醫院競爭對醫療成本和品質的影響

(1)

行政院國家科學委員會專題研究計畫成果報告

醫院競爭對醫療成本和品質的影響

計畫類別：個別型計畫

計畫編號： NSC94-2415-H-004-013-

執行期間： 94 年 08 月 01 日至 95 年 07 月 31 日

執行單位：國立政治大學財政系

計畫主持人：連賢明

報告類型：精簡報告

報告附件：出席國際會議研究心得報告及發表論文

處理方式：本計畫可公開查詢

中華民國 95 年 10 月 24 日

(2)

摘要本篇文章檢驗了在台灣，中風患者醫療費用與品質是否受到市場競爭的影響。對於醫療費用的衡量，我們有短期以及長期兩種方式，短期指該次住院所花費用，而長期指出院後一年內所花費的門診及住院費用；至於市場競爭程度的衡量，則採取傳統的 “變動半徑法” 賀氏指標，並以 75 百分比作為衡量基準。本研究選取的樣本為 1997~2001 年之間，在台灣新的住院中風病人。研究發現，市場競爭對醫療品質的影響，不論是以一個月或一年的死亡率來看，都不顯著；而在醫療費用方面，結果顯示當醫院所在市場的競爭度愈高，其所花費的住院醫療費用也愈高，但在長期的醫療費用上，影響程度較小，在統計上甚至是不顯著的。另一個我們的研究結果，與 MAR 的假設相反， MAR 的假設指出，醫院會透過提供一些不必要的醫療檢查來競爭，但我們的研究發現，醫院並不是透過昂貴醫療儀器的使用，如電腦斷層掃描 (CT) 或核磁共振 (NRI) ，來增加住院的醫療費用，而是導因於病人的住院天數增加。關鍵字: 競爭，醫院，全民健保

(3)

Abstract

This paper examines whether market competition affects treatment expenditure and health outcome of stroke treatment in Taiwan. Our measure of treatment expenditure is the hospital expenditure paid at the index admission (short-term) and the sum of inpatient and outpatient expenditures paid in the subsequent year (long-term). Our measure of health outcome is the

probability of death in 1 and 12 months after the hospital’s discharge. Our measure of competition is the conventional “variable radius” HHI index calculated at 75 percentile. Using data of patients hospitalized for new stroke treatment between 1997 and 2001 in Taiwan, we find that market competition results in insignificant impacts on health outcome, either measured by 1 or 12 months survival rates. In terms of treatment expenditure, our results indicate that hospitals facing more competition incur higher inpatient expenditures, but the increase is smaller or even insignificant for the long-term expenditure. Contrary to the MAR hypothesis arguing hospitals compete through the provision of inefficient medical technology, we find the higher inpatient expenditure is not driven by the increase of treatment intensity or the use of expensive medical technology such as CT or MRI, but by the increase in the patient’s length of stay.

(4)

I. Introduction

In the last two decades, the landscape of hospital industry worldwide has undergone a dramatic transformation. In the United States, the number of acute hospital beds fell by 12% between 1988 and 1998, of which 40% was due to the closure of 496 general hospitals over the sample period (Lindrooth, et. al., 2003). In Canada, more than 275 hospitals have been closed, merged, or converted to other types of facilities between 1996 and 2001 (CIHI, 2001). In Taiwan, about one quarter of the total seven hundred acute hospitals exited the market in the past ten years, despite the fact that the total number of hospital beds during the same period grows more than 40%.

The dramatic change in the hospital market through exits, ownership conversion, or

consolidations has led substantial research interests over the competitiveness of this industry (For comprehensive reviews, see Gaynor and Vogt, 2000; Dranove and White, 1994 and Dranove and Satterthwaite, 2000). On the one hand, there have been findings indicating competition among hospitals reduces treatment cost and improves service quality (Dranove et. al., 1992, 1993; Town and Vistnes, 2001). On the other hand have been studies arguing that the insensitivity to price due to health insurance has led hospitals to engage a “medical arm race“ (MAR) that compete through the provision of unnecessary care (Robison and Luft, 1985; 1987). Other studies have emphasized that the

informational imperfection in hospital markets leads to higher costs and excess capacities in face of an increase in the number of health providers (Frech, 1996; Fisher et. al, 1999). Clearly, the existing literature has no consensus as to whether hospital competition enhances social welfare.

