Comparing estimates of cost effectiveness submitted to the National In
http://www.100md.com
《英国医生杂志》
1 Health Economics Research Group, Brunel University, Uxbridge UB8 3PH, 2 Economics Department, University of York, Heslington, York YO1 5DD, 3 Health Policy Unit, Department of Public Health and Policy, London School of Health and Tropical Medicine, London WC1E 7HT, 4 Department of Community Health Sciences, St George's Hospital Medical School, London SW17 0RE
Correspondence to: A H Miners Alec.Miners@nice.nhs.uk
Abstract
Evidence suggests that profit making organisations are more likely to report favourable results and conclusions from clinical studies than non-profit organisations. The most recent of these studies concluded that for profit organisations were 5.3 (95% confidence interval 2.0 to 14.4) times more likely to recommend the use of an experimental drug compared with non-profit groups.1 Only three studies, however, have formally assessed this association with regard to the results from economic evaluations. One showed that evaluations of oncology products sponsored by drug companies were less likely to report unfavourable qualitative conclusions (5% v 38%; P = 0.04) compared with studies sponsored by non-profit organisations.2 The two other studies also reported similar findings.3-4
The technology appraisals programme of the National Institute for Clinical Excellence (NICE) provides guidance to the NHS in England and Wales on the use of new and existing health technologies. Manufacturers (or relevant UK agent) of the relevant technology, professional groups, and national groups representing patients submit evidence. Collectively, these three groups are known as consultees. An academic centre commissioned through the NHS's health technology assessment programme, called an assessment group, also assesses the evidence. Guidance to the NHS is formulated with regard to this evidence, which includes information on cost effectiveness.
The criterion traditionally used to assess cost effectiveness is the magnitude of a statistic known as the "incremental cost effectiveness ratio," defined as the difference in costs between two technologies divided by the difference in their benefits.5 The lower the incremental cost effectiveness ratio, the more cost effective a technology.
We evaluated the association between source of funding and the results from economic evaluations submitted to NICE's technology appraisals programme.
Methods
Of the 62 appraisals, we excluded 35 because the assessment group alone did not produce an incremental cost effectiveness ratio (n = 20), there was no manufacturer or the manufacturer did not produce an estimate (n = 3), the assessment group and the manufacturer did not produce an estimate (n = 8), or different measures of health benefit were used (n = 4). The 27 remaining appraisals contained 54 pairwise comparisons.
Over 80% of the manufacturers' incremental cost effectiveness ratios were between dominant (meaning that the appraised technology was considered to be less costly and more effective than the comparator technology) and ￡15 000 per unit of outcome (categories 0 and 1 combined) whereas the assessment groups' estimates were more uniformly distributed over the five categories (fig 1). Only four (7%) estimates from manufacturers were above ￡30 000 per unit of outcome compared with 19 (35%) estimates from the assessment groups.
Fig 1 Distribution of estimates of incremental cost effectiveness ratios from assessment groups and manufacturers according to category of cost effectiveness (see table)
There were 25 cases of negative rank (where the manufacturers' estimates were lower than those of the corresponding assessment group), 29 ties (where the estimates were in the same category), and no positive ranks (P < 0.01; see box for example). Sensitivity analysis around the category definitions did not alter the significance of this result. Figure 2 plots the log of these pairwise comparisons; the 45° line indicates identical incremental cost effectiveness ratios. Points above this line indicate that a manufacturer's estimate was higher (less favourable) than the estimate produced by the assessment group, whereas points below this line indicate the manufacturer submitted a lower (more favourable) estimate than that of the assessment group. The analysis conducted with the second method reduced the 54 pairs of incremental cost effectiveness ratios to 27 pairs, of which two were a tie. The analysis showed that in 21 instances the manufacturers' estimates were lower than the assessment groups' estimates and in the four remaining they were higher (P < 0.001).
