A range of specific policies designed to increase both patient choice and hospital competition has been introduced in, amongst other countries, England, Denmark, Sweden, Norway, and the Netherlands. A primary concern arising from such reforms is the effectiveness of hospital competition to provide improvements in quality, responsiveness, and efficiency. Theory would suggest that if hospital prices are not fixed but endogenously determined by the hospitals themselves, and quality is not easily observed or verifiable, then hospitals may react to increased competition for funds by offering lower quality at a given price, thus chiseling on quality, attracting higher volume and funding but producing lower quality output. Competition may be introduced, but it may not produce the desired effect.
Theory also suggests, however, that if prices are set exogenously increased competition will lead to higher quality, although, it has also been noted that, if provider preferences are sufficiently altruistic, high quality provision can also occur within a restricted competitive environment. Indeed, theoretically, if altruism is sufficiently high there may be a negative relationship between competition and quality provision. Thus examination of the incentive structures and the environment into which these are introduced is critical. This has been the subject of debate, at the core of which is the notion that, given a regime of fixed prices, hospitals will compete for patients and therefore revenue, through improving the quality of care offered. Fixed hospital prices are essentially associated with Diagnostic Related Group (DRG) prices for predefined case groupings. Those in favor of hospital competition argue that with fixed price competition for patients, efficiency and quality improve as hospitals increase their performance or risk losing their market share. Those against competition argue that such market-based reforms can destabilize hospitals, increase transaction costs, and possibly even harm patients.
This article examines the empirical evidence on patient choice and hospital competition to consider whether competition is associated with an improvement in hospital quality and patient outcomes. To do so, the general literature that considers hospital competition and quality is assessed. Before this examination of the literature however, the conceptual difficulties of measuring competition in this sector are discussed.
Issues In Measuring Competition
To assess the impact that hospital competition has on clinical quality there has to be an agreed definition of market power. The major challenge is the estimation of the size of the competitive market and the power exercised by individual hospitals. It is obvious that incorrect definition of the potential market would result in biased assessment of the impact of competition.
In product markets price relationships, in particular own-price and cross-price elasticities, may be examined to aid definition of the relevant market. In the hospital sector this is not relevant as prices, even if known, are highly regulated. Typically, investigators calculate hospital market size through concentrating on the definition of geographic area instead and do so in one of three ways. First, geographic market area may be defined as based on a fixed radius, defined by a largely arbitrary distance that creates a circular market of radius r. Investigators then calculate the degree of competition inside that market. Fixed radius measures have the possibility of both overestimating and underestimating the actual size of the market. The shortcomings of such fixed radius measures is that they do not take account of potential demand when they estimate market size. As a result, the fixed radius measures may suffer from urban density bias and overestimate competition in urban areas. However, an advantage of this type of fixed radius market definition is that the market size tends not to be endogenous to any other factors, such as hospital quality.
A second option is to create a variable radius market where the radius r that dictates the size of the market varies according to preexisting referral patterns, actual patient flows, or hospital catchment areas. For instance, a variable radius r could be set at a length that captures the home addresses of 75% of patients at a particular hospital. Variable radius measures tend not to be as affected by urban density bias but some argue that, when the radius r that defines the size of the market is based on existing referral patters or hospital catchment areas, the market size they estimate may again be biased. For example, a high performing hospital may have a larger catchment area than a lower quality competitor.
A third option is to create a radius that varies according to travel distance. An example of a travel-based radius would be to define radius r as the distance that captures the hospitals within a 30-min travel time from a particular patient’s home address. Market definitions based on existing referral patterns may be related to the real or perceived quality of local hospitals, but can suffer from referral patterns reflecting quality. Some argue that any estimates of competition that rely on actual patient flows may still be biased. Rather than using actual patient flows, predictions of patient flows to specific hospitals may be used to reduce this bias. Some studies have used predicted demand to estimate market size, based on travel distance for patients, arguing that their method mitigates the problems of traditional fixed and variable market measures of competition. However, in practice, sizes of markets defined using radii derived from travel distances tend to be highly correlated with the sizes of fixed radius markets. Because the two market definitions produce results, which are so closely correlated, they both tend to be affected by urban density bias. The key issue with both market definitions is that they require a largely arbitrary definition of the size of the market, such as 30 km for fixed measure and a 30-min travel time for time variable measure. Both market definitions may therefore either overestimate or underestimate the true size of the market depending on how the upper boundary of the market is set by researchers.
All three approaches have been applied to the hospital market; none is perfect. Each measure has its own strengths, weaknesses, and inherent bias. A practical approach in considering which method to employ is to assess the compatibility of the data with the various measures, to trade-off the inherent bias contained in each method by comparisons across a number of measures and to explore the use of instrumental variables to overcome any endogeniety.
