Population based cost utility study of interferon beta-1b in secondary progressive multiple sclerosis
BMJ 1999; 319 doi: https://doi.org/10.1136/bmj.319.7224.1529 (Published 11 December 1999) Cite this as: BMJ 1999;319:1529
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
EDITOR-We read with interest the population-based cost-utility study
of interferon beta-1b in secondary progressive multiple sclerosis1 but
were disappointed that the report was not placed in the context of a
systematic review of relevant published work. We have recently completed
such a review of the effectiveness, costs and utility of a range of
disease modifying drugs in multiple sclerosis for the Health Technology
Assessment programme (submitted for publication). As part of this, we
searched for published studies of the cost-utility of interferon beta in
multiple sclerosis, in electronic databases: Cochrane Database of
Systematic Reviews, Cochrane Controlled Trials Register, Database of
Abstracts of Reviews of Effectiveness and NHS Economic Evaluations
Database, Medline (Silverplatter), PubMed, Embase and National Research
Register (for the period 1980 to July 1999 and limited to English language
studies.) In addition, experts in the field were contacted.
In addition to the Forbes study, we identified three further studies:
one of beta interferons (1a and 1b) in both relapsing-remitting and
secondary progressive sclerosis2 as well as two further studies of
interferon in relapsing-remitting multiple sclerosis. 3,4
A summary of the results of the review is shown (see table). The
studies in secondary progressive multiple sclerosis are based on the same
trial (n=718) but use different cost-utility methodologies. Forbes’
conclusion that the cost per QALY was very high (£1,024,667) due to the
high drug cost and modest clinical effect concurs with the previous
study’s estimate of £874,600 per QALY. The studies in relapsing-remitting
multiple sclerosis also use the results of the existing RCTs in
conjunction with various sources of cost data, QoL measures and different
degrees of decision analytic modelling. Results give a high cost per QALY
per relapse avoided in all studies.
These cost-utility studies were critically appraised using standard
criteria for decision analysis and economic evaluations. Although all used
explicit and sensible processes to identify, select and combine the
evidence into probabilities, and to obtain utilities and costs, with
uncertainties being addressed by sensitivity analyses, none of them
provide a full economic comparison of all possible healthcare strategies.
The conclusion of Forbes and colleagues that ‘far more benefit would be
obtained from directing funds into improved supportive care’, though
plausible, is not supported by the evidence presented. Further economic
evaluations of a wider range of possible interventions, such as improved
supportive care, are needed in order to efficiently use limited resources.
Cost/QALY of interferons in MS
____________________________________________________
Study (Country) Drug
and patient group Cost/QALY
____________________________________________________
Forbes 1 (UK) 1999 IFNB-1b in SPMS £1,024,000
Nicholson 2, (UK) 1999 IFNB-1b in SPMS £874,600
Parkin 3 (UK) 1998 IFNB-1b in RRMS £809,000
Nicholson 2 (UK) 1999 IFNB-1a in RRMS £2,038,400
Otten 4 (Canada) 1998 IFNB-1a in RRMS $406,000 - $490,000
____________________________________________________
Jackie Bryant
Research Fellow
Andrew Clegg
Senior Research Fellow
Ruairidh Milne
Senior Lecturer in Public Health Medicine
Wessex Institute for Health Research and Development
University of Southampton,
Biomedical Sciences Building, Bassett Crescent East
Southampton SO16 7PX
1. Forbes RB, Lees A, Waugh N, Swingler RJ. Population based cost-
utility study of interferon beta 1b in secondary progressive multiple
sclerosis. BMJ 1999;319:1529-33.
2. Nicolson T, and Milne R. Beta Interferons (1a and 1b) in relapsing-
remitting and secondary progressive multiple sclerosis. Southampton:
Wessex Institute for Health Research and Development 1999 June.
Development and Evaluation Committee Report No. 98. 1999.
3. Parkin D, McNamee P, Jacoby A, Miller P, Thomas S, and Bates D. A cost-
utility analysis of interferon beta for multiple sclerosis. Health Technol
Assessment 2(4). 1998.
