BMA response to the Interim Report of the Health Trends Review
25 January 2002
Introduction
1. The Interim Report of the Health Trends Review [the Wanless report **] was published on 27 November 2001. The BMA has welcomed the report, which echoes some of the findings of its own year long review of health care funding published in February 2001 [Full text]. There have been numerous previous attempts to determine the level of resources which need to be allocated to health care in the United Kingdom over the last twenty years and the BMA has itself been involved in many of these. In 1985, for example it published a discussion document looking at the resources needed to meet demand over a ten year period. This concluded that health care expenditure needed to grow at 2.5% per annum in volume terms (that is after service specific pay and price inflation) if demand was to be met. Interestingly, this equates to a real increase (after general inflation) of 3.7% per annum or a little above the average rate experienced over the last two decades. The fact that this growth has manifestly not met demand is illustrative of the problems facing those who seek to forecast such trends.
**The text of the review can be found on the Treasury web-site
Our response
2. In responding to the Interim Report we have concentrated on the issues raised in it as questions for consultation rather than use it as a vehicle for reiterating BMA policy in this area. Some of the points we raise are methodological and others are of a more general nature. We hope the response will prove helpful to the Health Trends Review Team in preparing its final report.
General approach
3. The Review has been charged with considering levels of resources required for the health service over the next twenty years. This places the emphasis on cost. As we suggest when dealing with technological change elsewhere in this memorandum, it would be wrong to see health expenditure solely as a cost to the economy. The Interim Report implies (paragraph 2.3) that health care expenditure is in many ways an investment and identifies some of the financial benefits which could flow from better health outcomes. Not only does the NHS contribute to the maintenance of a healthy population and thus workforce; it is in itself a major source of employment and contributor to economic growth through the multiplier effect. We would like to see the final report include some more precise estimate of the benefits of increased health care expenditure so that the Health Trends Review can work towards something resembling a net cost approach.
Funding methods
4. Although its terms of reference obliged the Health Trends Review to consider the resources required for a publicly funded, comprehensive service based on clinical need rather than ability to pay, the Interim Report has rightly considered whether the method of funding is itself a factor in determining the required resources. It has concluded (paragraph 2.21) that 'there is no evidence that any alternative financing method to the UK's would deliver a given quality of health care at a lower cost to the economy.' It also concludes that alternatives do not appear to offer scope for increased equity though they might offer more choice.
5. Our own review also looked at alternatives to the present funding arrangements. It concluded that the way that a country’s healthcare system is funded is largely a matter of historical accident. Where universal or near-universal health cover is the aim, the choice of method lies between funding from taxation or funding through compulsory social insurance. There is little to choose between these two methods, although there is some evidence that funding from taxation is slightly more efficient and equitable. Any movement from one system to another has tended to be in the direction of general tax funding and away from social insurance. Furthermore, the method of funding appears to have little impact on the outcomes of the health system, or on patient satisfaction. These are influenced far more by the level of funding and the efficiency of the delivery mechanisms employed within the country concerned.
Expectations
6. In research undertaken for our own review we found that the concept of a healthcare system which is primarily free at the point of use and provides equal access to the same standard of care for all was strongly supported. However, there was also a growing awareness that the principle of equal access was under strain. On the basis of our research, the public seems firmly committed to retaining and strengthening the core principles and values of the NHS. Although delivery is seen to be failing, and reducing waiting times is therefore a high priority for funding, the underlying concern is that standards of care, and access to treatment, are not perceived to be consistent throughout the country. There was therefore a clear commitment to equity.
7. We differentiated between the concepts of universality (providing services to all) and comprehensiveness (providing all possible services). We concluded that these twin goals were mutually exclusive given current and likely future funding levels. The Interim Report concludes that on equity grounds clinically necessary treatment should not be available on grounds of ability to pay. It argued however that non-clinical services might be available in a way that provided more choice, It also gave NICE the major role of identifying what it was clinically necessary for the NHS to provide. For our part we came down on the side of universality as against comprehensiveness. There will always be treatments which the NHS cannot provide to everybody. At present the examples are few - tattoo removal and infertility treatment are often among those cited. There are undoubtedly more which at present are identified only by unacceptably long waiting times. Conversely, there are treatments provided which perhaps should not be. These can for example be treatments of marginal effectiveness (i.e. slightly better than alternatives but only for a limited number of patients or in a purely theoretical way) and high marginal cost. Alternatively, they can result from therapeutic drift (gradually extending the use of a treatment beyond those who really need it).
8. The Interim Report does, not we feel, deal adequately with the issue of comprehensiveness. We would like to see it and particularly the issue of implicit rationing dealt with in the final report since the distinction between clinical need and non-clinical services is an incomplete one.
