Rationing health care in the UK from an economic perspective
After having discussed the concept of rationing in the health care system of the United Kingdom (UK) from an epidemiological point of view, this chapter focuses on several rationing issues in the UK from an (health) economic perspective. Although some authors conclude differently in several case studies (see for example Stanton 1999), health economics have been (and still is) of great relevance to the notion of rationing health care and helpful in finding solutions for important rationing problems. That one cannot maintain the position that economics plays only a relatively minor role in rationing health care, is underlined by the work of several well-known academic scholars (see for example Bloor et al. 2003 and Maynard et al. 2004) and the fact that many rationing specialists have a background in economics. Therefore, reporting on health care priority setting in a country like UK without addressing some economic aspects would not result in the desired multidisciplinarity (see above).
The organization of the chapter is as follows. First, the levels of spending on health care in the UK’s National Health Service (NHS) will be discussed in relation to expenditure on health care in other (predominantly European) countries. Then, the means of raising funds in the NHS will be explored. The emphasis in this section will mainly lie on the distribution of UK health care expenditure in user charges and public and private funds. The third section deals with the various rationing mechanisms which are applied in the NHS. Implicit and explicit forms of priority setting and supply-side and demand-side rationing mechanisms will, among others, be discussed in this section. The last part of this chapter consist of some equity considerations. Finally, some concluding remarks will follow.
Thus, the following main topics will be leading in this chapter:
- Levels of health care spending in the UK
- Means of raising funds in the NHS
- Rationing mechanisms in the NHS
- Equity considerations
Levels of spending
Table 3.1 shows the total expenditure on health care in the United Kingdom for selected years. The table includes expenditure in current and constant prices; per capita expenditure in US dollars purchasing power parities at current prices; health expenditure as a share of GDP; and public expenditure as a share of total expenditure (EOHCS 1999). All expenditure series display a general upward trend. Health expenditure as a share of GDP rose between 1970 and 1992 from 4.5% to 6.9%. Thereafter it remained stable. Public expenditure on health as a proportion of total expenditure fell from 91.1% to 84.1% between 1975 and 1990. Thereafter, however, the public share has remained fairly constant. Public expenditure on the NHS dominates expenditure on health in the United Kingdom (Ibid.).
The next step is to look where this money is spend on. Table 12 shows some of the main categories of spending on health in de UK, as a proportion of total expenditure, over the same period as table 3.2.
The first thing one notices is that the spending on inpatient care has decreased from 53.5% in 1980 to 42.2% in 1995. Further the spending on pharmaceuticals displays a long-term upward trend, which has reached a percentage of 17.3 of the total expenditure by 1997 (Ibid.). The other noticeable trend has been the long-run decline of public investment as a percentage of the total investment, and a more modest decline in total investment since 1990 (Ibid.).
Spending on health in the UK is also rising at a much faster rate than other public services. This is indicated by figure 3.1. While other parts of the public sector stay behind (defence, housing), the NHS (commencing from the same starting point) leaves everything behind, at least when it comes to spending.
Although UK spending on health as a proportion of gross domestic product is still lagging behind that of other European countries, the report of the Office of Health Economics shows that by 2008, it will be only 0.2% lower than it is in France (Towse & Sussex 2000). This is in contrast to 1997, when France’s spending as a proportion of GDP was almost 3% higher than that of the United Kingdom: 9.4%, compared with 6.8% (ibid.).
The target of the prime minister for total health expenditure was to reach 8% of GDP by 2006. In the article of Adrian Towse and Jon Sussex (2000) it is estimated that an annual growth rate of 5.8% is required form the fiscal year 2001-2002 onwards to achieve an 8% share of GDP (Ilis 2000). Further research is needed to evaluate whether this goal is achieved.
