- Email: [email protected]
Economic evaluations of drug therapy: attitudes of primary care prescribing advisers in Great Britain
Health Policy 41 (1997) 61 – 72
Economic evaluations of drug therapy: attitudes of primary care prescribing advisers in Great Britain Tom Walley a,*, Stuart Barton a, Jonathan Cooke b, Michael Drummond c a
Department of Pharmacology and Therapeutics, Uni6ersity of Li6erpool, 70 Pembroke Place, Li6erpool L69 3GF, UK b South Manchester Uni6ersity Hospitals NHS Trust, Manchester, UK c Centre for Health Economics, Uni6ersity of York, York, UK Received 21 February 1997; accepted 25 March 1997
Abstract All health authorities in Great Britain have both medically or pharmaceutical qualified staff to advise both the authority and the local primary care medical practitioners about drug use and prescribing. This study used a piloted postal questionnaire to assess the attitudes of these advisers to economic evaluations of drug therapy, and their perceptions of the barriers to achieving cost effective prescribing by use of these evaluations. There was a 65% response rate to the questionnaire. Economic issues were rated by advisers to be less important than clinical issues, but were considered at most meetings between advisers and primary care medical practitioners. Advisers wished to consider true cost effective prescribing but often felt obliged to consider drug acquisition costs and risks of budgetary overspends. The perceived inflexibility of existing structures within the British National Health Service and the lack of credibility of the evaluations (often perceived as pharmaceutical industry marketing) were the major barriers to the application of the evaluations. The paper concludes that advisers were keen to use economic evaluations to promote cost effective prescribing but were impeded by the perceived bias of existing studies and by rigid current NHS structures. © 1997 Elsevier Science Ireland Ltd. Keywords: Economic evaluation; Drug therapy; National Health Service; Attitudes
* Corresponding author. Tel.: +44 151 7948125; fax: + 44 151 7948126; e-mail: [email protected] 0168-8510/97/$17.00 © 1997 Elsevier Science Ireland Ltd. All rights reserved. PII S 0 1 8 - 8 5 1 0 ( 9 7 ) 0 0 0 1 3 - 4
62
T. Walley et al. / Health Policy 41 (1997) 61–72
1. Introduction Faced with a rising primary care drug bill and uncertain of the value for money in spending on prescribing, the British government has introduced a range of initiatives, including the indicative prescribing scheme and the prescribing incentive schemes [1]. General practice fundholding is a related reform which also limits prescribing cost rises [2]. The stated aim of these initiatives was not cost containment, but improving the quality of prescribing and the value for money from prescribing [3]; hence there have also been educational initiatives, such as the production of drug information bulletins. To administer these initiatives and to advise both general medical practitioners and their health authorities on prescribing, the government funded a medical prescribing adviser within each local health authority [4]. There are now often pharmacist advisers either in addition to or instead of the medical advisers. The government stated that pharmaceutical companies could help to improve the use of resources in prescribing by providing evidence of the cost effectiveness of their products, in particular new drugs [3]. Companies were quick to seize on this, and many produced economic evaluations of their products [5]. Economic evaluations compare alternative treatments, including drug therapy, in terms of their costs and consequences. In theory, such evaluations could help to determine the best use of health service resources, particularly within prescribing. To the average prescriber however, such evaluations have little meaning, since he or she may bear no true responsibility for resource utilisation. In contrast, such evaluations could be of great value to a health authority or to a general practice fundholder, who have responsibility for budgets covering many areas of health services. To be of use, these evaluations must be reliable and easily understood by users. Also there must be scope for users to act on the basis of the study findings. We surveyed health authority primary care prescribing advisers on whether and how they used economic evaluations in their advice on prescribing to both general practitioners and their health authorities.
2. Methods A focus group meeting between one author (TW) and a small number of prescribing advisers generated hypotheses concerning their use of economic evaluations to be tested in this study. Based on these, a questionnaire was developed which covered four main themes: knowledge of economics, the importance of economic and financial issues, barriers to the use of economic analysis and awareness of published studies. This questionnaire (appendix) was piloted and subsequently sent by post to all 273 and pharmaceutical prescribers in England, Wales and Scotland in July 1995. These were identified using the database
T. Walley et al. / Health Policy 41 (1997) 61–72
63
of the Medical Advisers’ Support Centre. The questionnaire was anonymised and no follow up of nonresponders was undertaken. To test awareness of published studies, advisers were given a list of some recently published studies [6 – 15] and asked whether they had read the study and whether it had altered their advice to their authority or to general practitioners. The validity of the answers to this part was assessed by the use of two plausible control studies [16,17], based on clinical papers which in fact had contained no health economic assessment. Results were analysed by SPSS.