9. We note also that there is no mention of the role of the alternative or complementary sector. Although, there may be little robust data on the amounts spent in this sector, it would be useful to see if there were marked international variations and whether this had any relationship with spending on conventional health services.
Delivering high quality
10. There are four main issues here:
- The cost and benefits of delivering the principal National Service Frameworks
- The cost of capital investment in hospital facilities and information technology
- The cost of clinical governance and its effects on non-conformance
- Patient centred care and choice
12. There are difficulties in assessing the likely impact of the third and fourth however. Patient requirements seem, according to the Picker Institute research highlighted by the Interim Report, to involve increased exposure to doctors' time. Clinical governance on the other hand will require more of that time to be used for other purposes and the costs will be 'time costs.' Whether it is feasible for clinical governance to take up 10% of clinical time and still move towards meeting patient expectations is debatable. Furthermore, it is doubtful whether the increased cost of litigation is a sufficient proxy for the costs of non-conformance. Longer than necessary length of stay and some more direct measure of poorer outcomes would seem to be equally strong candidates.
13. The Review should not lose sight of input quality. In this respect, the training of doctors should also be seen as an integral part of a high quality health service. This has implications for the level of resources both direct and indirect devoted to postgraduate training from within health budgets.
Demography and future trends in morbidity
The Interim Report identifies a number of key considerations in relation to demography. First among these is a rather lower impact of changes in the age-distribution of the population on health care expenditure compared with previous studies. This, the report indicates, will be because its methodology will separate the impact of an ageing population from that of proximity to death. It is the latter which imposes the greater cost pressure and past studies have conflated the two. The second issue is compression as against expansion of morbidity as life expectancy increases. The report rightly points up the widely different cost implications of these alternatives and the ambivalent research base. The third issue revolves around differential access to health services. There is evidence of age discrimination in access to health services and our own review cited evidence (from Scotland) of lower access to some services by those in the most socio-economically deprived groups.
14. There are a number of other considerations which the Review could usefully examine in the area of demographic change.
- The prospect of smaller numbers of taxpayers facing a burden of supporting greater numbers of dependants has led many to question the future of tax-funded welfare in general and healthcare in particular, although the dependency ratio may also be affected by changing attitudes to work, education and training.
- The proportion of women in the workforce is predicted to increase, and this may indicate that the trend towards later births will continue. Delaying pregnancy can lead to greater complications, suggesting a need for more complex and specialist maternity medical and obstetric services.
- The effects on the supply of unpaid care of changes in family structure brought about by falls in birth rates, higher divorce rates, re-marriage, greater family mobility and less living together of families across generations. This was a particular concern of the Royal Commission on Long Term Care
- The effect of increasing disparities in income distribution in the UK which have been identified since the 1970s, specifically in the growing disparity between those with well-paid work and those without. Increasing flexibility in employment practices are expected to lead to reduced job security punctuated by periods of unemployment for many people. The link between income and health has been well documented but specific groups of the population are also more likely to be poor, for example those over 70, lone parents (who are almost overwhelmingly women), couples with three or more children and households headed by a member of an ethnic minority group as well as refugees
Methodology
15. The Health Trends Review has approached the identification of the cost of technological change through a top-down approach in line with the Project HOPE methodology used for this purpose in the United States. Briefly, cost increases due to technological change are estimated as the residual after demographic and inflationary elements have been removed from total costs. The US system is, however, a fee for service one and the residual isolated there is more accurately identified as real changes in the per capita use of health services - a measure which reflects the effects of technology. A change in the volume of fee for service items per head may be a function of availability of treatments or simply of reimbursement practice. Nevertheless, the assumption that it is affected by changes in technology is broadly sensible. Whether this conclusion extends to the UK where the residual contains the volume increase in inputs including staff numbers is not as clear cut. We would like to see further analysis aimed at isolating technological change from other elements of the residual.
16. Notwithstanding its size, the Interim Report has expressed surprise at the tendency for technological change to perpetually increase costs. This is not as surprising as it may appear. The overall impact of technological change can be viewed as an envelope curve subsuming individual technologies each of which has the capacity to become less expensive over time. The problem is that at the same time as the real cost of one technology begins to decline that of a new one is increasing. A good example of this is information technology. The role of NICE in determining which technologies are sufficiently cost-effective as to be publicly funded will be a powerful influence on the continuation of this pattern in the future.
Cost drivers
17. The Interim Report has asked whether it has correctly identified the main drivers of future increases in the cost of technology. In our own review we categorised the main areas of technological development likely to affect healthcare as:
- advances in genetics; identifying genes causing diseases such as cystic fibrosis and genetic susceptibility to disorders such as coronary artery disease
- developments in biotechnology which will enable advances in genetics to be exploited development of bioengineering to produce artificial body parts and organs, predicted to replace transplantation within the next three decades
- further developments in minimal access and image guided surgery, exploiting developments in magnetic resonance imaging
- use of robotics in surgery, increasing accuracy and consistency, and in rehabilitation
- further developments in transplantation, with more success in controlling rejection
- IT and telecommunications, including decision support systems, developments in medical record keeping and transfer and its impact on public access to information.
