reported significant relationships between LOS and hospital ownership for seven DRGs. Younis et al.
[17]concluded that hospital ownership was a signif-icant determinant of hospital profitability under PPS. In Taiwan, Lin et al.[18]found that FP hospitals had shorter LOS compared to NFP under case payment. The average costs per discharge of uncomplicated thy-roidectomy and two other diagnoses covered by case payment were higher in public hospitals than in FP hospitals, with the reverse being true in respect of three diagnoses that were reimbursed by cost-plus reim-bursement[19]. Similarly, Yang reported that uncom-plicated appendectomy that is covered by cost-plus re-imbursement cost more in FP compared to public hos-pitals[20]. These studies support the assumptions of the property rights theory under PPS reimbursement conditions. While institutional differences in costs per discharge between the FP and public sector are docu-mented, to date, there is no documentation on patient transfers in Taiwan, which represent yet another form of profit maximization behavior under case payment pressures. We used nation-wide data on cesarean sec-tion, femoral/inguinal hernia operasec-tion, and hemor-rhoidectomy to explore the association between hos-pital ownership and inpatient to outpatient transfers under the case payment system.
2. Methods
2.1. Data sources and study sample
The data were sourced from the National Health Insurance Research Database (NHIRD) for the year 2000, covering all medical benefit claims for Taiwan’s population of over 23 million. The NHIRD provides one principal diagnosis (ICD-9-CM code) and up to four secondary diagnoses for each patient.
Patients admitted to hospitals between 1 January 2000 and 31 December 2000 with a diagnosis of 0371A (cesarean section), 0163A (hernia operation), and 0158A (hemorrhoidectomy alone or with anal fistulectomy) were included in the study. Cesarean section and hernia operation were selected since they were among the earliest procedures brought under case payment (1995 and 1997, respectively), and we expected that the institutional responses should have stabilized by 2000. Hemorrhoidectomy was selected due to high volume of the procedure throughout Tai-wan. All three diagnoses also satisfied the condition of relative homogeneity in clinical severity within each diagnosis.
Discharge status in the NHIRD is classified into recovered, transferred to outpatient care in the same hospital, dead, discharged at the patient’s request, and transferred to another hospital. Transfer to outpatient care does not include patients asked to come for rou-tine follow-up, suture removal or allied rourou-tine outpa-tient procedures. After excluding dead, discharged at patient’s request, and transferred to another hospital to their very low volumes, a total of 59,364 cesarean sec-tion cases, 18,675 hernia operasec-tion cases, and 22,691 hemorrhoidectomy cases were included in the study (total 100,730 patients).
2.2. Statistical analysis
Multiple logistic regression analysis was conducted separately for each diagnosis to determine the ad-justed odds of being discharged to the hospital’s out-patient unit relative to being classified as recovered. The key independent variable of interest was hospi-tal ownership, classified as FP, public, and NFP. The control variables were patient, hospital, and hospital competitive environment variables. Patient variables included age, LOS, (both continuous variables), gen-der and presence of co-morbidities (nil, one, two, or three-plus secondary diagnoses). The latter two were coded as categorical variables. Hospital variables in-cluded geographic location (north, central, south, and east), hospital level (medical centers with >500 beds, regional hospitals, 250–499 beds, district hospitals, 20–249 beds, and private ob-gyn clinics,<10 beds), and teaching status. The degree of hospital compe-tition is known to be an important predictor of hos-pital behavior in respect of cost reduction, pricing,
and offering advanced technology [21–23]. In this study, degree of hospital competition is measured by Herfindal–Hirschman index (HHI) s2i where si is the market share of total discharges of each hospital competing in a given healthcare market network, and the summation runs over all hospitals in the individ-ual healthcare network[24]. In Taiwan, the healthcare market is classified by the Department of Health into 17 healthcare networks, for which the Herfindal index was calculated and used as a control variable. Since information on clinical severity is not available, the number of secondary diagnoses was used a proxy, rep-resenting the potential medical need for outpatient care following surgery. LOS was used as another control variable to represent the dimension of clinical severity not captured by the secondary diagnoses, for exam-ple, anesthetic or post operative complications, tran-sient hypertension, etc. that may be resolved before discharge and therefore fail to qualify as a discharge diagnosis. A significance level ofP < 0.05 was used.