3. Results A total of 178 responses were received (65% response rate). Response rates and responses were similar from both medical and pharmaceutical advisers and therefore no distinction between the two is made in the results. There was no systematic difference in the age, sex or experience of responders and nonresponders. The advisers had been in post an average of 33 months. A figure of 40% claimed to have some formal training in health economics, for the most part a distance learning package produced especially for this group [18]. The responses for those claiming such training were similar to those without and again no distinction is made in the results.
3.1. Importance of economic issues Only 5% considered economic considerations the most important issue in deciding how to advise general practitioners. Most (91%) considered economic issues important but secondary to issues of clinical effectiveness. A total of 92% of advisers discussed economic issues with general practitioners in most of their visits to practices. Only 4% considered economic factors totally unimportant. In advising their authorities, the pattern of response was similar: 11% thought economic issues the most important, and only 2% considered them unimportant, with most again believing that economic issues were secondary to clinical issues.
3.2. Economic issues considered There were differences between what economic issues advisers actually considered in discussions on prescribing either with general practitioners or with their health authority (drug acquisition costs and risks of budgetary overspends), and what advisers would ideally like to consider (cost effectiveness of drugs in the wider health service) (Table 1).
64
T. Walley et al. / Health Policy 41 (1997) 61–72
3.3. Sources of information The sources which the advisers claimed to use to assess the cost effectiveness of a therapy were clinical trials in peer reviewed journals (96%), formal economic evaluations (77%), and opinions of colleagues (49%). Almost all (98%) were keen to make more use of economic evaluations of drug therapy; the advice they would wish to offer as a result of such studies was broader than just choice of drug, and included informing purchasing decisions, such as a choice between drug therapy and other services.
3.4. Barriers to application of economic e6aluations Advisers ranked a list of possible barriers to the use of economic evaluations (Table 2). The major barriers were the perceived inflexibility of existing NHS structures preventing the application of the evaluations, and the lack of credibility of the evaluations.
3.5. Awareness of current e6aluations The two control studies were claimed to have been seen by 16 and 21% of advisers, to have convinced 14 and 7%, and to have encouraged 7 and 2% to have altered their advice on prescribing. Claimed awareness of the other published studies ranged from 58 [6] to 14% [7]. The awareness of seven of the ten studies [6,9–11,14–16] was significantly higher than that of the controls, judged by a result more than two standard deviations from the predicted value. The studies most convincing and influential were those with perhaps the best clinical basis, i.e. enalapril in heart failure [8] (61% found convincing, 34% altered the advice given) and H pylori eradication in duodenal ulceration [9] (45% convincing, 34% altered advice). Another highly regarded study was an Effective Health Care bulletin on antidepressants [6] (44% convincing and 26% altered advice given).
Table 1 Key economic issues for prescribing advisers
Acquisition costs of drug Risk of budgetary overspend for GP Risk of budgetary overspend for health authority Cost effectiveness-better clinical outcomes as a compensation for higher drug costs Cost effectiveness-savings elsewhere in the NHS as a compensation for higher drug costs
Currently consider as a factor (%)
Ideally would consider as a factor (%)
87 70 64 68
8 14 20 41
35
T. Walley et al. / Health Policy 41 (1997) 61–72
65
Table 2 Obstacles considered most important to the use of cost effectiveness studies of drug therapy in the NHS 1. Studies unrealistic because existing NHS structures prevent moving resources between budgets (e.g. from secondary care to prescribing in primary care) 2. Industry funded studies not credible 3. Studies too open to bias because of need to make assumptions 4. The agenda of the Department of Health is not long term cost effective prescribing, but short term containment of the drug bill 5. Personal difficulty in interpreting the studies 6. Studies unreal-‘savings’ are theoretical and are unlikely ever to be achieved 7. Advisers or health authorities unable to free resources to follow the suggested approach to therapy 8. Health authorities as a whole don’t really understand the process 9. Difficulties (of general practitioners) in interpreting the studies 10. Advisers or health authorities unable to take a long term view-need to meet budgetary constraints now 11. Dept of Health funded studies are not credible 12. Studies relate to overall NHS costs-Advisers or health authorities only concerned with primary care drug costs 13. Inability to act about using such studies-no personal power a
Ranked according to scoring by advisers.