19. With the increasing sophistication of technology, expertise is likely to be concentrated within a smaller number of larger centres. The Review has not covered potential changes in the organisation of the hospital service. In 1997, the BMA's Health Policy and Economic Research Unit conducted a study of possible future models for acute hospital services. This analysis examined five possible options for serving a population of two million people, ranging from a single regional provider of services to a rationalisation of existing provision with fewer providers, utilising current sites and where possible concentrating services. It concluded that a reconfiguration of services on a smaller number of sites would be realistic and acceptable, particularly a model which has five district hospitals and three additional hospital units. The rationalisation of current sites to a smaller number is likely to create greater efficiency and to benefit from some economies of scale with least disruption to local communities. This needs to be balanced against the provision of services to remote and rural areas. We believe that the Review will need to take some view on service configuration.
20. As we suggested earlier, one issue, which the Review does not address as fully as we would like, is that of cost effectiveness. While technology increases spending, the health benefits tend to more than justify the added costs. Many of the benefits do not accrue to the health care system although some (e.g. more years of healthy life) clearly do. Either way, some means of highlighting more general cost savings and quality of life gains would seem appropriate.
The future workforce
Numbers of doctors and nurses
21. The Interim Report deals with a number of workforce issues principal among which are:
- The low numbers of doctors and nurses relative to other countries and the likely continuation of this position over the next 20 years.
- The likelihood of a shift of work from doctors in training to fully trained specialists
- The likelihood of a shift in work from doctors in general to nurses
- The extent to which doctors are not presently involved in patient contact and the impact of this on productivity
23. We have three main observations to make on the Review's work in this area to date. First we believe the Interim Report has paid insufficient attention to the likely decline in medical participation rates over the next twenty years. The increasing numbers of women doctors, the impact of the European Working Time Directive and the desire for shorter working hours among all doctors will need to be factored into calculations on the potential gap between supply and demand. Secondly, the number of nurses available to feed skill mix changes is a crucial variable. The Review team should undertake some sensitivity analysis to determine whether the likely numbers will be available and the consequences of their not being so. Thirdly, the trend towards greater regulation of doctors will undoubtedly continue. This may reduce rather than increase contact time and restrict the potential for productivity gains.
Pay
24. The Interim Report's observations on relative pay levels echo much that the BMA has said to the Review Body on Doctors' and Dentists' Remuneration over the last two decades. We would however quarrel with the report's conclusion that the pay of doctors has largely maintained its relative position in the overall earnings hierarchy. The report's own data belie this finding. The pay of career grade doctors, it claims, has declined from the 94th to the 90th percentile since 1980-84. The steepness of the earnings gradient at these levels means that this is equivalent for non-manual males to a 28% difference in earnings. Furthermore, the period between 1980 and 1984 was one of significant relative decline and the peak level of doctors' earnings in 1980 coincided broadly with the 97.5th percentile. The gap between this and the 90th is of the order of 60%. Given the relative decline in nursing pay also, we believe that there will need to be major restructuring of pay levels in the period covered by the Review and that this will have major implications for funding levels. Without such increases, the required staffing levels and associated productivity increases are unlikely to be forthcoming. This may be a step change and will have timing implications as well as resource ones. Clearly, the more extended the process of restructuring the greater the capacity of the service to absorb it.
Differences within the UK
25. We note the intention of the Review to explore the implications of the research on social deprivation and resource allocation for the constituent countries of the UK. The Interim report asks also about the English regions. In our view, the resource allocation formulae operating in England are moving steadily towards a more equitable distribution of resources although we believe there is scope for more work on the relationship between primary and secondary care (in particular the resource implications of treating them as substitute or complementary services).
26. The level of health expenditure by the constituent countries is however determined politically once the various 'block' allocations have been made under the Barnett formula. Thereafter, the funds are distributed by formula. The Review could usefully look at the overall commitment of resources in the various countries relative to need but it will need to bear in mind that the purpose of the Barnett formula is to give the countries the power to determine their own public expenditure priorities. The fact that Scotland, Wales and Northern Ireland choose to spend more on explicit health expenditure clearly reflects greater levels of morbidity. As has been pointed out on many occasions however, health gains are not only a function of health expenditure and the Review might wish to look at the levels of expenditure in a wider context. It should also bear in mind that one consequence of the Barnett formula is to effect convergence of block expenditure per head over time among the constituent countries and that present differences, overall and in health expenditure, may not survive into the long run unless funding arrangements alter.