Why are implications of hospital competition so mixed? In styled markets of basic economic models, such a debate could be solved by examining whether the cost synergy is able to offset the price hike of the supplied services. In the hospital industry, however, applying this standard is difficult due to several limitations. Most problematic is the lack of strong price competition due to the presence of insurance and information asymmetry in the healthcare market. Moreover, treatment outcome is not only difficult to measure, but difficult to define per se. Without reliable measures of treatment cost and outcomes, as pointed out by Kessler and McClellan (2000), “there are virtually no previous research has identified these effects on both health care costs and patient’s health outcomes.”

(5)

payers; the presence of multiple payers increases the difficulty to collect price information since private payers often negotiate their payments privately with hospitals. Consequently, results of most competition studies are derived based on data of one large public payer (e.g.: Medicare or Medicaid). Nonetheless, there have been studies indicating that competition produces different effects for different payers, especially when private payers were involved (Gowrisankaran and Town, 2003)1; findings based on one payer may be difficult to generalize. Moreover, hospitals may shift their costs toward other payers in face of the price reduction of one payer (e.g. Dranove, 1988)2. In that case, the true impact of hospital competition is likely to be contaminated by the cost-shifting behaviors of hospitals.

The second problem relates to the payment system in the states. Beginning in 1987, following the practice of Medicare, almost all payers have replaced cost-based payment with prospective payment system (PPS). Although PPS is effective in containing the rising hospital expenditure, it also increases the extent of risk selection since hospitals are paid for each patient a fixed amount based on the patient’s diagnosis, not by the hospital’s cost of treatment. This is especially prevalent for hospitals under severe competition pressure (Meltzer et al, 2002);3 the strong incentive for hospitals to select healthy patients at admissions complicates the analysis, and distorts findings of hospital competition.

In this paper we analyze the impact of competition among hospitals on the treatment expenditure and health outcome of stroke treatment in Taiwan. We use Taiwan data because Taiwan implemented National Health Insurance (NHI) in 1995, the latest developed country to offer insurance coverage. Because NHI employs the single payer, and pays hospitals on the fee-for-service basis, the data provide an excellent opportunity to investigate the impact of hospital competition. We analyze the stroke disease---hemorrhage strokes as well as ischemic strokes---in part because of its importance and cost----stroke is a leading cause for hospital admission and inpatient expenditures in Taiwan4----but

1

Gowrisankaran and Town (2003) found that competition for Medicare and HMO patients may have different impacts on hospital quality: the former reduced the welfare but the latter improves the welfare.

2

Dranove (1988) found out that hospitals raised their prices for privately insured patients in response to the reduction of Medicare or Medicaid payments.

3

Meltzer et al. (2002) found that increasing competition in the context of prospective payment is associated with selective reductions of expenditures for the most expensive patients.

4

The inpatient expenditure for cerebrovascular disease, namely stroke, is 4.78% in 2003, next to heart disease (7.18%) and cancers (12.40%).

(6)

also because the severity of the illness is unlikely to be known by an individual prior to its occurrence. Therefore, the distribution of disease severity is less likely to be correlated with an individual’s residence location. As we seen below, this feature is important in defining the competition measure used in the analysis.

We use longitudinal hospital claims of NHI receipts diagnosed as new stroke cases during 1997 and 2001. Since NHI covers almost all stroke related treatment, we measure the health expenditure not only by the payment paid during the treatment episode (short-term), but also the total expenditure (inpatient and outpatient) paid in the year following the health shock (long-term). Because stroke is the second leading cause of death in Taiwan, we measure the short and long-term health outcome using the incidence of death within 1 and 12 months after admission. For competition measure, we use the conventional “variable radius” HHI index calculated using the standard of 75 percentile. To account for differences of patient’s severity mix at admission and hospital’s quality of care, the estimation not only includes a detailed set of patient’s and hospital’s characteristics, but also employs the hospital as well as patient’s residence zip-code fixed effect to capture some unobserved quality differences (e.g. reputation). As a result, our identification comes from variations of HHI index for a given hospital over years.