4. Discussion Advisers were generally keen to make best use of health service resources, and saw more use of economic evaluations of drug therapy as a means to this end. However, they felt constrained by a number of limitations, the most important of which were the rigidity of NHS structures and difficulties in reallocating NHS resources, and the perceived current unreliability of such evaluations. At present, increased spending on drug therapy in primary care cannot be funded by any savings which might result in a hospital budget (e.g. by use of inhaled steroids in asthma, or by ACE inhibitors in heart failure, either of which may reduce hospital admissions). This is largely the result of the funding arrangements for drugs in primary care, which are funded by a separate parliamentary vote from other NHS spending [19]. Furthermore to actually convert the possible savings in hospital spending, often identified in economic evaluations, into cash which can be used elsewhere, would require closing beds and cutting staff. This might not be practically possible or politically acceptable. This limits the applicability of pharmacoeconomic studies for a health authority, which instead finds such studies used, not to make better use of limited resources, but to justify demands for more resources in drug therapy. Faced with a rising drug bill and no savings elsewhere in health spending, the focus for a health authority or government with a limited budget may move to cost containment. This was reflected in the opinion of the advisers that cost containment was a higher priority for the government truly cost effective prescribing. Such an approach would not be acceptable professionally to the large majority of
66
T. Walley et al. / Health Policy 41 (1997) 61–72
advisers, and this is reflected in their rating of clinical effectiveness ahead of economic issues. The difficulty of taking a long term view also reduces the value of economic evaluations. Many studies may identify benefits over 5–10 years in exchange for increased spending now: if as is often the case, the health authority is in danger of overspending now, investment in a policy for long term savings may not be possible. General practice fundholders have in theory more freedom to move money between budgets than health authorities, but are also constrained by the need to keep within their annual budget which may prevent them from applying studies. Finally, managers are often responsible for only one portion of the overall budget, and may not be able to see the wider economic efficiency. Advisers with responsibilities only for the primary care drug budget might therefore be reluctant to see an overspend in their budget for the sake of benefits elsewhere. It is reassuring that only 14% of the respondents to our survey accepted this approach. Given the perception of advisers of pressures to contain cost, this approach may be more common than admitted. At the time of our survey, health authorities dealing with primary care and for whom prescribing advisers worked were separate from the health authorities responsible for secondary care had not formally merged in England and Wales, and this may have exacerbated this problem. These health authorities have now merged. The second major problem identified was that the studies themselves were not seen as credible, because of the nature of the study itself or its funding, or because of the lack of understanding on the part of the adviser, the general practitioner or the health authority. The interpretation of economic evaluations is new to clinicians, and hence suspicions of bias and of industry funded studies grow [20,21]. Attempts have been made to address these problems internationally [22] and in Great Britain [23]. The majority of the studies seen by the advisers had been given to them either directly or indirectly by pharmaceutical company representatives, increasing their already inherent suspicion. Few advisers were aware of existing specific British guidelines [23] (produced jointly by the Association of the British Pharmaceutical Industry and the Department of Health) for the conduct of economic evaluations, and of those who were aware, none considered these guidelines useful to them. This survey was undertaken before the recent publication of guidelines for the review of such studies by the British Medical Journal [24]. Scientific rigour of this kind will enhance the credibility of such studies in Britain. Are there ways around this impasse for economic evaluations? One comment from many advisers was that there should be a central means of validating studies for use within the NHS [25], perhaps even as part of the drug licencing or reimbursement process [26]. The NHS Centre for Reviews and Dissemination partly fulfils this role through its databases on reviews of effectiveness and economic evaluations [27], but the advisers were largely unaware of it. There is also a need to increase the training advisers in health economics, so as to allow them to become critical and intelligent readers of economic evaluations.
T. Walley et al. / Health Policy 41 (1997) 61–72
67
It is important to acknowledge the limitations of surveys of this kind: the answers given by respondents may reflect how they believe they should respond rather than what they actually do or think. Some bias may have been introduced by the non responders, although the response rate is good for a single postal survey without reminders. We did not explore the respondents’ wider understanding of health economics. The barriers identified are those perceived by those who took part in the initial interviews: no new barriers were identified in free text by respondents. How real these barriers are is not clear, since there was no evidence of any attempt by the advisers to actually use an economic evaluation to inform purchasing decisions. This survey considers only one of the possible targets of economic evaluations of drug therapy: others would include perhaps fundholding general practitioners, directors of public health and senior hospital pharmacists: the latter two have also been surveyed and will be reported separately. If economic evaluations of drug therapies are to be used, they must be credible and accessible to decision makers. However current NHS structures seem to make the actual implementation of economic evaluations relating to prescribing, even if accepted as correct, largely impossible. More consideration should be given to exploring these structural barriers and how they might be overcome to actively promote true cost effective prescribing.
Acknowledgements This work was partly funded by Duke University and by Care Research and Development Centre
Appendix See following pages.
