11.1.3. The Performance of Health Systems

 

The impact of health care (or a health system) on the population’s health has been a topic of much research and attention since Thomas McKeown’s work in the 1970s (McKeown, 1979). And while McKeown’s argument that, rather than health care, it was in fact wider ranging social policies (e.g. education leading to cleaner water and a better understanding of nutrition) that were responsible for allowing individuals to live longer, healthier lives, this was a finding conditioned by the reality of the day; the reality has since changed as the scope of health care and range of interventions and pharmaceuticals is much greater today. As a result, the concept of avoidable mortality i.e. mortality which could have been prevented through access to adequate if not high quality health care, began finding currency in the 1980s and 1990s, spreading from initial ‘specific health event’ studies in the United States to more comprehensive assessments also in Europe (e.g. Rutstein et al, 1976; Mackenbach et al, 1990; Holland et al, 1994; Westerling, 1996). These studies provided evidence that (access to) health care had a positive impact on the population’s health – often by decreasing infant mortality rates – but it is to be acknowledged that measurement remains difficult. In assessing the contribution to health outcomes, disaggregating access to effective treatment from broader lifestyle changes is an especial challenge. We need therefore to be cautious in drawing conclusions from measures of health system performance that are based, at least in part, on broad population health indicators.

 

Nevertheless, the concept of avoidable or amenable mortality was then revisited by Nolte and McKee (2005), who aksed themselves whether this concept might be used as an indicator for health system performance measurement. Their findings and review of the literature demonstrate the link between access to health care and improved population health during the ’80s and ‘90s, primarily because of declining infant mortality rates and improved numbers of life years for the middle-aged and elderly, especially for what concerns women. Nolte and McKee call for more detailed studies on the causes of death where there is a link to health system performance. This has brought to a growing consensus towards the concept that high-performing health systems can help in achieving better population health outcomes; a consensus supported and in part driven by the WHO’s health systems framework.

 

In view of the above, health care system performance is a high priority on international and national policy agendas. Constantly increasing costs to fund desired activities on the one hand, and a deficiency of knowledge about the effectiveness of invested resources on the other hand, has created an increasing demand for more information on the performance of health systems as a basis for a continuous improvement cycle. Therefore, it is necessary to perfom an examination of the functions of health systems so as to understand how their improved performance can help tackle population health needs.

 

First of all, one should bear in mind that health systems can be grouped in many ways: by level of (de)centralization; by extent of government regulation or market orientation; and by outcomes in terms of health attainment. One framework used to categorize health systems across countries is through a combination of financing and organization characteristics. Thus, four broad types of health systems can be identified in Europe, varying in terms of revenue collection, pooling mechanisms and purchasing (Smith, 2004):

1) ‘unreformed’ social health insurance e.g. France, Austria, Greece, Slovenia;

2) social health insurance with competitive sickness funds e.g. Germany and the Netherlands;

3) devolved tax funded systems e.g. Sweden and Spain;

4) centralized tax funded systems e.g. UK and Italy.

This delineation provides a useful starting point for a comparative analysis, although as shown throughout this chapter, there are important differences between each system, even within the same broad category.

 

Recent health care trends in the European Union’s (EU) new Member States and Candidate Countries of central and Eastern Europe (CEE) paint a distinct picture. In a time of rapid political and economic transition, first following the fall of the Berlin Wall at the end of the ’80s, and more recently precipitated by the process of EU accession, many have experienced a fundamental reform of their governing and structural institutions as well as a shift from centralized to market-based mechanisms. Broadly, there was a move away from the centralized integrated Semashko model of health care to a more decentralized and contracted model of social health insurance (Bismarck model)2. This shift, pursued with varying degrees of success and ease among the countries, had numerous goals: increase the responsibility of individuals for their own health and costs of health care; improve efficiency by holding providers more accountable for the use of resources; and shift the responsibility of health services away from central government authorities (Preker et al, 2002).

 

While the past decade saw continuous efforts for improving efficiency in central and Eastern Europe involving, among other things, a reduced role of the government, in Western Europe the most recent era of health reform can be characterized by an increased attention on effectiveness through improved quality; in many cases through an increased role of government, either directly or through arms-length bodies. These efforts to ensure and improve quality of care are currently present alongside persistent concerns towards efficiency and cost containment (OECD 2007). Indeed, continuous pressure on health systems worldwide stem from the conflicts between the efficiency and cost-containment goals and equity objectives, in terms of both funding and access; these conflicts become more complicated as the cost of health care continues to rise (Cutler, 2002). Many countries have implemented numerous regulatory mechanisms to facilitate improvements in the quality of care and the achievement of quality standards. The focus on patient rights and patient safety has been integral in these reforms, with the introduction of clinical guidelines and linking payments to performance and quality all part of the new reform agenda.

 

Although not the focus of this chapter, within this context of increasing cost pressure on European health systems, the role of the individual or patient is another factor requiring consideration. Are patients to be seen as passive recipients (or consumers) of care provided via the health system, or are they more an active demand-side player, in turn putting further demands on health system sustainability? The answer is a combination of both, the balance in different countries depending on many factors from culture, to ability to pay or nature of the health system itself. The role of society and the contribution of the medical profession itself – the so-called medicalization of society as first developed by Illich (1975) – are also issues to be considered, but which again fall outside this chapter. Nevertheless, the increased involvement of individuals in their own medicine, and the pros and cons associated to the more recent trend towards patient-centered medicine (Tomes, 2007), is another element to be borne in mind when discussing the characteristics, performance and responsiveness of health systems.