3. Results
3.1. Sample characteristics
Table 1 shows that 72.1, 92.8, and 91.8% of ce-sarean section, hernia operation, and hemorrhoidec-tomy cases, respectively, were transferred to outpatient treatment under the same hospital. NFP hospitals had the highest market share for all diagnoses, and public hospitals had the lowest market share. Hemorrhoids had the highest incidence of secondary diagnoses (69.7%) compared to cesarean section (47.7%), and 16% for the hernia operation group. The secondary diagnoses in the hemorrhoids group were mostly qualifying clauses about the hemorrhoid itself, rather than true complications with patho-physiological or LOS implications (39.7% thrombosed hemorrhoids, 20% with residual skin tags, 20% with both internal and external hemorrhoids, etc.) It is essential to bear this in mind when evaluating the study results. 3.2. Association between hospital ownership and patient transfer
Bivariate statistics show higher rates of patient transfer for hemorroidectomy and hernia operations
Table 1
Descriptive statistics of the three case payment diagnoses DRG classification
(DRG code)
Cesarean section (0371A), n= 59,364;
n (%) or mean (S.D.)
Hernia operation (0163A),
n= 18,675; n (%) or mean (S.D.) Hemorrhoidectomy (0158A), n= 22,691; n (%) or mean (S.D.) Discharge status Transfer to O-P 42,809 (72.1) 17,328 (92.8) 20,820 (91.8) Recovered 16,555 (27.9) 1,343 (7.2) 1,871 (8.2) Gender Male 16,176 (86.6) 13,507 (59.5) Female 2,495 (13.4) 9,184 (40.5) Hospital ownership Public 10,412 (17.5) 5,036 (27.0) 5,155 (22.7) NFP 25,934 (43.7) 7,779 (41.7) 9,509 (41.9) FP 23,018 (38.8) 5,856 (31.4) 8,027 (35.4) Hospital location Northern 26,535 (44.7) 8,402 (45.0) 11,515 (50.7) Central 11,943 (20.1) 4,688 (25.1) 4,364 (19.2) Southern 19,831 (33.4) 5,035 (27.0) 6,557 (28.9) Eastern 1,044 (1.8) 546 (2.9) 255 (1.1) Hospital level Medical center 13,883 (23.4) 5,752 (30.8) 7,006 (30.9) Regional hospital 20,172 (34.0) 6,926 (37.1) 8,146 (35.9) District hospital 24,287 (40.9) 5,993 (32.1) 7,539 (33.2) Clinic 1,022 (1.7) Teaching status Yes 40,862 (68.8) 14,400 (77.1) 17,172 (75.7) No 18,502 (31.2) 4,271 (22.9) 5,519 (24.3)
Number of secondary diagnoses
0 31,056 (52.3) 15,682 (84.0) 6,877 (30.3) 1 15,721 (26.5) 1,958 (10.5) 10,132 (44.7) 2 8,381 (14.1) 671 (3.6) 3,727 (16.4) >3 4,206 (7.1) 360 (1.9) 1,955 (8.6) Age (years) 30 (5) 46 (26) 44 (14) LOS (days) Public 5.60 (1.36) 2.79 (1.56) 3.15 (2.06) NFP 5.16 (1.32) 2.38 (1.32) 2.68 (1.75) FP 5.16 (1.27) 2.19 (1.54) 2.59 (1.98) Total 5.24 (1.32) 2.43 (1.48) 2.75 (1.92)
O-P: outpatient; LOS: length of stay; NFP: not-for-profit; FP: for-profit.
in FP hospitals compared to public hospitals. Patient transfer rates were similar in FP and NFP hospitals for hemorroidectomy (97.4% versus 97.8%, respec-tively), and for hernia operations (98.9% versus 97%). In case of cesarean section, transfer rates are highest in NFP hospitals followed by FP and public hospitals (Table 2). However, these crude rates need to be ad-justed for clinical severity (operationalized as number
of secondary diagnoses) to evaluate the association between patient transfer and hospital ownership.