4. Otten, N. Comparison of drug treatments for multiple sclerosis. Ottawa:
Canadian Coordinating Office for Health Technology Assessment (CCOHTA).
1998.
Competing interests: Cost/QALY of interferons in MS ____________________________________________________Study (Country) Drug and patient group Cost/QALY____________________________________________________Forbes 1 (UK) 1999 IFNB-1b in SPMS £1,024,000Nicholson 2, (UK) 1999 IFNB-1b in SPMS £874,600Parkin 3 (UK) 1998 IFNB-1b in RRMS £809,000Nicholson 2 (UK) 1999 IFNB-1a in RRMS £2,038,400Otten 4 (Canada) 1998 IFNB-1a in RRMS $406,000 - $490,000____________________________________________________
Dear Sir,
The paper by Forbes et al1 is most welcome, given the
difficult debate on this topic; we would particularly like to endorse the
views expressed in the final paragraph.
However, we would also like to draw attention to methodological issues in
calculating cost effectiveness. Forbes et al1 have used the only trial
data available, curtailed at 36 months.2 For the outcome used by Forbes et
al, time to wheelchair dependence, the median point was not reached.
Instead the trial reports, rather vaguely, that the benefit was 'a delay
of up to 9 months' in time to wheelchair dependence.2 This appears to be
based on the 83rd centile: at 30 months for the treatment arm and 20.5
months for the control arm (see graph); as such it is an arbitrary measure
of the difference between the distribution curves, determined by the
length of the trial. As a second dimension of the effectiveness Forbes et
al1 use the number-needed-to-treat at 30 months (NNT = 18), based on the
relative risk reduction from the trial and the (lower) rate of reaching
wheelchair dependency from their Tayside cohort. This gives an average
delay for a patient of just 0.5 months for the Tayside patients (0.9
months for the trial cohort2, with a NNT of 10 calculated, using the graph
below, from the difference between the numbers of treated and control
patients at 30 months, [100/83-73]).
The graph plots the estimated probabilities of being not wheelchair bound
from the trial data2 as applied to a cohort of 100 patients. The area
under each curve was estimated by summing the areas of the polygons for
each month using a spreadsheet. The difference between them is the
difference in the wheelchair dependent time, and generates a true mean at
30 months of 1.7 months delay (c.f. the 0.9 months above). The method
used by Forbes et al,1 estimates only the (marginal) benefit as
represented by the area between the curves from 21 to 30 months.
This area-under-curve method, we believe, offers the best means of
calculating true benefit in most cases, especially when data are limited,
though it has its own problems. Further debate is needed as methodologies
for health economic assessments develop.
As for individual patients, they may be led to believe that treatment
offers a 9 months delay, where as in reality it is just 1.7 months on
average. Would this knowledge change their perspective?
Richard Richards
consultant public health physician
North Nottinghamshire Health Authority,
Ransom Hall,
Rainworth,
Mansfield NG21 0ER
Amanda Burls
senior clinical lecturer in public health & epidemiology
Department of Public Health and Epidemiology,
University of Birmingham,
Edgbaston,
Birmingham B15 2TT
Nick Payne
clinical senior lecturer
School of Health and Related Research (ScHARR),
University of Sheffield,
Regent Court,
30 Regent Street,
Sheffield S1 4DA
1 Forbes RB, Lees A, Waugh N, Swingler RJ. Population based cost
utility study of interferon beta-1b in secondary progressive multiple
sclerosis. BMJ 1999;319:1529-33
2 European Study Group on Interferon beta-1b in Secondary Progressive
MS. Placebo-controlled multicentre randomised trial of interferon beta-1b
in treatment secondary progressive multiple sclerosis. Lancet
1998;352:1491-97.
Competing interests: No competing interests
Simon J. Ellis
M.A. M.B. B.Chir. F.R.C.P. M.D.
Consultant Neurologist/Senior Clinical Lecturer
North Staffordshire Royal Infirmary, Keele University,
Princes Road, Hartshill, Stoke-on-Trent, Staffordshire, ST4 7LN, U.K.