Means of raising funds
This sections explores the ways in which in the United Kingdom money is raised to fund its health care system. The focus will lie on three main sources of finance: public sources (predominantly taxes), private sources and out-of-pocket payments. In the year 2000, 80.9 percent of total expenditure on health in the UK came from public sources, which leaves 19.1% for private means of finance, including payments paid out-of-pocket (Docteur & Oxley 2004). It is interesting to investigate how these sources are distributed across the National Health Service and the private sector. It must, however be noted that the used data are quite outdated and were available for different years for the relevant categories. Although this hampers interpretation to a certain degree, the presented figures are nonetheless sufficiently indicative for the present situation of funding health care in the UK.
All citizens of the United Kingdom are entitled to services delivered in the National Health Service (NHS). With the enactment of the National Health Service Act in 1977, the scope of the UK’s main health care system was specified and as the European Observatory on Health Care Systems (EOHCS) states, the act “requires the Secretary of State to promote a comprehensive health service designed to secure improvement in the physical and mental health of the population and to develop services for the prevention, diagnosis and treatment of illness” (EOHCS 1999:35). In principle, all hospital and specialist services are to be provided for free at the point of use, although co-payments can be charged for, among others, prescriptions and dental care. The UK’s NHS is probably the best known example of a tax-funded health care system which has as a result that the UK predominantly, but not solely, relies on public sources in order to fund its health system. Table 3.5 summarises the sources of finance for the NHS. The main sources of spending on the NHS in 1996-1997 were drawn from general taxation (81,5%) and national insurance contributions (12,2%). Further, a total of 6.3 percent of total spending was raised through other sources, like user charges – mainly for drugs and dental care – (2,1%) and Capital funds from NHS Trusts (3%) (EOHCS 1999).
Although the data presented in table 3.5 are not very recent they might nonetheless be indicative for today’s figures because the shares of the different sources of total spending on the NHS remain fairly constant. However, one should still be cautious when interpreting these data in the light NHS financing in 2007 because (relatively small) changes in the distribution did occur in the presented ten-year period: total public funding declined with 1.5 percent whereas total funding from other sources rose with 1.5 percent.
Private sources of finance
In addition to the services provided through the NHS, citizens are allowed to insure themselves privately (voluntary private insurance). In 1996, approximately 14.6 percent of health care spending in the UK was based on private sources, (EOCHS 1999). Within this percentage, user charges paid within the NHS framework are not included which partly explains the difference of 4.5 percent in relation to the earlier mentioned 19.1 percent. The remaining difference is possibly the result of more people buying private health insurance in the period 1996-2000 but this is, however, not likely (see table 3.6). Although quite modest compared with other developed countries, not to be neglected amounts were paid out of pocket in the private sector. “This took the form of payments for private medical care, payments for long-term care and co-payments for pharmaceuticals, dental and ophthalmic services” (EOHCS 1999:34).
A further observation made in the literature are the trends in equally growing public and private health expenditures in the 1990s. These trends are depicted in figure 3.2. “Increases in private expenditure on health have kept pace with public expenditures but have not exceeded them” (ASI 2000:4).
The Adam Smith Institute (ASI) state in their report (in which they advocate substantial changes in the current health care funding mechanisms) that although private expenditures on health have grown over the last couple of decades, their growth has recently stagnated around a point of approximately 15 percent of total expenditure on health. They conclude that, given the difficulty to obtain sufficient support of the public for increases in public expenditures (taxes) or cuts on spending in other public domains, the goal to increase NHS expenditure up to average health care spending in the EU (see above) may be achieved through encouraging private expenditure (ASI 2000).
According to Robinson (2002), total out-of-pocket payments as a share of total expenditure on health (both public and private) were about 10.8 percent in 1998 (see also Mossialos & Le Grand 1999). It is important to note that Robinson included both direct forms (like co-payments and deductibles) and indirect forms of cost-sharing (like coverage exclusions such as IVF; see also New 1996 for a discussion on this topic) in his calculations. Compared with other countries like Portugal (44.6% in 1995) Greece (40.4% in 1992) and Italy (23.5% in 1999) cost-sharing in the UK seems quite modest. However, some countries appear to have even lower shares of out-of-pocket payments for health care. Examples are France (10% in 1999), Luxembourg (7.4% in 1997) and The Netherlands (5.9% in 1998). Nevertheless, the level of cost-sharing in the UK remains below that of most other developed countries. Perhaps partly because of this fact, some are advocating increases in the level of cost-sharing. The Adam Smith Institute (ASI) is an example: “…we support a greater role for co-payments (encouraging individuals to take more responsibility for their health care)” (ASI 2000:8).