 

11.1.3.1. Cost pressures and performance measurement

 

Since the 1970s, spending on health care has grown more quickly than national income due to the limited constraints on spending and growing technical sophistication of medical practice. On average, public expenditure on healthcare per capita increased by almost 100% between 1990 and 2004, although there is considerable diversity between countries. While Eastern European countries spend less than US$1000 per capita on healthcare, the majority of other European countries spent between US$1,000 and US$2,500 per capita in 2004 (table 11.1). Luxembourg spends almost twice as much as any other EU27 country per capita at US$4,603.

 

Table 11.1. Private and Public expenditure ($US 2004) on healthcare per capita (1990-2004)

 

As a response, there was a gradual shift in the 1980s away from ensuring the generosity and equality of coverage to cost containment and increasing efficiency, in large part by introducing tighter regulatory mechanisms (Mossialos and Le Grand, 1999; Cutler, 2002) (Docteur and Oxley 2003). For instance, in the hospital sector, countries that relied on global budgets restricted the budgetary limits to contain costs and in many cases introduced activity-based payments to increase productivity (e.g. in Italy). For physicians, depending on the policy objectives in different countries, some measures to contain costs and improve efficiency consisted of tightening the fee schedules and introducing prescription drug budgets (e.g. in Germany).

 

Despite some success in putting downward pressure on increasing health expenditures, these cost-containment policies and the rationed model became less popular. In some countries, these reforms were associated to: increased difficulties in accessing care, such as long waiting lists; a lack of incentives to increase efficiency; and only short-term cost savings (Cutler, 2002). Health reform in the 1990s thus focused more on introducing incentives that were expected to improve efficiency largely by: a) increasing cost sharing; b) introducing competition between insurance funds in systems with social insurance; c) separating purchasing from provision; and d) reforming payment mechanisms.

 

In order to measure the health system’s contribution to socially desirable goals in general, and to the health status of citizens in particular, estimating the efficiency with which the resources are used to attain wanted outcomes – and, ultimately, to support continuous health system improvement efforts – is a key element for assessing the performance of health systems. The concern with measuring the performance of health systems dates back to the 1800s when Florence Nightingale initiated the practice of comparing hospital outcomes data in order to better understand and improve performance. Only in the last decade or so has the vision of making use of large-scale data sources to measure and improve health system performance been realized. Performance data serve broadly two purposes: identify the works to achieve the health system objectives and identify specific areas of competence (Smith, 2002).

 

Since the publication of the World Health Report by the WHO in 2000, which yielded a ranking of health systems in 191 countries, most governments have been concerned with the measurement of their own system’s performance and comparing strengths and weaknesses with other countries. One of the major criticisms of any approach to measure performance based on aggregate data is that they provide little or no indication of what needs to be done to improve the system when faced with evidence of sub-optimal performance. It is vital that more detailed analyses of the elements of health care systems are conducted in order to disentangle the specific issues that require attention; some recent OECD projects on performance measurement show some promise in this regard (e.g. the OECD Health Project 2004, and Smith 2002).

 

The World Health Report does, however, provide a useful starting point for measuring health system performance. It outlines three fundamental dimensions of performance: health attainment as measured by healthy (or disability-free) life expectancy, responsiveness to the needs of the population, and fairness of financing. The report further delineates four functions through which countries can achieve these goals: service provision, resource generation, financing and stewardship. One of the major contributions of this report was the development of an instrument to assess the responsiveness of health systems to the population’s needs based on seven dimensions: autonomy, confidentiality, dignity, prompt attention, quality of basic amenities, access to social support networks during care and choice of providers. While conceptually this is clearly important to consider when evaluating a country’s health system, these indicators are incredibly difficult to measure and even more difficult to compare across countries.

 

Other macro-level analyses of performance have been conducted, relying heavily on health status indicators to measure performance (Retzlaff-Roberts et al, 2004; Social and Cultural Planning Office of the Netherlands, 2004; The Conference Board of Canada, 2004). These, among others, have sought to develop a single indicator of performance. However, it is highly unlikely that a single estimate could possibly capture the multitude of costs and benefits of the health system. Furthermore, a composite indicator compounds the inaccuracy of its component measures (Naylor et al, 2002). While rankings may be popular among policy makers, those based on a single measure may be misleading and uninformative, whereas those based on disaggregated indicators such as infant mortality, waiting lists or other specific measures, are likely to be more easily related to policies or practices (Navarro, 2000).

 

The increasing attention paid to measuring and reporting performance data results from multiple influences. On the supply side, the massive advances in information technology have greatly enhanced the feasibility and reduced the price of collecting and processing data. On the demand side, a growing popular and political scepticism of health professions and institutions has emerged. This calls for greater independent audit and accountability. Alongside the increased potential for deploying performance measurement tools in modern health systems, and the large number of experiments underway, there are still many unsolved debates on how to measure and use performance data in the best possible way. Moreover, many systems of performance assessment, some noted above, have been developed in a piecemeal and opportunistic fashion. As effective performance assessment relies on a more pragmatic approach that depends on the health system being assessed, the availability and reliability of data, and the intended use of the performance indicators, more research is needed to better understand how to measure performance and how to integrate performance data into the regulation and governance of health care. The remainder of this section identifies some of the key areas to address in performance measurement based on the objectives of the health system; these include: access to care, quality and appropriateness of care, technical efficiency and the patient’s experience.