General Evidence On The Relationship Between Hospital Competition And Clinical Quality
The largest volume of literature assessing the relationship between hospital competition and quality comes from the USA (see Gaynor, 2006 for an overall review). The bulk of the existing US literature has investigated the relationship between competition, prices, and capacity and is rather out of date. There is a related small, but growing literature in the US that looks directly at the impact of hospital competition on clinical performance. A number of studies consider endogenous price environments and, unsurprisingly, the general finding with respect to the influence of increased competition on outcome quality is ambiguous.
A smaller number of recent studies on competition and quality tends to the conclusion that, under exogenously determined fixed-price competition, higher levels of competition generally lead to improvements in clinical performance. The bulk of this US literature on hospital competition and clinical quality examines the outcomes of Medicare beneficiaries and within the timeframe of these studies Medicare operated an exogenously determined DRG pricing scheme. Findings generally support a positive relationship between in-hospital mortality and increased hospital concentration (Kessler and McClellan (2000) is a prime example). One study found that competition was associated not only with improved outcomes in the Medicare population but also with more intensive treatment for sicker patients and less intensive treatment for healthier patients who needed less care.
The literature outside the US is smaller but supports the general findings. There is a growing, recent literature on hospital competition within the National Health Service (NHS) in England, for example. It is based on the introduction of a purchaser–provider split, where GP practices purchased secondary hospital care on behalf of their patients. As initially introduced, these reforms were said to have created an internal market in health care. They were based on various contractual arrangements. Hospital prices were generally not fixed and can therefore be assumed endogenous. There is a wide consensus that the internal market never created high-powered incentives for hospitals or developed a significant degree of competition. Notwithstanding this criticism, there is some evidence that prices fell during the internal market. One study also found that, during the initial phase of the internal market, higher competition was not associated with lower quality.
Examination of the impact of the NHS internal market on patient waiting times and length of stay for hip replacement from 1991 to 1994/5, using survival analysis to look at hospital level data during the internal market reform period, found that waiting times for hip replacements fell and so did patients’ average length of stay. This study found that, after the internal market was introduced, patients were more likely to be transferred to another facility rather than remaining in the hospital where they had the surgery until they were ready to be discharged home.
The strongest evidence on the impact of hospital competition on patient quality in the NHS comes from a number of English studies. This article considers various aspects of increased competition on hospital quality. The dominant quality measure, 30-day AMI mortality, was chosen because, being tied to an emergency treatment and largely associated with in-hospital mortality, it is not easily manipulated by hospital admission policies. The mechanism through which AMI-mortality may be used as a proxy for general hospital quality is not always made explicit, but hinges on the presumed correlation between the management of AMI treatment and wider hospital practices. One study of the impact of the internal market (presumed competitive) on hospital quality as it had been before 1999, i.e., a period before the fixing of hospital prices, used a 30-min drive time from ward centers as the competitiveness measure. Using hospital level data and controlling for hospital and local area characteristics, it was found that the internal market led to a small but statistically significant increase in 30-day AMI mortality, the adopted measure of quality (Propper, 1996).
A further study (Propper et al., 2008) used a longer time period to assess whether more competitive areas had higher or lower AMI mortality over the period 1991–1999. Once again this is a period of endogenously determined prices. Similar to the findings from their previous work, the report that higher competition during periods of competition was associated with higher AMI mortality, i.e., higher competition is associated with lower hospital quality in this dimension. They argue that it is not credible that hospitals deliberately sought to curtail quality in this manner – hospitals did not deliberately worsen 30-day AMI mortality. Rather it is suggested that as the internal market increased competitive pressures hospital resources were shifted from quality domains that were not fully observable and verifiable such as the impact of hospital care on health outcomes, to those, such as waiting times for elective procedures that were easily measured and were being targeted.
The introduction of DRG-type prices into the English NHS in 2005/06 fixed hospital tariffs at the same time as competition within the NHS was strengthened. Two recent studies have used difference-in-difference estimators to examine the impact of this increase in competition on hospital quality using 30-day AMI as the measure of hospital quality. Cooper et al. (2011) found that AMI mortality decreased more quickly for patients living in more competitive areas than that in less competitive areas. Specifically in the three-year period after the reforms were introduced, a one standard deviation increase in hospital competition was associated with approximately a 1% decrease in AMI mortality. Gaynor et al. (2010) found a similar impact of the increase in competition on hospital quality, again measured through 30-day AMI death rates, over the period 2003 to 2007. Both studies, therefore, find that increased competition under a fixed price regime within the English NHS over the period 2002–8 improved hospital quality even though a different aggregation of data and different methods are used.