Cost-sharing in the form of out-of-pocket payments or co-payments (direct cost-sharing) is used extensively by a lot of countries as an important device for rationing health care. The next section explores and discusses the various rationing mechanisms adopted in the UK’s health care system. Cost-sharing is, apparently, one of these mechanisms.
This section deals with the various devices which are used by the UK government in order to ration the health care system. Most scholars, committees and institutes agree that rationing is inevitable and that it needs to take place, even if the necessary information is scarce. “There is no question of there being insufficient information to ration” (New 1996:23). The UK government has always underlined this point of view by stating, in 1996, that “rationing was part of the system” (De Gooijer 2000). More disagreement, however (especially in the UK) has arisen about whether rationing should take place implicitly or in a more explicit manner. Proponents of the latter seems to advocate a rationing policy based on public consensus and informed by relevant stakeholders like the medical profession and patient organisations (see for example New 1996, British Medical Association 1996 (in: the Gooijer 2000)). Proponents of the former argue that “the reality is not conducive to the serious adoption of such a stance [i.e. explicit rationing]” (Hunter 1997:8). Although the UK government historically mostly used implicit rationing devices (see below), the last decade or so more emphasis is being placed on more explicit forms of rationing (Mechanic 1995, an opponent of explicitness).This section does not aim to discuss the pros and cons of both types of rationing. Rather, it gives a description of the various rationing devices which are actually used in the UK. It will become clear that in the UK, both forms are employed (Purdy 1996). The section begins by investigating the presence of rationing mechanisms on the demand-side whereupon several supply-side applications of the concept will be discussed. It must be noted that much of the illustrative data is coming from England due to the lack of available data from other countries within the UK.
A way of controlling health care costs is by introducing user charges in some form (deductible, co-payment, etc.). Because of insurance individuals are insensitive for price and as a result they may consume more care than strictly necessary. This phenomenon is generally referred to as ‘moral hazard’. User charges may thus limit demand, since people become sensitive to prices, having them trade-off costs (as far as charged) and consequences (Brouwer 2007a). In the NHS system user charges are adopted for prescriptions, dental care and optical care. For the General practitioner and specialist care no co-payments are charged (Robinson 2002).
Approximately 80 percent of people aged between 18 and 60 have to pay prescription charges (Citizens Advice 2005). For pharmaceuticals, co-payments of £6 per item (2000) are charged (Robinson 2002). It is also possible to get a “season ticket” of approximately £70. However many people are exempt for this scheme (Docteur 2004). The exemptions differ from charges for children under the age of 16, elderly people, those on low income levels, to people with specific chronic conditions and for specified uses. By 1995/1996, 84% of prescriptions were dispensed to people claiming exemptions (EOHCS 1999). Dental charges work very differently from prescription charges. For example people over 60 are not exempt from charges, and tapered help is available with dental charges under the Low Income Scheme (Citizens Advice 2005). Charges are not levied on certain patient groups, mainly children, those on low incomes and pregnant or nursing mothers (EOHCS 1999). For dental care patients pay 80% of the cost of a course of treatment up to a maximum of £384 (Citizens Advice 2005). In 1998/1999 the average full cost of a course of NHS dental treatment was approximately £34 (EOHCS 1999). The system for charging for optical services represents another approach to charging. Here the service is basically delivered privately, but help is provided for people on low incomes through a complex system of vouchers (Ibid.). In principle there are no co-payment for Inpatient care, with exception for superior National Health Service (amenity) beds (Robinson 2002). Altogether Robinson calculated that 10.8% of total expenditure on health is out-of-pocket payment (Ibid).