Our findings indicate that higher market concentration (or lower hospital competition) results in higher short-term treatment expenditures, but has no consequential impact on patient’s mortality rate, either short or long-term. Although our results to some extent are consistent with predictions of MAR hypothesis, we do not find direct evidence showing that higher market concentration increases the inpatient treatment intensity or use of expensive medical technology such as computer tomography (CT) or magnetic resonance imaging (MRI). Instead, the higher hospital expenditure is driven by the increase of length of stay. Because post-stroke patients usually suffer from disability, immediate therapies are necessary for them to restore their functional status (e.g. motor, language abilities). Stroke patients may prefer a longer hospital stay to reduce the subsequent burden of nursing on themselves and family members and to continue their rehabilitations in hospitals rather than through outpatient visits. We suspect hospitals facing higher competition pressure are more likely to meet this specific demand of stroke patients, resulting a higher length of stay.

(7)

This article is organized as follows. We begin in Section 2 with a discussion on previous literature on hospital competition. Section 3 briefly introduces NHI and the hospital market in Taiwan. Section 4 describes the data and sample we analyze, and Section 5 shows the econometric strategy we adopt. In Section 6, we present basic results, robustness checks and possible mechanisms affecting treatment expenditure and health outcome. Section 7 concludes.

II. Previous Literature on Hospital Competition

As compared to the competitive markets described in the industrial literature, healthcare markets usually differ in a number of important aspects: products are differentiated, information is imperfect, government regulation is extensive and services are provided by non-profit organizations (Gaynor and Vogt, 2000). Not surprisingly, in face of various complications, the impact of competition in the healthcare market may not conform to the conventional wisdom that competition leads to lower prices and higher consumer welfare.

The theoretical justification for the way in which hospital competition may reduce social welfare focuses on the absence of price competition. As a direct result of insurance, the full burden of

healthcare expenses is not borne by consumers, a factor that diminishes the importance of price shopping. In addition, if hospitals were reimbursed on a “cost-plus” basis, they did not bear the

marginal costs of intensive treatment decisions. Therefore, hospitals compete to provide excessive care or expensive medical technology, adding substantially to overall healthcare costs. Furthermore, the provision of medical service may fail to take advantage of scale and learning effects, leading to worsening outcomes. In light of these adverse effects, the literature refers this type of competition as MAR since competition results in social waste (Robinson and Luft 1985; 1987).

However, the emergence of managed care such as health managed organization (HMO) or preferred provider organization (PPO) after the mid-1980s has changed the nature of hospital competition. HMO uses financial incentives to hospitals to limit utilizations. More importantly, managed care induces hospitals through selective contracting to engage in price competition (Dranove et al., 1993; Town. and Vistnes, 2001). Although the extent of price competition still depends on the information imperfections in the hospital market as well as the profit-maximizing incentives of hospitals, there are no doubts that managed care has substantially increased the importance of price

(8)

competition in the healthcare marketplace.5 As a consequence, the welfare implications of hospital competition now become ambiguous.

Since the theoretical literature does not offer a clear picture on the effect of hospital competition, the consequences of hospital competition need empirical justifications. A vast empirical literature has examined the consequences of competition in markets for hospital services [For comprehensive reviews, see Gaynor and Vogt (2000); Dranove and White (1994); Dranove and Satterthwaite (2000)]. In short, empirical research using the U. S. data prior to the mid-1980s found results that were more consistent with the predictions from the MAR hypothesis (e.g. Joskow, 1980; Robinson and Luft, 1985, 1987; Robinson, 1988; Robinson et. al., 1988; and Noether, 1988) 6. However, studies using more recent data generally found that hospital competition results in effects consistent with the traditional views described in the textbooks (e.g. Gruber, 1994; Dranove, Shanley and White, 1993; Kessler and McClellan, 2000). By comparisons, fewer attempts were made to examine the consequence of health outcome, of which the findings between health outcomes and market competition were often mixed (e.g. Shortell and Hughes, 1988; Propper et. al. 2004).