Multiple regression analysis that controls for clin-ical severity besides other variables, clarifies the relationship between patient transfer and ownership, as shown inTable 3. Patients admitted to FP hospitals had greatly increased odds of being transferred to the hospital’s outpatient unit relative to public hospitals
Table 2
Bivariate associations between hospital ownership and patient transfer
Diagnosis Discharge status χ2-test
Recovery n (%) Transfer to outpatient unit n (%) Total
Cesarean section (0371A) 2988a
Public hospital 4,509 (43.3) 5,903 (56.7) 10,412
NFP hospital 4,467 (17.2) 21,467 (82.8) 25,934
FP hospital 7,579 (32.9) 15,439 (67.1) 23,018
Hernia operation (0163A) 1901a
Public hospital 1,042 (20.7) 3,994 (79.3) 5,036 NFP hospital 236 (3.0) 7,543 (97.0) 7,779 FP hospital 65 (1.1) 5,791 (98.9) 5,856 Hemorrhoidectomy (0158A) 3493a Public hospital 1,451 (28.1) 3,704 (71.9) 5,155 NFP hospital 244 (2.6) 9,265 (97.4) 9,509 FP hospital 176 (2.2) 7,851 (97.8) 8,027 NFP: not-for-profit; FP: for-profit. a P < 0.001.
(ORs = 5.11, 14.69, and 27.62, respectively, for
ce-sarean, hernia, and hemorrhoidectomy operations). NFP hospitals also had higher odds of patient transfer compared to public hospitals (ORs = 3.29, 10.82, 23.26).
Geographic location was significantly associated with patient transfer, with central Taiwan showing the highest odds relative to northern Taiwan. The Herfindal index had odds ratios of about 1 (not a sig-nificant predictor) for all three diagnoses. Teaching hospitals were more likely to transfer cases compared to non-teaching hospitals. Patient transfer was not related to LOS (OR = 1), although LOS is often an indirect indicator of severity of illness. This is possi-bly because of collinearity with the other variable for severity of illness, the number of secondary diagno-sis. Since the odds ratio for LOS was 1, we deleted LOS from the model.
The odds ratios for the number of secondary diag-noses represent a mixed bag. As discussed earlier, the predominance of trivial qualifications of the primary condition among the secondary diagnosis, particularly in respect of hemorrhoid cases clearly confounded our attempt to represent severity of illness through a count of the secondary diagnoses. Only the >3 secondary diagnoses group shows higher odds of pa-tient transfer relative to the “no secondary diagnoses” group for cesarean section and hernia cases. For hemorrhoidectomy cases, the odds ratios show that the “no secondary diagnosis” group had higher odds
of patient transfer relative to those with any addi-tional diagnosis. These findings probably reflect the inadequacy of this conceptualization of severity of illness.
4. Discussion
This study supports the hypothesis that hospital ownership, particularly FP ownership, and to a lesser extent, NFP status significantly predicts patient trans-fer to a hospital’s outpatient unit. In the hernia and hemorroidectomy groups, the effect size of ownership (magnitude of odds ratios) is the highest compared to any other institutional or demographic variable eval-uated (as shown inTable 3), indicating that of all the variables, ownership is perhaps the strongest predic-tor of patient transfer in Taiwan. Even in respect of cesarean section, it is almost the leading predictor of patient transfer.
The way the PPS is presently designed, there are fi-nancial incentives to minimize inpatient care and use the route of patient transfer to cushion the effects of PPS on hospital profits. LOS statistics are also con-sistent with this explanation. While the average LOS for hernia operations was 2.43 days, it was 2.79, 2.38, and 2.19, respectively, for public, NFP, and FP hospi-tals; similarly, average LOS were 3.15, 2.68, and 2.59 days for hemorrhoidectomy cases, and 5.60, 5.16 and 5.16 days, respectively, for cesarean deliveries.