There is also a small empirical literature that considers the impact of increased hospital competition on equity and patient access. The hypothesis is that competition may have a detrimental effect on equality of access for NHS patients. Waiting times for patients having an elective hip replacement, knee replacement and cataract repair over the period 1997 and 2007 in England seem to have generally decreased as competition increased, with the variation in waiting times for those procedures across socioeconomic groups also greatly reduced. Cookson et al. (2010) examined the impact of the internal market on equity, measured as the association between patient deprivation and hospital utilization. They compared competitive and noncompetitive areas, where competition was measured using a Herfindahl–Hirschman (HHI) index in a fixed radius market and also found that there was no evidence that competition had a worsening effect on socioeconomic health care inequality.
Conclusions
This short review has confirmed what was to be expected from theory: Under exogenous fixed-price regimes health care reforms, which increase competition among hospital providers, can lead to improved outcome of quality. There is not a large volume of empirical evidence that can be used to test this theoretical conclusion but what does exist is rather robust. The methods used tend to be similar and reliant on robust estimation procedures, including difference-in-difference estimation and large data sets. One criticism of these findings is that a large number of studies use a similar proxy measure of hospital quality: 30-day AMI mortality. There are justifiable reasons for the choice of this measure: It is associated with an emergency admissions and treatment, which is difficult to manipulate by the hospital providers. It is nonetheless a one-dimensional measure of quality and the generalizability of the empirical findings rest on a belief that there is a strong correlation between this dimension and other less verifiable dimensions of hospital quality. It is perhaps not too difficult to buy into the belief that if hospitals have good management structures all dimensions of quality will trend in a similar manner. Other empirical research has indeed found that hospitals with better overall management skills had lower mortality from AMI. Moreover, recent studies show that this measure of hospital quality (30day AMI mortality) is indeed correlated with other hospital outcome measures. The policy implications appear clear that with a fixed price regime competition can be improving. That this is not found when prices are set endogenously is perhaps an unsurprising lesson.
References:
- Cooper, Z. N., Gibbons, S., Jones, S. and Mcguire, A. (2011). Does hospital competition save lives? Evidence from the NHS patient choice reforms. Economic Journal 121, F228–F260.
- Gaynor, M. (2006). Competition and quality in health care markets. Foundations and Trends in Microeconomics 2, 441–508.
- Gaynor, M., Moreno-Serra, R. and Propper, C. (2010) Death by market power: Reform, competition and patient outcomes in the National Health Services. CMPO Working Papers. UK: University of Bristol.
- Kessler, D. P. and Mcclellan, M. B. (2000). Is hospital competition socially wasteful? Quarterly Journal of Economics 115, 577–615.
- Propper, C. (1996). Market structure and prices: The responses of hospitals in the UK National Health Service to competition. Journal of Public Economics 61, 307–335.
- Propper, C., Burgess, S. and Gossage, D. (2008). Competition and quality: Evidence from the NHS internal market 1991–1996. Economic Journal 118, 138–170.
- Baker, L. C. (2001). Measuring competition in health care markets. Health Services Research 36, 223–251.
- Bloom, N., Propper, C., Seiler, S. and Van Reenan, J. (2010). The impact of competition on management quality: Evidence from public hospitals. CEP Working Paper – 14 February 2010 Draft. London, UK: Centre for Economic Policy, London School of Economics.
- Breeke, K., Siciliani, L. and Straume, O. (2009). Hospital competition and quality with regulated prices. CESinfo Working Paper – 2010. Munich, Germany: Centre for Economic Studies and Ifo Institute for Economic Research (CESifo).
- Cookson, R., Dusheiko, M., Hardman, G. and Martin, S. (2010). Competition and inequality: Evidence from the English National Health Service 1991–2001. Journal of Public Administration Research and Theory 20, 181–205.
- Cooper, Z. N., McGuire, A., Jones, S. and Le Grand, J. (2009). Equity, waiting times, and NHS reforms: Retrospective study. British Medical Journal 339, b3264.
- Gaynor, M. (2004). Competition and quality in hospital markets. What do we know? What don’t we know? Economie Publique 15, 3–40.
- Kessler, D. P. and Geppert, J. J. (2005). The effects of competition on variation in the quality and cost of medical care. Journal of Economics and Management Strategy 14, 575–589.
- Le Grand, J. (2009). Choice and competition in publicly funded health care. Health Economics, Policy and Law 4, 479–488.
- Tay, A. (2003). Assessing competition in hospital care markets: The importance of accounting for quality differentiation. Rand Journal of Economics 34, 786–814.