National Primary
68
T. Walley et al. /Health Policy 41 (1997) 61–72
T. Walley et al. /Health Policy 41 (1997) 61–72
69
70
T. Walley et al. /Health Policy 41 (1997) 61–72
T. Walley et al. /Health Policy 41 (1997) 61–72
71
72
T. Walley et al. / Health Policy 41 (1997) 61–72
References [1] Walley T, Wilson R, Bligh J. Current prescribing in primary care in the UK: effects of the indicative prescribing scheme and GP fundholding. Pharmacoeconomics 1995;7:320 – 31. [2] Wilson R, Buchan I, Walley T. Alterations in prescribing by GP Fundholders; an observational study. Br Med J 1995;311:1347–50. [3] Department of Health. Improving prescribing. London: HMSO, 1990. [4] Walley T, Bligh J. FHSA medical advisers: friend or foe?. Br Med J 1992;304:133 – 4. [5] Walley T, Edwards RT. Health economics in primary care in the UK: containment of drug costs. Pharmacoeconomics 1993;3:100–6. [6] Anon. The treatment of depression in primary care. Effective Health Care Bull 1993;5:1 – 12. [7] Knill Jones R, McGhee S, McDonald F. An economic evaluation of enteric-coated and standard formulations of naproxen from a gastrointestinal perspective. Br J Med Econ 1994;7:185 – 96. [8] Hart W, Rhodes G, McMurray J. The cost effectiveness of enalapril in the treatment of chronic heart failure. Br J Med Econ 1993;6:91 – 8. [9] Jones R, Bosanquet N, Johnson NJ, et al. Cost effective management strategies for acid-peptic disorders. Br J Med Econ 1994;7:99 – 114. [10] Jonsson B, Bebington PE. What price depression? The cost of depression and the cost-effectiveness of pharmacological treatment. Br J Psychol 1994;164:665 – 73. [11] Akehurst R, Piercy J. Cost-effectiveness of the use of Nicorette nasal spray to assist quitting smoking among heavy smokers. Br J Med Econ 1994;7:155 – 84. [12] Cull R, Wells N, Miocevick M. Economic costs of migraine. Br J Med Econ 1992;5:103 – 15. [13] Knill Jones R. An economic evaluation of Arthrotec in the treatment of arthritis. Br J Med Econ 1992;5:51–8. [14] Lancaster Smith M, Gough K, Wells N, et al. An economic analysis of ranitidine versus cimetidine in the prevention of duodenal ulcer recurrence. Br J Med Econ 1992;2:25 – 36. [15] Bates CM, Richardson PDI. Cost-effectiveness of omeprazole in the management of gastroesophageal reflux disease in clinical practice. Br J Med Econ 1994;7:81 – 97. [16] Greening AP, Ind PW, Northfield M, Shaw G. Added salmeterol versus higher-dose corticosteroid in asthma patients with symptoms on existing inhaled corticosteroid. Lancet 1994;344:219 – 24. [17] Dahlof B, Hansson L, Lindholm LH, Schersten B, Wester PO, Ekbom T, Hedner T, de-Faire U. STOP-Hypertension 2: a prospective intervention trial of ‘newer’ versus ‘older’ treatment alternatives in old patients with hypertension. Swedish Trial in Old Patients with Hypertension. Blood Press 1993;2:136–41. [18] Edwards, R. Distance learning programme in health economics Vol. 1 – 6. Medical Advisers Support Centre, Liverpool, 1993. [19] Crump BJ, Panton R, Drummond MF, Marchment M, Hawkes R. Transferring the costs of expensive treatments from secondary to primary care. Br Med J 1995;310:509 – 12. [20] Horton R. Pharmacoeconomical with the truth. Lancet 1993;341:752. [21] Hillman A, Eisenberg JM, Pauly MV, Bloom BS, Glick H, Kinosian B, Schwartz JS. Avoiding bias in the conduct and reporting of cost-effectiveness research sponsored by pharmaceutical companies. New Eng J Med 1991;324:1362–5. [22] Task Force on Principles for Economic Analysis of Health Care Technology. Economic analysis of health care technology a report on principles. Ann Intern Med 1995;122:61 – 70. [23] Department of Health press release 94/251. Guidelines for the economic evaluation of pharmaceuticals. London, 19 May 1994. [24] Drummond M, Jefferson T. Guidelines for authors and peer reviewers of economic submissions to the BMJ. Br Med J 1996;313:275–83. [25] Walley T, Edwards RT. Is there a need for an independent centre for pharmacoeconomics in the UK? Pharmacoeconomics 1994;5:93– 100. [26] Freemantle N, Henry D, Maynard A, Torrance G. Promoting cost effective prescribing. Br Med J –6. [27] NHS Centre for Reviews and Dissemination. Structured databases on line. York: NHS CRD, 1996.