Table 3
Adjusted relationship between hospital ownership and patient transfer
Variable Cesarean section
OR (95% CI) Hernia operation OR (95% CI) Hemorrhoidectomy OR (95% CI) Hospital ownership
Public hospital (reference group)
NFP 3.29 (3.11–3.47) 10.82 (9.18–12.75) 23.26 (17.74–30.51)
FP 5.11 (4.76–5.49) 14.69 (10.83–19.91) 27.62 (23.54–32.41)
Hospital location North (reference group)
Center 2.08 (1.96–2.21) 4.09 (3.16–5.31) 6.98 (5.44–8.96) South 1.49 (1.42–1.57) 1.86 (1.60–2.17) 3.78 (3.19–4.48) East 0.91 (0.76–1.09) 0.58 (0.43–0.79) 0.38 (0.25–0.58) Hospital level Medical center 2.05 (1.90–2.22) 0.15 (0.11–0.20) 0.09 (0.07–0.13) Regional hospital 1.16 (1.09–1.24) 0.41 (0.31–0.54) 0.97 (0.74–1.28) District hospital (reference group)
Clinic 2.39 (2.09–2.75)
Teaching status
Yes 5.28 (4.94–5.64) 5.51 (4.11–7.38) 11.18 (8.19–15.26)
Number of secondary diagnosis 0 (reference group) 1 1.29 (1.23–1.35) 0.98 (0.81–1.17) 0.51 (0.43–0.59) 2 1.33 (1.25–1.42) 0.75 (0.58–0.96) 0.53 (0.44–0.63) 3 1.40 (1.28–1.53) 1.32 (1.11–1.82) 0.49 (0.39–0.63) Gender Male 0.99 (0.81–1.22) 1.18 (1.05–1.34)
Female (reference group)
HHI 1.00 (1.00–1.01) 0.99 (0.99–1.00) 0.99 (0.99–1.00)
Age 0.99 (0.99–1.00) 0.99 (0.99–1.00) 0.99 (0.99–0.99)
Constant 0.10 5.64 1.01
n 59,364 18,675 22,691
2 8679a 2386a 5130a
OR: odds ratio; NFP: not-for-profit; FP: for-profit; HHI: Herfindal–Hirschman index.
aP < 001.
The index of hospital competitive environment, the Herfindal index, which is documented as a ma-jor predictor of institutional behavior, failed to show any significance in our study (OR = 1 for all three diagnoses). This could be due to two reasons. We con-trolled for geographic location, which itself serves as a proxy for density of hospital bed capacity. Therefore, the Herfindal index could have failed to show signif-icance due to collinearity with region specification. Second, it could be that the Herfindal index is highly correlated with FP and NFP ownership. Reasonably, it could be expected that regions with high density of hospital bed capacity also have a high density of FP and NFP beds. With the ownership effect on patient
transfer overshadowing all other institutional effects (highest magnitude of ORs), our finding of nil effect of competitive environment is reasonable.
Our findings on patient transfer in FP hospitals are consistent with the studies of Bellandi et al.[25], You-nis et al.[26], and Shukla et al.[27], essentially pro-viding empirical support for the property rights theory. The property rights theory suggests that FP hospitals have a greater financial motive to maximize profits than non-profit hospitals [14]. The fixed reimburse-ment rates for inpatient care without concomitant re-strictions on outpatient care carries a built-in incentive to generate more outpatient visits to maximize rev-enues, a response that is more likely in the FP sector.
Our study shows that FP hospitals have very high odds of patient transfer relative to public hospitals. But the NFPs are not far behind. This can be explained by the hospital ownership situation and physician compen-sation system in Taiwan. Public hospitals are owned and financed by either the Department of Health of Taiwan, or by the county and city governments. After realizing the NHI reimbursement, any residual budget deficit is made up by the government. Physicians are compensated with a base salary, with some supple-mentation from volume-driven NHI revenues accruing to the hospital. The bonus is not directly proportional to the patient revenue generated by each physician. By law, FP hospitals can be owned only by physicians. There are no corporate hospitals in Taiwan. Of total 637 hospitals in Taiwan, 419 are physician-owned FP hospitals. In these hospitals, profits and physician income are solely driven by reimbursements realized from NHI and from out-of-pocket payments by pa-tients. NFPs are affiliated with non-profit religious organizations or other non-profit institutions. They are exempt from taxes, receive donations from phi-lanthropists, and often, lump sum grants from govern-ment to continue to serve in under-served areas. Most NFPs compensate their physicians in proportion to the volume-driven revenues generated by each physician. Our findings of highest odds of patient transfer by FP hospitals, followed closely by NFPs is in line with the ownership and physician compensation system, assuming the dynamics of the property rights theory. Although the NFP hospital managements have no in-centive to maximize profits, physicians have an incen-tive (allied with the assumptions of the property rights theory) to transfer patients. Our finding of slightly lower odds in NFP hospitals compared to FP hospitals may be reflecting the higher percentage of revenues paid to physicians compared to the NFP sector, a stance that is possible due to their relative financial advantages, such as direct financial support from the government, and exemption from property and insti-tutional income taxes. On the contrary, physicians in public hospitals, mostly compensated with high base salaries, have little individual incentive to provide the additional clinical effort to provide outpatient care.