Fig 2 Logged pairwise comparison of incremental cost effectiveness ratios. Only 45/54 pairs of recommendations have been plotted because remaining nine pairs contained at least one negative estimate and could not be logged. However, in six of nine pairs estimates from the two sources were broadly similar and in three pairs estimates reported by manufacturers were much more favourable towards the technology than those reported by the assessment groups
Discussion
Als-Nielsen B, Chen W, Gluud C, Kjaergard LL. Association of funding and conclusions in randomized drug trials: a reflection of treatment effect or adverse events? JAMA 2003;290: 921-8.
Friedberg M, Saffran B, Stinson TJ, Nelson W, Bennett CL. Evaluation of conflict of interest in economic analyses of new drugs used in oncology. JAMA 1999;15: 1453-7.
Lexchin J, Bero LA, DJulbegovic B, Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ 2003;326: 1167-1170
Azimi NA, Welch HG. The effectiveness of cost-effectiveness analysis in containing costs. J Gen Intern Med 1998;10: 664-9.
Drummond MF, O'Brien B, Stoddart GL, Torrance GW. Methods for the economic evaluation of health care programmes. Oxford: Oxford University Press, 1997.
NICE. Guidance for manufacturers and sponsors. www.nice.org.uk/pdf/technicalguidanceformanufacturersandsponsors.pdf (accessed 6 May 2004).
Devlin N, Parkin D. Does NICE have a cost-effectiveness threshold and what other factors influence its decisions? A binary choice analysis. Health Econ 2004;13: 437-52.
Raftery J. NICE: faster access to modern treatments? Analysis of guidance on health technologies. BMJ 2001;323: 1300-3.
Briggs A. Handling uncertainty in cost-effectiveness models. PharmacoEconomics 2000;17: 479-500.
Briggs AH, Goeree R, Blackhouse G, O'Brien BJ. Probabilistic analysis of cost-effectiveness models: choosing between strategies for gastroesophageal reflux disease. Med Decis Making 2002;22: 290-308.
Van Hout BA, Al MJ, Gordon GS, Rutten FF. Costs, effects and C/E ratios alongside clinical trials. Health Econ 1994;3: 309-19.
Garrison LP. The ISPOR good practice modelling principles—a sensible approach: be transparent, be reasonable. Value in Health 2003;6: 6-8.
Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost-effectiveness in health and medicine. New York: Oxford University Press, 1996.(A H Miners, visiting research fellow1, M)
Correspondence to: A H Miners Alec.Miners@nice.nhs.uk
Abstract
Evidence suggests that profit making organisations are more likely to report favourable results and conclusions from clinical studies than non-profit organisations. The most recent of these studies concluded that for profit organisations were 5.3 (95% confidence interval 2.0 to 14.4) times more likely to recommend the use of an experimental drug compared with non-profit groups.1 Only three studies, however, have formally assessed this association with regard to the results from economic evaluations. One showed that evaluations of oncology products sponsored by drug companies were less likely to report unfavourable qualitative conclusions (5% v 38%; P = 0.04) compared with studies sponsored by non-profit organisations.2 The two other studies also reported similar findings.3-4
The technology appraisals programme of the National Institute for Clinical Excellence (NICE) provides guidance to the NHS in England and Wales on the use of new and existing health technologies. Manufacturers (or relevant UK agent) of the relevant technology, professional groups, and national groups representing patients submit evidence. Collectively, these three groups are known as consultees. An academic centre commissioned through the NHS's health technology assessment programme, called an assessment group, also assesses the evidence. Guidance to the NHS is formulated with regard to this evidence, which includes information on cost effectiveness.
The criterion traditionally used to assess cost effectiveness is the magnitude of a statistic known as the "incremental cost effectiveness ratio," defined as the difference in costs between two technologies divided by the difference in their benefits.5 The lower the incremental cost effectiveness ratio, the more cost effective a technology.
We evaluated the association between source of funding and the results from economic evaluations submitted to NICE's technology appraisals programme.
Methods
Of the 62 appraisals, we excluded 35 because the assessment group alone did not produce an incremental cost effectiveness ratio (n = 20), there was no manufacturer or the manufacturer did not produce an estimate (n = 3), the assessment group and the manufacturer did not produce an estimate (n = 8), or different measures of health benefit were used (n = 4). The 27 remaining appraisals contained 54 pairwise comparisons.
