athy: it might more generally benefit patients, by more clearly focusing the process of obtaining and analys- ing evidence on the performance of different health.
British Homeopathic Journal (2000) 89, Suppl 1, S20±S22 ß 2000 Macmillan Publishers Ltd All rights reserved 0007±0785/00 $15.00 www.nature.com/bhj
Assessing the cost-effectiveness of homeopathic medicines: are the problems different from other health technologies? M Buxton1* 1
Health Economics Research Group, Brunel University, Uxbridge, Middlesex UB8 3PH, UK
Evidence-based medicine and health technology assessment All health care systems around the world are facing the problems of an increasing range of technologies competing for positions within the packages of care that can be provided to health care recipients. Health care systems based on private insurance and national systems based on public funding are alike in needing to ®nd ways to limit health care provision to that which can be afforded within the limited funds available.1 This necessity is an unavoidable reality. What is open to discussion are the criteria which should be used to decide about what health care to provide, and what not to provide, for particular groups of patients.2 The criteria may, of course, be implicit and unspoken, with the decisions made behind closed doors. The decisions may re¯ect the political power within the health care system and maintain the status quo. Budget allocations, and the health care provision they permit, may simply re¯ect historic decisions with budgets rolled forward (or in hard times rolled back) from previous allocations. Alternatively, they can be more explicit and potentially open to challenge and debate. Increasingly, the currency for that debate is not opinion, precedent and prejudice, but scienti®c evidence of clinical effectiveness and cost-effectiveness. Two parallel movements have gained considerable momentum. The growth of the evidence-based medicine (EBM) movement has focused the attention of practising clinicians on the need to use to best advantage the existing evidence base for the clinical decisions they make concerning the treatment of individual patients.3 EBM emphasises the prime role of good evidence from well designed trials, or better
*Correspondence: M Buxton, Director, Health Economics Research Group, Brunel University, Uxbridge, Middlesex UB8 3PH, UK.
still from systematic reviews of multiple trials, as a basis for judgments as to the likely outcomes of different treatments. In parallel the growth of formal health technology assessment (HTA) is increasing the relevant evidence base. HTA provides a systematic framework for addressing not only the effectiveness, but also the cost-effectiveness, of new and existing technologies through primary studies and secondary research.4 These movements are international. For example, Cochrane Review Groups are undertaking reviews of trial evidence on a very wide range of interventions and are revealing where good evidence exists and where it is lacking.5 Many countries now have formal arrangements for technology assessment and national technology assessment bodies cooperate internationally. The UK now has a very signi®cant programme of technology assessment, most of it focused on assessing cost-effectiveness and so requiring consideration of both costs and patient outcomes.6 The NHS HTA programme aims to identify, on behalf of the NHS, the most important topics for evaluation and then commissions high quality research to address these questions. Few proponents of particular technologies are totally enthusiastic about this need to provide good evidence. Pharmaceutical manufacturers, who are now well accustomed to the necessity for licensing purposes to provide evidence of ef®cacy, complain that the requirements add signi®cantly to the costs of producing new drugs, and delay their market entry.7 Now, as a number of countries additionally seek information on cost-effectiveness prior to making drugs available within the publicly funded system, the pharmaceutical industry complains that these requirements impose additional burdens that may discourage future research and development. But they get on and provide the evidence. Similarly, proponents of other technologies plead special cases. As the attention swings towards the cost-effectiveness of many devices, manufacturers
Assessing cost-effectiveness M Buxton
protest that the economics of the relatively small-scale production of many devices means that they do not have the resources or pro®t margins to enable them to begin to undertake such studies. Surgeons have argued that surgical techniques are not easily amenable to randomised trials, and where trials are undertaken true effectiveness can be obscured by problems of the learning curves through which surgeons individually and collectively discover how best to perform particular surgery and how best to manage surgical patients.8 There certainly are challenges involved in undertaking health technology assessments, but equally both the methodology of assessment and the organisational arrangements for evaluation are continually being developed and re®ned. For example, in terms of methodological development, there is increasing experience and understanding of the value of large multi-centre pragmatic (rather than explanatory) trials. In such trials, subject to minimal protocol restrictions, patients can be treated in a manner that provides a reasonable indication of the effectiveness of the technology in normal use rather than a measure of its ef®cacy in highly controlled circumstances. Institutionally, the advent of the National Institute for Clinical Excellence, for example, offers a context in which good evidence, where it exists, can be incorporated into formal clinical guidelines.9 Where there is no good evidence of cost-effectiveness, the diffusion of technologies can be discouraged until such evidence becomes available.