Economic evaluations of drug therapy: attitudes of primary care prescribing advisers in Great Britain Tom Walley a,*, Stuart Barton a, Jonathan Cooke b, Michael Drummond c a
Department of Pharmacology and Therapeutics, Uni6ersity of Li6erpool, 70 Pembroke Place, Li6erpool L69 3GF, UK b South Manchester Uni6ersity Hospitals NHS Trust, Manchester, UK c Centre for Health Economics, Uni6ersity of York, York, UK Received 21 February 1997; accepted 25 March 1997
Abstract All health authorities in Great Britain have both medically or pharmaceutical qualified staff to advise both the authority and the local primary care medical practitioners about drug use and prescribing. This study used a piloted postal questionnaire to assess the attitudes of these advisers to economic evaluations of drug therapy, and their perceptions of the barriers to achieving cost effective prescribing by use of these evaluations. There was a 65% response rate to the questionnaire. Economic issues were rated by advisers to be less important than clinical issues, but were considered at most meetings between advisers and primary care medical practitioners. Advisers wished to consider true cost effective prescribing but often felt obliged to consider drug acquisition costs and risks of budgetary overspends. The perceived inflexibility of existing structures within the British National Health Service and the lack of credibility of the evaluations (often perceived as pharmaceutical industry marketing) were the major barriers to the application of the evaluations. The paper concludes that advisers were keen to use economic evaluations to promote cost effective prescribing but were impeded by the perceived bias of existing studies and by rigid current NHS structures. © 1997 Elsevier Science Ireland Ltd. Keywords: Economic evaluation; Drug therapy; National Health Service; Attitudes
* Corresponding author. Tel.: +44 151 7948125; fax: + 44 151 7948126; e-mail: [email protected] 0168-8510/97/$17.00 © 1997 Elsevier Science Ireland Ltd. All rights reserved. PII S 0 1 8 - 8 5 1 0 ( 9 7 ) 0 0 0 1 3 - 4
62
T. Walley et al. / Health Policy 41 (1997) 61–72
1. Introduction Faced with a rising primary care drug bill and uncertain of the value for money in spending on prescribing, the British government has introduced a range of initiatives, including the indicative prescribing scheme and the prescribing incentive schemes [1]. General practice fundholding is a related reform which also limits prescribing cost rises [2]. The stated aim of these initiatives was not cost containment, but improving the quality of prescribing and the value for money from prescribing [3]; hence there have also been educational initiatives, such as the production of drug information bulletins. To administer these initiatives and to advise both general medical practitioners and their health authorities on prescribing, the government funded a medical prescribing adviser within each local health authority [4]. There are now often pharmacist advisers either in addition to or instead of the medical advisers. The government stated that pharmaceutical companies could help to improve the use of resources in prescribing by providing evidence of the cost effectiveness of their products, in particular new drugs [3]. Companies were quick to seize on this, and many produced economic evaluations of their products [5]. Economic evaluations compare alternative treatments, including drug therapy, in terms of their costs and consequences. In theory, such evaluations could help to determine the best use of health service resources, particularly within prescribing. To the average prescriber however, such evaluations have little meaning, since he or she may bear no true responsibility for resource utilisation. In contrast, such evaluations could be of great value to a health authority or to a general practice fundholder, who have responsibility for budgets covering many areas of health services. To be of use, these evaluations must be reliable and easily understood by users. Also there must be scope for users to act on the basis of the study findings. We surveyed health authority primary care prescribing advisers on whether and how they used economic evaluations in their advice on prescribing to both general practitioners and their health authorities.
2. Methods A focus group meeting between one author (TW) and a small number of prescribing advisers generated hypotheses concerning their use of economic evaluations to be tested in this study. Based on these, a questionnaire was developed which covered four main themes: knowledge of economics, the importance of economic and financial issues, barriers to the use of economic analysis and awareness of published studies. This questionnaire (appendix) was piloted and subsequently sent by post to all 273 and pharmaceutical prescribers in England, Wales and Scotland in July 1995. These were identified using the database
T. Walley et al. / Health Policy 41 (1997) 61–72
63
of the Medical Advisers’ Support Centre. The questionnaire was anonymised and no follow up of nonresponders was undertaken. To test awareness of published studies, advisers were given a list of some recently published studies [6 – 15] and asked whether they had read the study and whether it had altered their advice to their authority or to general practitioners. The validity of the answers to this part was assessed by the use of two plausible control studies [16,17], based on clinical papers which in fact had contained no health economic assessment. Results were analysed by SPSS.