Over 80% of the manufacturers' incremental cost effectiveness ratios were between dominant (meaning that the appraised technology was considered to be less costly and more effective than the comparator technology) and ￡15 000 per unit of outcome (categories 0 and 1 combined) whereas the assessment groups' estimates were more uniformly distributed over the five categories (fig 1). Only four (7%) estimates from manufacturers were above ￡30 000 per unit of outcome compared with 19 (35%) estimates from the assessment groups.
Fig 1 Distribution of estimates of incremental cost effectiveness ratios from assessment groups and manufacturers according to category of cost effectiveness (see table)
There were 25 cases of negative rank (where the manufacturers' estimates were lower than those of the corresponding assessment group), 29 ties (where the estimates were in the same category), and no positive ranks (P < 0.01; see box for example). Sensitivity analysis around the category definitions did not alter the significance of this result. Figure 2 plots the log of these pairwise comparisons; the 45° line indicates identical incremental cost effectiveness ratios. Points above this line indicate that a manufacturer's estimate was higher (less favourable) than the estimate produced by the assessment group, whereas points below this line indicate the manufacturer submitted a lower (more favourable) estimate than that of the assessment group. The analysis conducted with the second method reduced the 54 pairs of incremental cost effectiveness ratios to 27 pairs, of which two were a tie. The analysis showed that in 21 instances the manufacturers' estimates were lower than the assessment groups' estimates and in the four remaining they were higher (P < 0.001).
Fig 2 Logged pairwise comparison of incremental cost effectiveness ratios. Only 45/54 pairs of recommendations have been plotted because remaining nine pairs contained at least one negative estimate and could not be logged. However, in six of nine pairs estimates from the two sources were broadly similar and in three pairs estimates reported by manufacturers were much more favourable towards the technology than those reported by the assessment groups
Discussion
Als-Nielsen B, Chen W, Gluud C, Kjaergard LL. Association of funding and conclusions in randomized drug trials: a reflection of treatment effect or adverse events? JAMA 2003;290: 921-8.
Friedberg M, Saffran B, Stinson TJ, Nelson W, Bennett CL. Evaluation of conflict of interest in economic analyses of new drugs used in oncology. JAMA 1999;15: 1453-7.
Lexchin J, Bero LA, DJulbegovic B, Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ 2003;326: 1167-1170
Azimi NA, Welch HG. The effectiveness of cost-effectiveness analysis in containing costs. J Gen Intern Med 1998;10: 664-9.
Drummond MF, O'Brien B, Stoddart GL, Torrance GW. Methods for the economic evaluation of health care programmes. Oxford: Oxford University Press, 1997.
NICE. Guidance for manufacturers and sponsors. www.nice.org.uk/pdf/technicalguidanceformanufacturersandsponsors.pdf (accessed 6 May 2004).
Devlin N, Parkin D. Does NICE have a cost-effectiveness threshold and what other factors influence its decisions? A binary choice analysis. Health Econ 2004;13: 437-52.
Raftery J. NICE: faster access to modern treatments? Analysis of guidance on health technologies. BMJ 2001;323: 1300-3.
Briggs A. Handling uncertainty in cost-effectiveness models. PharmacoEconomics 2000;17: 479-500.
Briggs AH, Goeree R, Blackhouse G, O'Brien BJ. Probabilistic analysis of cost-effectiveness models: choosing between strategies for gastroesophageal reflux disease. Med Decis Making 2002;22: 290-308.
Van Hout BA, Al MJ, Gordon GS, Rutten FF. Costs, effects and C/E ratios alongside clinical trials. Health Econ 1994;3: 309-19.
Garrison LP. The ISPOR good practice modelling principles—a sensible approach: be transparent, be reasonable. Value in Health 2003;6: 6-8.
Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost-effectiveness in health and medicine. New York: Oxford University Press, 1996.(A H Miners, visiting research fellow1, M)