The priorities for homeopathy My question then, for the proponents of homeopathy, is in what way, if any, is homeopathy special or different? Should it be treated in exactly the same way as the rest of the competitors for scarce health care pounds or dollars? I can only guess at your views. My own conclusion, from my admittedly limited but growing knowledge of homeopathy, and a very extensive experience of collaborating on assessments of technologies and undertaking numerous cost-effectiveness studies in health care, is that homeopathy is not fundamentally any different, and should not expect to be treated differently. In which case, what are the priorities for building an evidential base? The ®rst point I would make is addressed to those who seek to promote homeopathy through simply looking at costs. It is an empirical question, for which there is currently little evidence, as to whether patients treated with homeopathy generally cost more or less than equivalent patients treated with `conventional' therapies. We could embark on costing studies and might conceivably ®nd that homeopathy is cheaper. But it is my conviction that it is not primarily issues of cost that currently prevent the wider acceptance of homeopathy. I believe that the primary problem here, as in other areas where there is a
reluctance to provide a particular form of care, is the lack of good evidence on effectiveness.10 In this context, no amount of evidence on costs alone is likely substantively to change that perception. Critics can always point to a variety of forms of ineffective care that are cheap: not least, no care at all! I would argue that the ®rst priority must be to provide sound evidence of effectiveness, and therefore you might well argue that it is not the help of health economists that is needed. It is possible however that the economist's perspective may offer some help, in the form of the economist's traditional focus on establishing whether patients judge themselves to be in a preferred state following an intervention. The now well-rooted idea that we should measure and focus on patients' own perceptions of their quality of life and well-being, in terms of generic and comparable measures of patients' health-related quality-oflife and, more recently, measures of patients' utility, may provide a more appropriate way of measuring effectiveness of homeopathy than do some of the more traditional and limited measures of clinical outcome.11 Patients' assessments of their own utility may re¯ect a wide range of factors other than narrowly de®ned clinical effectiveness. They will certainly also take account of their concerns or lack of them about concurrent side effects, and their current utility may also be affected by the possibility of currently unrecognised future risks. In addition, the process of care is certainly important to patients. In both respects, some non-traditional medicine may offer greater patient utility.
The role of conjoint analysis Within health economics, studies employing the techniques of conjoint analysis (CA) are increasingly being undertaken to identify what characteristics of treatments patients value, and what relative values these characteristics have.12 CA has been extensively used in areas other than health care as a means to understand how respondents value different aspects of a good or service. Typically it requires respondents to make choices between pairs of options with different combinations of characteristics. In health care, conjoint analysis is being used to look at the trade-off that patients state they are willing to make between process and outcome variables.13,14 With Drs van Haselen and colleagues from the Royal London Homoeopathic Hospital, the Health Economics Research Group at Brunel University is about to undertake a study that will use conjoint analysis to address the question of the relative values patients place on different attributes of their treatment for asthma. In this study, we will be comparing the values expressed by a group of patients who have opted for homeopathic medicine as well as a group of patients who are being treated in a convenBritish Homeopathic Journal
Assessing cost-effectiveness M Buxton
tional therapy clinic. This may of course show that patients who have opted for a different setting do have a different set of values to those who have not. Equally it may show that both groups of patients have very similar relative values which are differentially being met by the different therapies. Such information can then feed back into the process of appropriately de®ning outcome measures and process variables to be measured in future trials and studies. Then, once there is appropriate evidence on patient outcomes, the questions of cost become relevant. It is then possible to establish whether the technology is dominant (or dominated): that is, it is both cheaper (more expensive) and with better (poorer) overall outcomes, in which case the evidence is clear. If not, it will be necessary to measure its incremental cost-effectiveness ratio (ICER): that is, the cost per additional unit of patient bene®t, and then to argue that, comparative to other uses of additional resources, this ICER is acceptable.