One major shortcoming in the previous research, as noted by Kessler and McClellan (2000), is that these studies identified the effect of market competition on treatment cost and health outcome

separately. Without examining costs and benefits jointly, it is impossible to evaluate the welfare effect of market competition. One exception is Kessler and McClellan (2000) that found that prior to 1991 competition had led to higher costs and lower rates of adverse health outcomes for elderly heart disease patients, but after 1991 competition had led both to substantially lower costs and reduced rates of adverse outcomes. In other words, market competition unambiguously improves social welfare, at least in years after 1991. In the same spirit, this analysis seeks to examine the relationship between

5

Dranove and Satterthwaite (1992; 2000) showed that the effect of competition on quality and price depends on the effect of competition on the price and quality elasticity of demand. When the information is imperfect, if competition increases the price elasticity alone, then market competition will decrease quality. If competition increases both elasticities of price and quality, then the effect on quality is ambiguous, depending on the relative magnitudes of elasticities.

6

Joskow (1980) found that higher market concentration leads to a lower reservation quality in hospitals, whilst studies based on 1972 data (Robinson and Luft, 1985) and 1982 data (Robinson and Luft, 1987), both found that average costs per admission, and costs per patient day, were substantially higher in those hospitals operating in more competitive markets. In addition, earlier studies had also revealed that hospitals located in more

competitive markets were more likely to offer percutaneous transluminal coronary angioplasty (PTCA) and coronary-artery bypass surgery (CABG) (Robinson, Garnick and McPhee, 1987) as well as employ a substantially higher number of employees (Robinson, 1988).

(9)

market competition and both treatment expenditures and health outcomes.

VII Discussions

This paper investigates the effect of hospital competition on treatment expenditure and health outcome for stroke treatment. Using the NHI enrollees hospitalized for stroke treatment between 1997 and 2001 in Taiwan, we examine whether hospitals facing different market concentrations incur different health expenditure or mortalities after treatment. Our results indicate that market

concentration does not result in a significant effect on patient’s mortality rate, either measured by 1 and 12 month death rate. For treatment expenditure, our findings suggest that higher market concentration significantly lowers the health expenditure at the index admission, but the effect is smaller or even insignificant when accounting for the total health expenditure in the subsequent year. Moreover, we investigate the sources causing differences in health expenditure. Our findings show that the main difference comes from differences in length of stay, rather than differences in treatment intensity or use of expensive diagnostic equipment such as CT or MRI.

Our findings differ from previous studies on two aspects. First, though the finding in this paper confirms that higher concentration results in higher expenditure, the utilization and expenses on expensive medical technologies are not found to be significantly different across market concentration. In other words, our evidence here does not support the MAR type of argument. Second, a number of studies have reported that the implementation of NHI has encouraged the adoption and utilization of expensive medical technology in Taiwan, of which one likely reason is the quality competition among hospitals (e.g. Chou et al., 2004; Tsai and Li, 2002). While the universal health insurance could still contribute the adoption of new technology, our results do not support the claim of hospital

competition.

We now discuss several limitations of our study. First, the study employs the conventional “variable-radius” competition measure. Although this is probably the most widely used competition measure, the measure is likely to suffer from endogeneity bias despite controlling for hospital fixed effect. Recent studies (e.g. Kessler and McClellan, 2000; Town and Vistnes, 2001; Gowrisankaran and Town, 2003) have proposed to use the predicted than actual patient flows as the competition measure to solve the problem of endogeneity. Unfortunately, due to the data constraint, this study cannot

(10)

conduct the measure based on predicted patient flow. We caution readers this potential bias. Second, our focus of this analysis is stroke treatment. Comparing with other inpatient services, advances in health technology for stroke treatment have been relatively slow in recent years; results of hospital competition may be different when applying the analysis to other treatments, particularly the ones experiencing substantial technological progress (e.g. AMI). Finally, and most importantly, we measure a patient’s outcome by his mortality rate, not accounting for his or her functional status. For post-stroke patients, a valuation that incorporates the patient’s mortality and functional status may be a better outcome measure (e.g.: quality of adjusted year of life).7 A more careful analysis that combines the patient’s functional status may be necessary to evaluate the effect of hospital competition on social welfare.