A problem of user charges is that it may scare people away who are in need of care, but do not wish to face the financial consequences of receiving care. Citizens Advice concluded in 2001 that many people faced financial difficulty in accessing the health care they needed because of the impact of health charges (Citizens Advice 2005). Another problem of user charges is that it may have a higher impact on low-income groups than on high-income groups. If the reduction in care consumption is relatively strong in low-income groups, inequity is created. Equity considerations are further discussed in section 3.5.
Basic benefits package
Another way of limiting demand is to select which treatments are covered by the national scheme. By limiting this, more care is left to own payment or supplementary insurance. The aim of many countries is to have comprehensive care, but increasingly, it is considered necessary to keep some things out of the package in order to ensure incorporation of others (Brouwer 2007a). The scope of the NHS is specified in the National Health Service Act 1977. Under section one of the 1977 Act, all hospital and specialist services are to be provided free-of-charge, unless the law expressly permits charges to be made (EOHCS 1999). As already discussed in section 3.4.1, user charges are made for drugs, optical and dental services. The NHS does not specify an explicit list of services to be provided. At a general level, the 1977 Act imposes a number of responsibilities on the Secretary of State in relation to the provision of hospital and community health services (Ibid.).
In the case of pharmaceuticals, the scope of NHS benefits is more explicit than in other areas. Schedule 10 to the National Health Service (General Medical Services) Regulations 1992 lists drugs which may not be prescribed on the NHS by general practitioners; Schedule 11 to the same regulations lists drugs which may only be prescribed to the specified types of patient or for the specified condition(s) (EOHCS 1999).
An interesting case in this is the so-called ‘Child B case’ which caused a lot of public commotion. New (1996:1596) summarises it as follows:
Jaymee Bowen, aged 10 (…) had acute myeloid leukaemia. She was given some initial treatment, including a bone marrow transplant (…), but after a remission her cancer recurred. NHS clinicians (…) at Cambridge decided that further bone marrow transplantation was inappropriate: that the probability of a successful outcome was very slight (2.5%) and that treatment would cause considerable pain and distress. However, (…) Jaymee’s father pressed for another transplant, this time from another hospital in London. Cambridge Health Authority refused to pay for the extra contractual referral that this entailed on the basis that clinicians at both hospitals thought the treatment was unlikely to succeed and would cause considerable pain and distress. Jaymee’s father took the case to the high court, where the health authority was required to reconsider. However, on appeal the health authority’s decision was upheld. Cambridge Health Authority consistently argued that financial matters did not enter its decision. Treatment was finally offered in the private sector, (…) but again Cambridge Health Authority declined to pay. (…) the treatment ultimately provided was not bone marrow transplantation but a leading edge treatment [not performed on a child before]. The treatment was effective for a while and the cancer went into remission for over a year. It eventually recurred, however, and in May 1996 Jaymee died.
The second bone marrow transplantation came at a price of £75.000 at that time. The rejection of the health authority’s to pay this sum was, according to them, only based appropriateness (“the risks were too high and probabilities of benefits to be neglected”). There are, however, definitely rationing issues involved in this case. New (1996:1596) puts question marks on whether it is “ethically defensible to use resources in cases with very small probabilities of success and significant probabilities of harm: (…) does refusing to finance treatment in individual cases such as this damage the benefit of reassurance which the NHS provides? Are these sorts of judgments applied consistently across the NHS and is there sufficient explicitness to judge?”
Even with large increases in NHS expenditure, acute funding difficulties continue to emerge. It is essential that a national mechanism to prioritise new and existing technologies is available to inform decision making. The National Institute for Clinical Excellence (NICE) was created to meet this need (Maynard et al. 2004). The National Institute for Clinical Excellence issues guidance to local decision makers about services of proven effectiveness and recommended for adoption by the NHS (EOHCS 1999).
The role of the primary care physician
Since 1991 buyers of care can contract providers of care in the NHS (purchaser-provider split). The most important groups of buyers are the District Health Authorities (DHA’s) and the General Practice Fundholders (GP fundholders). They have the task to estimate the population’s need and to buy care conform the estimated need (Vermaas 1996). In introducing budgets for General Practitioners (GPs) and DHA’s it has brought rationing decisions closer to the service provider (Purdy 1996).