3. Results A total of 178 responses were received (65% response rate). Response rates and responses were similar from both medical and pharmaceutical advisers and therefore no distinction between the two is made in the results. There was no systematic difference in the age, sex or experience of responders and nonresponders. The advisers had been in post an average of 33 months. A figure of 40% claimed to have some formal training in health economics, for the most part a distance learning package produced especially for this group [18]. The responses for those claiming such training were similar to those without and again no distinction is made in the results.
3.1. Importance of economic issues Only 5% considered economic considerations the most important issue in deciding how to advise general practitioners. Most (91%) considered economic issues important but secondary to issues of clinical effectiveness. A total of 92% of advisers discussed economic issues with general practitioners in most of their visits to practices. Only 4% considered economic factors totally unimportant. In advising their authorities, the pattern of response was similar: 11% thought economic issues the most important, and only 2% considered them unimportant, with most again believing that economic issues were secondary to clinical issues.
3.2. Economic issues considered There were differences between what economic issues advisers actually considered in discussions on prescribing either with general practitioners or with their health authority (drug acquisition costs and risks of budgetary overspends), and what advisers would ideally like to consider (cost effectiveness of drugs in the wider health service) (Table 1).
64
T. Walley et al. / Health Policy 41 (1997) 61–72
3.3. Sources of information The sources which the advisers claimed to use to assess the cost effectiveness of a therapy were clinical trials in peer reviewed journals (96%), formal economic evaluations (77%), and opinions of colleagues (49%). Almost all (98%) were keen to make more use of economic evaluations of drug therapy; the advice they would wish to offer as a result of such studies was broader than just choice of drug, and included informing purchasing decisions, such as a choice between drug therapy and other services.
3.4. Barriers to application of economic e6aluations Advisers ranked a list of possible barriers to the use of economic evaluations (Table 2). The major barriers were the perceived inflexibility of existing NHS structures preventing the application of the evaluations, and the lack of credibility of the evaluations.
3.5. Awareness of current e6aluations The two control studies were claimed to have been seen by 16 and 21% of advisers, to have convinced 14 and 7%, and to have encouraged 7 and 2% to have altered their advice on prescribing. Claimed awareness of the other published studies ranged from 58 [6] to 14% [7]. The awareness of seven of the ten studies [6,9–11,14–16] was significantly higher than that of the controls, judged by a result more than two standard deviations from the predicted value. The studies most convincing and influential were those with perhaps the best clinical basis, i.e. enalapril in heart failure [8] (61% found convincing, 34% altered the advice given) and H pylori eradication in duodenal ulceration [9] (45% convincing, 34% altered advice). Another highly regarded study was an Effective Health Care bulletin on antidepressants [6] (44% convincing and 26% altered advice given).
Table 1 Key economic issues for prescribing advisers
Acquisition costs of drug Risk of budgetary overspend for GP Risk of budgetary overspend for health authority Cost effectiveness-better clinical outcomes as a compensation for higher drug costs Cost effectiveness-savings elsewhere in the NHS as a compensation for higher drug costs
Currently consider as a factor (%)
Ideally would consider as a factor (%)
87 70 64 68
8 14 20 41
35
T. Walley et al. / Health Policy 41 (1997) 61–72
65
Table 2 Obstacles considered most important to the use of cost effectiveness studies of drug therapy in the NHS 1. Studies unrealistic because existing NHS structures prevent moving resources between budgets (e.g. from secondary care to prescribing in primary care) 2. Industry funded studies not credible 3. Studies too open to bias because of need to make assumptions 4. The agenda of the Department of Health is not long term cost effective prescribing, but short term containment of the drug bill 5. Personal difficulty in interpreting the studies 6. Studies unreal-‘savings’ are theoretical and are unlikely ever to be achieved 7. Advisers or health authorities unable to free resources to follow the suggested approach to therapy 8. Health authorities as a whole don’t really understand the process 9. Difficulties (of general practitioners) in interpreting the studies 10. Advisers or health authorities unable to take a long term view-need to meet budgetary constraints now 11. Dept of Health funded studies are not credible 12. Studies relate to overall NHS costs-Advisers or health authorities only concerned with primary care drug costs 13. Inability to act about using such studies-no personal power a
Ranked according to scoring by advisers.