The main contribution of this paper is to add to the sparse literature providing empirical evidence as to how hospital competition affects treatment expenditure and health outcome in developing countries. Although a significant amount of attention has been paid over the past decade to the consequences of competition, the majority of them focus on the healthcare market in the U. S. that is difficult to generalize to health systems in other countries. By comparisons, our analysis uses Taiwan data, and investigates the short and long-term consequences of hospital competition. Our finding should be more applicable to other countries seeking to conduct health reforms through the introduction of market forces.

7

Muennig et al (2001) have found that the additional quality-adjusted years of life and absolute years of life are very different if strokes were eliminated.

(11)

References

Canada Institute for Health Information (2001) Health Care in Canada 2001

Baker, L. C. (2001) “Measuring Competition in Health Care Markets.” Health Services Research 36 (1): 223-51.

CEPD (1990) The Report of National Health Insurance Plan, CEPD.

Cheng, T. M.(2003) "Taiwan's new national health insurance program: Genesis and experience so far,"

Health Affairs 22 (3): 61-76.

Chou, S.-Y., J.-T. Liu, and J. K. Hammitt. (2004). “National Health Insurance and Technology Adoption: Evidence from Taiwan.” Contemporary Economic Policy 22 (1): 26-38.

Dranove, D. (1988). “Pricing by Non-profit Institutions: The Case of Hospital Cost-Shiftin g.” Journal

of Health Economics 7 (1): 47-57.

Dranove, D. and M. A. Satterthwaite. (1992) “Monopolistic Competition When Price and Quality Are Imperfectly Observable.” RAND Journal of Economics 23 (4): 518-34.

Dranove, D. and M. Satterthwaite. (2000). “The Industrial Organization of Health Care Markets.” In

Handbook of Health Economics, edited by A. J. Culyer and J. P. Newhouse, pp. 1094-139.

Dranove, D., M. Shanley, and C. Simon. (1992) “Is Hospital Competition Wasteful?” RAND Journal

of Economics 23 (2): 247-62.

Dranove, D., M. Shanley, and W. D. White. (1993) “Price and Concentration in Hospital Markets: The Switch from Patient-Driven to Payer-Driven Competition.” Journal of Law and Economics 36 (1): 179-204.

Dranove, D. and W. D. White. (1994) “Recent Theory and Evidence on Competition in Hospital Markets.” Journal of Economics and Management Strategy 3 (1): 169-209.

Fisher, E. S., Wennberg, J. E., Stukel, T. A., Skinner, J. S., Sharp, S. M., Freeman, J. L .and Gittelsohn, A. M. (2000) “Associations Among Hospital Capacity, Utilization, and Mortality of U.S.

Medicare Beneficiaries – Controlling for Sociodemographic Factors,” Health Services Research. 34 (6): 1351–1362

Frech, H.E. III (1996) Competition and Monopoly in Health Care (Washington, DC: AEI Press,. Garnick, D. W., H. S. Luft, J. C. Robinson, and J. Tetreault. (1987) “Appropriate measures of hospital

market areas.” Health Service Research 22 (1): 69-89.

Gaynor, M. and W. B. Vogt. (2000) “Antirust and Competition in Health Care Markets.” In Handbook

of Health Economics, edited by A. J. Culyer and J. P. Newhouse, pp. 1405-87.

(12)

for Ischemic Stroke Outcome Studies," Stroke. 35 (8): 1941-5.

Gowrisankaran, G. and R. J. Town. (2003) “Competition, Payers, and Hospital Quality.” Health

Services Research 38 (6): 1403-21.

Gresenz, C. R., J. Rogowski, and J. J. Escarce. (2004) “Updated Variable-Radius Measures of Hospital Competition.” Health Services Research 39 (2): 417-30.

Gruber, J. (1994) “The Effect of Competitive Pressure on Charity: Hospital Responses to Price Shopping in California.” Journal of Health Economics 13 (2): 183-211.