Conclusions It appears to me that, with respect to the needs for good evidence, homeopathy is no different from other areas of medicine, and cannot expect to be exempted from the need to show its cost-effectiveness. Homeopathy cannot divorce itself from the wider movements in health care. It must recognise that, if it is to survive and ¯ourish as part of mainstream modern health care, it bears the onus as the proponent of a particular technology to provide, or at least to help provide, the necessary scienti®c evidential basis to justify its use. Special pleading will not be accepted! The converse is that with such evidence there is a real opportunity to change opinions and patterns of provision. What is more, there is an opportunity to take a lead in addressing important questions about what patients value with regards to their health care. Such an analysis might not only be to the bene®t of homeopathy: it might more generally bene®t patients, by more clearly focusing the process of obtaining and analysing evidence on the performance of different health
British Homeopathic Journal
interventions on the factors that patients consider important.
References 1 Buxton MJ. Scarce resources and informed choices. In: Ashton D (ed.) Future Trends in Medicine (Congress and Symposium Series no 202. London: RSM, 1993. 2 New B, Le Grand J. Rationing in the NHS: Principles and Pragmatism. London: King's Fund, 1996. 3 Sackett DL, Rosenberg WM, Gray JAM, Haynes RB, Richardsom WS. Evidence based medicine: what it is and what it isn't. BMJ 1996; 312: 71±72. 4 Advisory Group on Health Technology Assessment. Assessing the Effects of Health Technologies: Principles, Practice, Proposals. London: Department of Health (Research and Development Division), 1992. 5 Chalmers I, Enkin M, Keirse MJ. Preparing and updating systematic reviews of randomised controlled trials of health care. Milbank Q 1993; 71(3): 411±437. 6 National Coordinating Centre for Health Technology Assessment. The Annual Report of the NHS Health Technolgy Assessment Programme 1998. London: Department of Health, 1998. 7 DiMasi JA, Hansen RW, Grabowski HG, Lasagna L. The cost of innovation in the pharmaceutical industry. J Health Econ 1991; 10: 107±142. 8 Horton R. Surgical research or comic opera: questions, but few answers. Lancet 1996; 347: 984±985. 9 NHS Executive Faster Access to Modern Treatment: How NICE Appraisal Will Work: A Discussion Paper. Leeds: Department of Health (Pharmaceutical and Prescribing Branch). Int J Technol Assess Health Care 1999; 15: 340±351. 10 Van Haselen R. Evidence in¯uencing British Health Authorities' decisions in purchasing complementary medicine. J Am Med Assoc 1998; 280(18): 1564±1565. 11 Gold MR, Seigel JE, Russell LB, Weinstein MC. Cost-Effectiveness in Health and Medicine. New York. Oxford University Press, 1966, Chap 4. 12 Ryan M. Using Consumer Preferences in Health Care Decision Making: The Application of Conjoint Analysis. London: Of®ce of Health Economics, 1996. 13 Bryan S, Buxton MJ, Sheldon R, Grant A. Magnetic resonance Imaging for the investigation of knee injuries: an investigation of preferences. Health Econ 1998; 7: 595±603. 14 Ratcliffe J , Buxton MJ. Patients' preferences regarding the process and outcomes of high technology medicine: an application of conjoint analysis to liver transplantation. Int J Technol Assess Health Care 1999; 15(2): 340 ± 351