4. Discussion Advisers were generally keen to make best use of health service resources, and saw more use of economic evaluations of drug therapy as a means to this end. However, they felt constrained by a number of limitations, the most important of which were the rigidity of NHS structures and difficulties in reallocating NHS resources, and the perceived current unreliability of such evaluations. At present, increased spending on drug therapy in primary care cannot be funded by any savings which might result in a hospital budget (e.g. by use of inhaled steroids in asthma, or by ACE inhibitors in heart failure, either of which may reduce hospital admissions). This is largely the result of the funding arrangements for drugs in primary care, which are funded by a separate parliamentary vote from other NHS spending [19]. Furthermore to actually convert the possible savings in hospital spending, often identified in economic evaluations, into cash which can be used elsewhere, would require closing beds and cutting staff. This might not be practically possible or politically acceptable. This limits the applicability of pharmacoeconomic studies for a health authority, which instead finds such studies used, not to make better use of limited resources, but to justify demands for more resources in drug therapy. Faced with a rising drug bill and no savings elsewhere in health spending, the focus for a health authority or government with a limited budget may move to cost containment. This was reflected in the opinion of the advisers that cost containment was a higher priority for the government truly cost effective prescribing. Such an approach would not be acceptable professionally to the large majority of
66
T. Walley et al. / Health Policy 41 (1997) 61–72
advisers, and this is reflected in their rating of clinical effectiveness ahead of economic issues. The difficulty of taking a long term view also reduces the value of economic evaluations. Many studies may identify benefits over 5–10 years in exchange for increased spending now: if as is often the case, the health authority is in danger of overspending now, investment in a policy for long term savings may not be possible. General practice fundholders have in theory more freedom to move money between budgets than health authorities, but are also constrained by the need to keep within their annual budget which may prevent them from applying studies. Finally, managers are often responsible for only one portion of the overall budget, and may not be able to see the wider economic efficiency. Advisers with responsibilities only for the primary care drug budget might therefore be reluctant to see an overspend in their budget for the sake of benefits elsewhere. It is reassuring that only 14% of the respondents to our survey accepted this approach. Given the perception of advisers of pressures to contain cost, this approach may be more common than admitted. At the time of our survey, health authorities dealing with primary care and for whom prescribing advisers worked were separate from the health authorities responsible for secondary care had not formally merged in England and Wales, and this may have exacerbated this problem. These health authorities have now merged. The second major problem identified was that the studies themselves were not seen as credible, because of the nature of the study itself or its funding, or because of the lack of understanding on the part of the adviser, the general practitioner or the health authority. The interpretation of economic evaluations is new to clinicians, and hence suspicions of bias and of industry funded studies grow [20,21]. Attempts have been made to address these problems internationally [22] and in Great Britain [23]. The majority of the studies seen by the advisers had been given to them either directly or indirectly by pharmaceutical company representatives, increasing their already inherent suspicion. Few advisers were aware of existing specific British guidelines [23] (produced jointly by the Association of the British Pharmaceutical Industry and the Department of Health) for the conduct of economic evaluations, and of those who were aware, none considered these guidelines useful to them. This survey was undertaken before the recent publication of guidelines for the review of such studies by the British Medical Journal [24]. Scientific rigour of this kind will enhance the credibility of such studies in Britain. Are there ways around this impasse for economic evaluations? One comment from many advisers was that there should be a central means of validating studies for use within the NHS [25], perhaps even as part of the drug licencing or reimbursement process [26]. The NHS Centre for Reviews and Dissemination partly fulfils this role through its databases on reviews of effectiveness and economic evaluations [27], but the advisers were largely unaware of it. There is also a need to increase the training advisers in health economics, so as to allow them to become critical and intelligent readers of economic evaluations.
T. Walley et al. / Health Policy 41 (1997) 61–72
67
It is important to acknowledge the limitations of surveys of this kind: the answers given by respondents may reflect how they believe they should respond rather than what they actually do or think. Some bias may have been introduced by the non responders, although the response rate is good for a single postal survey without reminders. We did not explore the respondents’ wider understanding of health economics. The barriers identified are those perceived by those who took part in the initial interviews: no new barriers were identified in free text by respondents. How real these barriers are is not clear, since there was no evidence of any attempt by the advisers to actually use an economic evaluation to inform purchasing decisions. This survey considers only one of the possible targets of economic evaluations of drug therapy: others would include perhaps fundholding general practitioners, directors of public health and senior hospital pharmacists: the latter two have also been surveyed and will be reported separately. If economic evaluations of drug therapies are to be used, they must be credible and accessible to decision makers. However current NHS structures seem to make the actual implementation of economic evaluations relating to prescribing, even if accepted as correct, largely impossible. More consideration should be given to exploring these structural barriers and how they might be overcome to actively promote true cost effective prescribing.
Acknowledgements This work was partly funded by Duke University and by Care Research and Development Centre
Appendix See following pages.