The DHA’s are the most prominent buyers of care and are responsible for hospital care and community health services (Vermaas 1996). Each DHA has his own population. The DHA’s often work together with family health service authorities (FHSA’s), who are responsible for GP care. The second group of buyers are the regional health authorities (RHA’s). They are responsible for care with a regional function. The third group are GP’s who voluntary can become fundholders if they meet the requirements. With the budget they can buy pharmaceuticals and some hospital care (Ibid.). The budget originally covered three parts; hospital care, including clinical care, policlinic care, day-care and non-clinical diagnostics (as for breast cancer); pharmaceuticals subscribed by the GP; and costs of non-medical personnel working in the practice. GP care is not included in the budget, but is financed separately (Ibid.). The budget is managed by the FHSA’s (Ibid.). By 1998 there were over 3500 fundholding practices covering 15 000 GPs (EOHCS 1999). From April 1999, as part of the nationwide primary care groups scheme, prescribing budgets for all GPs will be merged with hospital and community health service budgets and cash limited (Ibid.).
In addition to GP-fundholding, patient referral to hospital specialists made by GPs is another important rationing mechanism on the supply-side. This task of GPs is often referred to as their ‘gatekeeping’ role. Unlike many other countries, NHS patients do not have direct access to specialists other than in special circumstances, like emergency situations (EOHCS 1999).
This section explores a very controversial and heavily debated rationing device in the UK in more detail: waiting lists and waiting times. The UK’s NHS is probably the best known example a public health care system with very much government involvement. Rigid supply-side regulation with closed end budgets effectively restricts supply and therefore delivery of care (Brouwer 2007b). The rationale of such a system is primarily based on cost-containment motives. An important consequence of such stringent capacity (and price) restrictions is however, that (financially) unrestricted demand (a common feature of such systems) will exceed supply. Naturally, this will result in queuing in the form of waiting times and waiting list. Queuing is a distinct form of implicit rationing and is usually seen as a very effective and potentially equitable rationing tool in the absence of prices as rationing mechanism, provided that those with more severe conditions are given priority relative to those with less serious ailments (Purdy 1996). De Gooijer gives the following straightforward definition:
A waiting list for hospital admission (or waiting time for a first hospital outpatient appointment) is an instrument which seeks to square the circle of supply and demand. It is tempting to describe it as a valve by which pressure is dissipated. It comprises people whose cases are not urgent and life threatening so that those who need immediate treatment can receive it. So we are talking of rationing a proportion of the total demand. (2000:11)
It is not surprising that the NHS (having in mind its characteristics) is probably best known, or to put it more directly, most notorious for its relatively high waiting times and waiting lists (see table 3.5). In the UK, approximately a million people are waiting for elective surgery at any time (Martin & Smith 1999). Because fine-tuning the system in order to reach the optimal waiting time (which probably exist, see for example Siciliani & Hurst 2004) is a very difficult exercise, (groups of) people will find their ‘rescue’ by other means. The private sector in the UK and undergoing treatment abroad are examples (see for instance the recent Watts ruling of the European Court of Justice).
To illustrate the size of the by many perceive largest problem of the NHS, tables 3.6 and 3.7 present some empirical data. When interpreting these figures, it must be borne in mind that it is likely that waiting time varies widely between regions (Martin & Smith 1999). Table 3.6 shows that despite the slight reduction in total queue sizes, waiting lists for some specialities, like Urology, have grown in the indicated period.
Successive UK governments have always acknowledged the excessive waiting times and waiting lists as a major problem of the NHS and introduced several measures to tackle disproportionate waiting. A maximum waiting time of two years (later 18 months) became part of the Patient’s Charter in the early nineties (see also chapter 4). The Labour government elected in 1997 continued this policy by promising a reduction of 100.000 people waiting within the course of its administration and did in fact meet this target (EOHCS 1999, Oliver 2005). Table 3.7 indicates that although rising until March 1998, less people were waiting after 2 years of queuing policy (including a maximum waiting time of one year). However, even though the amount of people waiting between 12 and 18 months has fallen significantly since March 1998, still a substantial amount of people fell under that category (Ibid.). Some academics argue that these issues may be the result of counterproductive effects of introducing maximum waiting times (see for example Goddard & Tavakoli 1998).
