First, it is unlikely that a single measure can capture the multi-dimensional nature of health care quality and this limits the extension of any results to the whole health care system. For example, the mortality rate would not be the best indicator of the quality of cataract surgery, or maternity services.
Improving the quality of health care is important to service users and taxpayers, and therefore to Government and the media. Stories, usually focusing on poor quality of care, abound. And NHS quality initiatives are seemingly endless. Since the early nineties, policy makers have focused on increasing competition forces between providers to drive up the quality and to drive down the cost of health services. The positive effect of competition on productivity is well established in the literature on the private sector. Health care markets, however, differ from other markets in ways that do not allow direct translation of those results (Gaynor and Vogt, 2000). A key difference is that 'quality' is either very difficult or impossible to observe directly.
In both sets of studies, quality was measured by mortality rates following heart attacks. This quality has several advantages over other measures such as in-hospital complications, infection, readmission rates or waiting times: observation is easy; there is no dependency on patient behaviour after the medical procedure; it is not subject to gaming; comparison with another hospital is easy; and, it is correlated to good management practices (Bloom et al., 2010). Yet no other disease was studied, because it is difficult to build an argument for mortality rate as a good indicator for the quality of outcome for other conditions.
Despite these advantages, there are reasons to reject post-cardiac arrest mortality rates as an indicator of provider quality. First, it is unlikely that a single measure can capture the multi-dimensional nature of health care quality and this limits the extension of any results to the whole health care system. For example, the mortality rate would not be the best indicator of the quality of cataract surgery, or maternity services. The variability of hospital's death rates over time can be a challenge for econometric estimations, and the problem is aggravated in hospitals with low volume of admissions. Patients admitted in hospital x, which end up being treated in hospital y, if not well tracked in the data (when information is available), may distort hospital x's death rate. Also, early discharges of patients could give inaccurate measures of real mortality rates.
In response to these limitations, The English National Health Service (NHS England) has made great efforts to create new measures of quality. These measures attempt to incorporate not only clinical outcomes, but also consider patients' perception about the medical care they receive and the environment where this care was provided. These are specially relevant to assessing the quality of elective treatments or primary care services. The main quality measures available are:
· Quality and Outcomes Framework (QOF);
· Patient-Led Assessments of the Care Enviroment (PLACE);
· Data on Written Complaints in the NHS;
· Patient Reported Outcome Measures (PROMs);
The QOF is an incentive-based payment scheme that aims to measure and improve quality in primary care. It was implemented in 2004/5 and is composed of four domains: clinical, organizational, patient experience and additional services. Each domain has a set of indicators to proxy levels of achievement (148 indicators in 2012/13 for all domains). The participation of General Practices is voluntary. This measure provides valuable information to track improvements in the delivery of primary care. However, there are some drawbacks that should be considered. Pure comparisons of practice-level performance could be misleading and it is advisable to perform comparisons with additional information on social, demographic, and infrastructure characteristics of the practice. Given the anonymity in the data collection process, it is not possible to analyse co-morbidities. Lastly, QOF does not cover every clinical condition or every clinical areas.
PLACE are self-assessments of non-clinical services that impact patients experience of care inside organisations that supply NHS services. Participation started in 2013 and is also voluntary. The programme aims to promote some of the principles established in the NHS constitution. The assessment is composed by five areas: cleanliness; food and hydration; organisation and food ward; privacy, dignity and well-being; and, condition, appearance and maintenance. Efforts are made to obtain an unbiased measure of quality, as training is offered to the teams that undertake the assessments. Nevertheless, there is always an element of personal judgement in the patient assessor that cannot be questioned.
A dataset that counts the number of formal complaints made by patients (or their representatives) can be found in the Data on Written Complaints. It is collected for NHS Hospitals and Community Health Services (HCHSs) and Family Health Services (FHSs). The datasets have been published since 1997-98 and complaints are labelled by service area, profession and subject. Comparisons over time are complicated by the reorganisation the NHS has witnessed over the past 20 years. From 2011/12, the number of complaints upheld started being recorded as well. This is another worthy source of quality information on NHS England. It might be, however, subject to some level of selection bias as organisations that are more welcoming of patients' complaints, and publicize this fact, are prone to receive a higher number of formal objections than organisations with a more defensive approach to this matter.
PROMs are designed for the evaluation of four common elective inpatient surgical procedures: hip replacement, knee replacement, groin hernia and varicose vein surgery. The patients are invited to provide answers to pre- and post-operative questionnaires with questions about their general and condition-specific health. Those answers are then used to build generic and specific measures of health: EQ-5D index; EQ VAS; Oxford Hip Score; Oxford Knee Score; and, Aberdeen Varicose Vein Questionnaire. The data are released annually from 2009/10 and captures another important dimension of quality that may help to understand what drives its increase in the health markets.
These four measures together, offer a far better view of the quality of care provided by the NHS England than a pure rate of mortality could do. None of the indicators alone can be seen as substitute as they focus on different aspects of quality or different levels of care. Despite the fact that data are not available for the same periods of time, health economists can now produce more precise results about the determinants of quality in the health care sector.
In writing this post I do not intend to question the previous studies on the relation between competition and health care quality, but rather to highlight the fact that these findings apply to a specific set of hospital admissions. Also, to emphasize a need for the literature to evolve using more patient-reported outcomes as they are becoming more informative and covering a wider scope. Studies exploring choice and competition in primary care have already been done and more of these should take advantage of these new and rich sets of information (Gravelle et al., 2013 and Santos et al., 2013}.
Bloom N, Propper C, Seiler S, Reenen JV. 2010. The impact of competition on management quality: evidence from public hospitals. CEP Discussion Paper 983.
Cooper Z, Gibbons S, Jones S, McGuire A. 2011. Does hospital competition save lives? Evidence from the English NHS patient choice reform. Economic Journal 121: F228-260.
Gaynor M, Vogt W. 2000. Antitrust and competition in health care markets. In Culyer A and Newhouse J (Eds.), Handbook of Health Economics (1405-1487). Elsevier, edition 1, volume 1, chapter 27.
Gaynor M, Propper C, Seiler S. 2011. Free to choose? Reform and demand response in the English National Health Service. CEP Discussion Paper 1179.
Gaynor M, Moreno-Serra R, Propper C. 2013. Death by market power: Reforms, competition and patient outcomes in the National Health Service. American Economic Journal: Economic Policy 5 (4): 134-166.
Gravelle H, Santos R, Siciliani L. 2013. Does a hospital's quality depend on the quality of other hospitals? A spatial econometrics approach to investigating hospital quality competition. CHE Research Paper 82.
Propper C, Burgess S, Green K. 2044. Does competition between hospitals improve the quality of care? Hospital death rates and the NHS internal market. Journal of Public Economics 88: 1247-1272.
Propper C, Burgess S, Gossage D. 2008. Competition and quality: evidence from the NHS internal market 1991-99. Economic Journal 118: 138-170.
Santos R, Gravelle H, Propper C. 2013. Does quality effect patients' choice of doctor? Evidence from the UK. CHE Research Paper 88.