Apart from financial factors, another factor in pa-tient transfers could be the increasing anxiety about malpractice litigation in Taiwan. Physician uncertainty about health status at discharge following a short LOS
(which is also significantly shorter in FP hospitals) may also be contributing to patient transfer. The num-ber of malpractice lawsuits increased by 58.5% since 1995 when National Health Insurance was initiated (Department of Health, 2002).
Although past studies have documented patient transfers from inpatient to outpatient units as one institutional response to the implementation of a PPS[9,12], the differences in magnitude of the phe-nomenon by ownership have not been explored. Our study shows that over 97% of patients undergoing a hernia operation or hemorrhoidectomy in FP and NFP hospitals were transferred to outpatient care. The dis-proportionate transfer rate at FP and NFP hospitals is similar to DRG creep, a phenomenon documented by several authors in the United States[28,29]. DRG creep is an administrative syndrome where hospitals manipulate the diagnostic information in order to clas-sify patients into DRGs that are associated with higher rates and thereby obtain higher revenue[30]. We are not aware of the existence of DRG creep in Taiwan, but the data show the existence of patient transfer as one institutional response to the revenue-restricting effects of the case payment system.
Currently, it is not possible to assess the dollar value of the cost impact, as this would require patient care costs to be tracked through all levels of care. The NHIRD does not release patient identifier information on grounds of patient confidentiality issues, nor does it provide comprehensive, linked data on patients over time and across claims databases. Outpatient care, in-patient care, and other levels of care are documented in unlinked databases.
Despite the above limitations, our study findings have important policy and research implications. The magnitude of patient transfer, our earlier documen-tation about costs per discharge, and documendocumen-tation by other authors emphasize the need for the NHIRD to move towards providing patient-level data linked across databases to researchers. Until then, past stud-ies showing cost reduction for case payment items may be causing a false sense of complacence that case payment is the panacea to the problem of surg-ing healthcare costs. From a policy perspective, our study suggests that case payment diagnoses patients treated in the FP and NFP sectors are possibly incur-ring greater healthcare costs than in the public sector. However, our study does not take into account the
budget supplementation by governments to the public hospitals, and the costs of foregone taxes on account of tax exemptions to the NFP sector. Considering the sums of money involved, the government should sponsor detailed costing studies to establish the real costs and efficiencies of patient care in the FP, public, and NFP sector, taking into account all inputs (such as public sector salaries and tax exemptions). Pending that, it is difficult for us to recommend any long-term policy initiatives to address the question of patient transfer in the private healthcare sector.
Acknowledgements
This study is based in part on data from the Na-tional Health Insurance Research Database provided by the Bureau of National Health Insurance, Depart-ment of Health and managed by the National Health Research Institutes. The interpretations and conclu-sions contained herein do not represent those of the Bureau of National Health Insurance, Department of Health or the National Health Research Institutes.
References
[1] Yang CF. The effect of the implementation of case payment system on expenditure and quality of care—a case study of coronary artery bypass graft surgery. M.A. Thesis. Kaohsiung (Taiwan): Kaohsiung Medical University; 2002.
[2] Tseng IL. The study of cesarean section utilization under case payment policy of the Nation Health Insurance. M.A. Thesis. Taoyuan (Taiwan): Chang Gung University; 1998. [3] Hsueh SC. The impact of fee-for-service and case payment
on cost and quality of care-example of total hip replacement. M.A. Thesis. Kaohsiung (Taiwan): Kaohsiung Medical University; 1998.
[4] Romeis JC. US nursing-home population composition in the PPS-era. Health Serv Mgmt Res 1994;7:111–9.
[5] Lin YR. Impacts of case payment on health provider’s behavior: a case study of transurethral prostatectomy in a regional hospital. M.A. Thesis. Taipei (Taiwan): Yang-Ming University; 1998.