National Primary
68
T. Walley et al. /Health Policy 41 (1997) 61–72
T. Walley et al. /Health Policy 41 (1997) 61–72
69
70
T. Walley et al. /Health Policy 41 (1997) 61–72
T. Walley et al. /Health Policy 41 (1997) 61–72
71
72
T. Walley et al. / Health Policy 41 (1997) 61–72
References [1] Walley T, Wilson R, Bligh J. Current prescribing in primary care in the UK: effects of the indicative prescribing scheme and GP fundholding. Pharmacoeconomics 1995;7:320 – 31. [2] Wilson R, Buchan I, Walley T. Alterations in prescribing by GP Fundholders; an observational study. Br Med J 1995;311:1347–50. [3] Department of Health. Improving prescribing. London: HMSO, 1990. [4] Walley T, Bligh J. FHSA medical advisers: friend or foe?. Br Med J 1992;304:133 – 4. [5] Walley T, Edwards RT. Health economics in primary care in the UK: containment of drug costs. Pharmacoeconomics 1993;3:100–6. [6] Anon. The treatment of depression in primary care. Effective Health Care Bull 1993;5:1 – 12. [7] Knill Jones R, McGhee S, McDonald F. An economic evaluation of enteric-coated and standard formulations of naproxen from a gastrointestinal perspective. Br J Med Econ 1994;7:185 – 96. [8] Hart W, Rhodes G, McMurray J. The cost effectiveness of enalapril in the treatment of chronic heart failure. Br J Med Econ 1993;6:91 – 8. [9] Jones R, Bosanquet N, Johnson NJ, et al. Cost effective management strategies for acid-peptic disorders. Br J Med Econ 1994;7:99 – 114. [10] Jonsson B, Bebington PE. What price depression? The cost of depression and the cost-effectiveness of pharmacological treatment. Br J Psychol 1994;164:665 – 73. [11] Akehurst R, Piercy J. Cost-effectiveness of the use of Nicorette nasal spray to assist quitting smoking among heavy smokers. Br J Med Econ 1994;7:155 – 84. [12] Cull R, Wells N, Miocevick M. Economic costs of migraine. Br J Med Econ 1992;5:103 – 15. [13] Knill Jones R. An economic evaluation of Arthrotec in the treatment of arthritis. Br J Med Econ 1992;5:51–8. [14] Lancaster Smith M, Gough K, Wells N, et al. An economic analysis of ranitidine versus cimetidine in the prevention of duodenal ulcer recurrence. Br J Med Econ 1992;2:25 – 36. [15] Bates CM, Richardson PDI. Cost-effectiveness of omeprazole in the management of gastroesophageal reflux disease in clinical practice. Br J Med Econ 1994;7:81 – 97. [16] Greening AP, Ind PW, Northfield M, Shaw G. Added salmeterol versus higher-dose corticosteroid in asthma patients with symptoms on existing inhaled corticosteroid. Lancet 1994;344:219 – 24. [17] Dahlof B, Hansson L, Lindholm LH, Schersten B, Wester PO, Ekbom T, Hedner T, de-Faire U. STOP-Hypertension 2: a prospective intervention trial of ‘newer’ versus ‘older’ treatment alternatives in old patients with hypertension. Swedish Trial in Old Patients with Hypertension. Blood Press 1993;2:136–41. [18] Edwards, R. Distance learning programme in health economics Vol. 1 – 6. Medical Advisers Support Centre, Liverpool, 1993. [19] Crump BJ, Panton R, Drummond MF, Marchment M, Hawkes R. Transferring the costs of expensive treatments from secondary to primary care. Br Med J 1995;310:509 – 12. [20] Horton R. Pharmacoeconomical with the truth. Lancet 1993;341:752. [21] Hillman A, Eisenberg JM, Pauly MV, Bloom BS, Glick H, Kinosian B, Schwartz JS. Avoiding bias in the conduct and reporting of cost-effectiveness research sponsored by pharmaceutical companies. New Eng J Med 1991;324:1362–5. [22] Task Force on Principles for Economic Analysis of Health Care Technology. Economic analysis of health care technology a report on principles. Ann Intern Med 1995;122:61 – 70. [23] Department of Health press release 94/251. Guidelines for the economic evaluation of pharmaceuticals. London, 19 May 1994. [24] Drummond M, Jefferson T. Guidelines for authors and peer reviewers of economic submissions to the BMJ. Br Med J 1996;313:275–83. [25] Walley T, Edwards RT. Is there a need for an independent centre for pharmacoeconomics in the UK? Pharmacoeconomics 1994;5:93– 100. [26] Freemantle N, Henry D, Maynard A, Torrance G. Promoting cost effective prescribing. Br Med J –6. [27] NHS Centre for Reviews and Dissemination. Structured databases on line. York: NHS CRD, 1996.