The effects of these policies on waiting time of patients admitted and waiting time of patients on the list are graphically shown in figures 3.3 and 3.4. Whereas the latter have fallen substantially, the former remained quite the same.
More recently, the government determined a maximum inpatient wait of 18 weeks from referral by a general practitioner (GP) to treatment by the end of 2008. However, current policy aims at treating as many people as possible within seven weeks. Greater patient choice of provider at the point of referral and higher expenditures on health (see section 3.2) should contribute in meeting these ends (Oliver 2007). Recent research resulted in the following outcomes on inpatient waits (Ibid.):
|Number of people waiting for over 26 weeks for treatment||Number of people waiting for over 20 weeks for treatment||Number of people waiting for over 13 weeks for treatment|
The amount of people waiting for inpatient care in January 2007 was 774,000. According to Oliver (2007:3) this number “was one of the lowest since comparable records began in 1988”. Overall, these figures (except those for 26 weeks or more one could say) appear to be quite stable. The government probably expected more but remains confident that it will meet the set goals within the indicated time period (Ibid.). Further research should clear things up in about one year.
Some equity considerations
The UK tries hard to avoid inequity and tries to eliminate economic health differences. A policy statement in a recent report entitled ‘tackling health inequalities 2002: cross-cutting review’ underlines this with the following words:
For us, it is unacceptable that the opportunity for a long and healthy life today is still linked to social circumstances, childhood poverty, where you live, what job you do, how much your parents earned, your race and your gender. Our vision is of a country in which everyone has the same chance of good health, regardless of where they live or their social circumstances. To achieve it, we must tackle health inequalities where they occur now, and break the inter-generational cycle to prevent inequalities in future (Ministry of Health 2002:i).
This means there is a commonly held position at the UK governmental level that the poor, disadvantaged and chronically sick should not be financially ruined or socially excluded from health care because of their life circumstances, and that this is accepted by society.
Hurst et al. have done research about the generalist physicians’ perspectives on resources allocation and its consequences in Norway, Switzerland, Italy and the UK. The UK performed worst with a 7.2 score on a scale ranged from 3 to 15. The physicians in the UK reported less equity than physicians in other countries. They also reported more pressure to ration and reported more adverse events attributed to scarcity (Hurst et al. 2007). Most respondents (78.7%) reported that at least one group of patients was more likely than others to be denied beneficial care on the basis of cost in their health care environment. The most frequently identified groups were respectively patients who are mentally incapacitated, patients who require chronic care, illegal immigrants, and patients who are old (Ibid.).
These results are quite remarkable, since equity is considered a strong point of taxed based health care systems. Revenues are drawn from general tax receipts so individual contributions to the costs of health care are dependent on their tax payments. The rich subsidise the poor and the sick do not pay more than the healthy. There is also no problem of determining whether an individual is entitled to health care services, as the whole population is automatically covered (ASI 2000).
Coming back to the equity considerations with regard to rationing in the UK, several issues come to fore. For example, one intuitively will raise questions regarding equity when confron-ted with the aforementioned Child B case. This is also true for user charges. Recently, Lexchin and Grootendorst (2004) surveyed literature on this matter from a range of countries including England. They concluded by stating: “Virtually every article we reviewed supports the view that cost sharing through the use of co-payments (charges) or deductibles decreases the use of prescription medicines by the poor and the chronically ill”. Although many patients are (partly) exempt from (prescription) charges dependent on their age, receipt of various benefits, pregnancy status and degree of disability or medical exemption criteria (see above and also Watson 2005), it is nonetheless possible – as was concluded from the RAND health insurance experiment in the US – that user charges eventually might lead to deteriorated health outcomes for lower income people with bad health, which will probably not be accepted by society for equity reasons if these consequences were more visible. This argument is supported by evidence fr