[6] Huang SH. The impact of case payment system on health resources utilization—a case study of open reduction of close fracture in a regional hospital. M.A. Thesis. Taichung (Taiwan): Chinese Medical College; 2000.
[7] Chang SJ. The effects of implementation of case payment on the use of health resources—taking appendectomy as an example. M.A. Thesis. Kaoushung (Taiwan): Kaoushung Medical University; 1988.
[8] Lave JR, Frank RG. Hospital supply response to prospective payment as measured by length of stay. Adv Health Econ Health Serv Res 1990;11:1–25.
[9] Kominski GF, Biddle AK. Changes in follow-up care for Medicare surgical patients under the prospective payment system. Med Care 1993;31:230–46.
[10] Freiman MP, Sederer LI. Transfers of hospitalized psychiatric patients under Medicare’s prospective payment system. Am J Psychiatry 1990;147:100–5.
[11] Morrisey MA, Sloan FA, Valvona J. Medicare prospective payment and posthospital transfers to subacute care. Med Care 1988;26:685–98.
[12] Trinh HQ, Beguan JW. Hospital response to the implementation of prospective payment. Health Serv Mgmt Res 1988;11:163–73.
[13] Furubotn E, Pejovich S. Property rights and economic theory: a survey of recent literature. J Econ Lit 1972;10:1137–62. [14] Menke TJ. The effect of chain membership on hospital costs.
Health Serv Res 1997;32:177–96.
[15] Friedman B, Shortell S. The financial performance of selected investor-owned and not-for-profit system hospitals before and after Medicare prospective payment. Health Serv Res 1988;23:237–67.
[16] Sorrentino EA. Hospitals vary by LOS, charges, reimburse-ments and death rates. Nurs Mgmt 1989;20:54–6, 58, 60. [17] Younis M, Rice J, Barkoulas J. An empirical investigation
of hospital profitability in the post-PPS era. J Health Care Finance 2001;28:65–73.
[18] Lin HC, Chang HC, Lee SC, Yang CS, Tung YC. Comparisons of LOS for patients undergoing a transurethral prostatectomy (TURP) by hospital level, hospital ownership and hospital location under the case payment system in Taiwan. N Taipei J Med 2002;4:25–33.
[19] Lin HC, Yang CS, Tung YC, Chang WY. Comparing costs and hospital ownership for thyroidectomy patients. N Taipei J Med 2002;4:99–107.
[20] Yang YC. The effect of hospital payment system on the inpatient medical expenditures. M.A. Thesis. Taichung (Taiwan): Feng Chia University; 2000.
[21] Robinson JC, Luft HS. Competition and the cost of hospital care, 1972 to 1982. JAMA 1987;257:3241–5.
[22] Melnick GA, Zwanziger J. Hospital behavior under competition and cost-containment policies. The California experience, 1980 to 1985. JAMA 1988;260:2669–75. [23] Luft HS, Robinson JC, Garnick DW, Maerki SC, McPhee
SJ. The role of specialized clinical services in competition among hospitals. Inquiry 1986;23:83–94.
[24] Keeler EB, Melnick G, Zwanziger J. The changing effects of competition on non-profit and for-profit hospital pricing behavior. J Health Econ 1999;18:69–86.
[25] Bellandi D, Kirchheimer B, Saphir A. Profitability a matter of ownership status. For-profit systems see earnings rise, while not-for-profits lag. Mod Healthc 2000;30:24–6.
[26] Younis M, Rice J, Barkoulas J. An empirical investigation of hospital profitability in the post-PPS era. J Health Care Finance 2001;28:65–73.
[27] Shukla RK, Pestian J, Clement J. A comparative analysis of revenue and cost-management strategies of not-for-profit and for-profit hospitals. Hosp Health Serv Adm 1997;42: 117–34.
[28] Assaf AR, Lapane KL, McKenney JL. Possible influence of the prospective payment system on the assignment of
discharge diagnosis for coronary heart disease. N Engl J Med 1993;329:31–6.
[29] Roger FH. DRGs and documentation: the information problem. Health Policy 1988;9:193–204.
[30] Simborg DW. DRG creep: a new hospital-acquired disease. N Engl J Med 1981;304:1602–4.