Halm, E. A., C. Lee, and M. R. Chassin. (2002) “Is volume related to outcome in health care? A systematic review and methodological critique of the literature.” Annuals of Internal Medicine 137 (6): 511-20.

Joskow, P. L. (1980) “The Effects of Competition and Regulation on Hospital Bed Supply and the Reservation Quality of the Hospital.” Bell Journal of Economics 11 (2): 421-47.

Kessler, D. P. and M. B. McClellan. (2000) “Is Hospital Competition Socially Wasteful?” Quarterly

Journal of Economics 115 (2): 577-615.

Lien, H., S.-Y. Chou, and J.-T. Liu (2005) “Hospital Ownership under Universal Health Insurance: Evidence from Taiwan”. Working Paper.

Lindrooth, R. C., Lo Sasso A. T., and Bazzoli, G. T. (2003) “The Effect of Urban Hospital Closure on Markets”, Journal of Health Economics 22: 691-712

Lu, J. R. and Hsiao, W. C. (2003). "Does universal health insurance make health care unaffordable? Lessons from Taiwan," Health Affairs 22 (3): 77- 88.

Meltzer, D., J. Chung, and A. Basu. (2002) “Does Competition under Medicare Prospective Payment Selectively Reduce Expenditures on High-Cost Patients?” RAND Journal of Economics

33(3):447-68.

Muennig, P. A. and M. R. Gold. (2001) “Using the years-of-healthy-life measure to calculate QALYs.”

American Journal of Preventive Medicine 20 (1): 35-9.

Noether, M., (1988) “Competition Among Hospitals,” Journal of Health Economics, VII, 259-284. Paolucci, S., G. Antonucci, M. G. Grasso, M. Bragoni, P. Coiro, D. De Angelis, F. R. Fusco, D. Morelli,

V. Venturiero, E. Troisi, and L. Pratesi. (2003) “Functional outcome of ischemic and hemorrhagic stroke patients after inpatient rehabilitation: a matched comparison.” Stroke 34 (12): 2861-5. Phibbs, C. S. and J. C. Robinson. (1993) “A variable-radius measure of local hospital market

structure.” Health Service Research 28 (3): 313-24.

(13)

Quality of Care? Hospital Death Rates and the NHS Internal Market.” Journal of Public

Economics 88 (7-8): 1247-72.

Robinson, J. C. (1988) “Market Structure, Employment, and Skill Mix in the Hospital Industry.”

Southern Economic Journal 55 (2): 315-25.

Robinson, J. C. and H. S. Luft. (1985) “The Impact of Hospital Market Structure on Patient Volume, Average Length of Stay, and the Cost of Care.” Journal of Health Economics 4 (4): 333-56. Robinson, J. C. and H. S. Luft. (1987) “Competition and the cost of hospital care, 1972 to 1982.”

JAMA 257 (23): 3241-5.

Shih et. al. (2004) “The Design, Contents, Operation and the Characteristics of the Respondents of the 2001National Health Interview Survey in Taiwan.” Taiwan Journal Public Health, (22): 419-430. Shortell, S. M. and E. F. Hughes. (1988) “The effects of regulation, competition, and ownership on

mortality rates among hospital inpatients.” New England Journal of Medicine 318 (17): 1100-7. Town, R. and Vistnes, G. (2001) "Hospital Competition in HMO Networks,” Journal of Health

Economics, 20 (4): 3-53.

Tsai, W.-D. and I.-h. Li. (2002) “Hospital Nonprice Competition and Market Structure: An Empirical Study of Hospitals' Acquisition of High-Tech Medical Equipment. (In Chinese. With English summary.).” Academia Economic Papers 30 (1): 57-78.

Wong, H. S., C. Zhan, and R. Mutter. (2005) “Do Different Measures of Hospital Competition Matter in Empirical Investigations of Hospital Behavior?” Review of Industrial Organization 26 (1): 61-87.

Zwanziger, J. and G. A. Melnick. (1988) “The Effects of Hospital Competition and the Medicare PPS Program on Hospital Cost Behavior in California.” Journal of Health Economics 7 (4): 301-20.

醫院競爭對醫療成本和品質的影響

行政院國家科學委員會專題研究計畫 成果報告