Telephone: 01782 554343 Fax: 01782 555008 E-mail:
Simon@Northesk.Demon.co.UK
Friday, 24 December 1999.
Cost Utility Analysis an analysis too far.
Dear Editor,
The paper by Rorbes et al. (1) gives the appearance of applying
scientific methodology to the challenge of resource allocation. However,
if the methodology is flawed such papers have the potential of influencing
the political debate resulting in patients being deprived useful therapy.
The paper takes a local audit of persons with multiple sclerosis,
applies the results from a single double blind placebo controlled trial
(2) using a single aspect of disability (wheelchair dependence) and then
extrapolates QALYs to which a price tag is attached. This results in the
apparently exorbitant cost per QALY of £832399.
The study cited of the use of beta-1b interferon was designed as an
efficacy study. Any further analysis of the data has to be viewed as
hypothesis generating rather then strong evidence in its own right. Post
hoc manipulations of data are notoriously unreliable as was demonstrated
by the unfortunate safety scare over the use of Selegeline in Parkinson's
Disease (3) which has not been confirmed in other studies.
Although wheel chair dependence is an important aspect of disability
in multiple sclerosis it is not the only one. Any study into the overall
benefit of a therapy in MS that only used this one aspect of disability
seriously underestimated the problems associated with this disease.
The assessment of the effect of medical interventions is problematic
(4) and the underlying assumption behind QALYs that badness of a
disability is a fixed quantity is unlikely to be true. Many healthy people
might say they would prefer to be dead rather than live with a significant
disability, but if you ask disabled people they generally prefer to be
alive. The badness of a disability depends on ones perspective and is not
fixed.
What particularly worries me about Rorbes et al.'s paper is that it
will be used as strong evidence to limit availability of treatment to
patients and may constitute another example of the misuse of so called
evidence based medicine for rationing purposes (5). This sort of
exploration should be used as hypothesis generating in order to inform the
design of a proper cost benefit trial.
Yours sincerely,
Simon J. Ellis.
1. Rorbes RB, Lees A, Waugh N, Swingler RJ. Population based cost
utility study of beta-1b in secondary progressive multiple sclerosis. BMJ
1999;319:1529-33.
2. European Study Group on Interferon beta-1b in Secondary
Progressive Multiple Sclerosis. Placebo-controlled multicentre randomised
trial of interferon beta-1b in treatment of secondary progressive multiple
sclerosis. Lancet 1998; 352:1491-7.
3. Ben-Shlomo Y, Churchyard A, Head J, Hurwitz B, Overstall P,
Ockelford J, Lees AJ. Investigation by Parkinson's Disease
Research Group of United Kingdom into excess mortality seen
with combined levodopa and selegiline treatment in patients
with early, mild Parkinson's disease: further results of randomised trial
and confidentialinquiry:BMJ.316(7139):1191-6,1998.
4. Wright JC, Weinstein MS. Gains in life expectancy from medical
interventions - standardizing data outcomes. New England Journal of
Medicine 1998; 339:380-6.
5. Kryger M. Sleep apnoe and the misuse of evidence based medicine.
Lancet 1997; 349:803-4.
Competing interests: No competing interests
QALY approach is inappropraiate in disabled populations
Editor - A recent cost utility analysis of interferon beta-1b in
secondary progressive multiple sclerosis concluded that the cost per QALY
gained by these disabled patients is high, while clinical effect was
modest (1). This conclusion is definitely misleading, as the QALY approach
as a measure of treatment effect size is inappropriate in disabled
populations (2).
Health status weights of utility instruments are usually
derived from the general population. Comparison of non-disabled and
disabled populations puts the later at an additional disadvantage, as
disabled people have less room to improve in the QALY model (3, 4).
Recent
psychometric evaluation of the EQ-5D demonstrated a non-Gaussian
distribution of scores (5), indicating that the use of mean instead of
median values would distort any calculation of treatment effect size in
terms of QALYs.
Finally, if utility measures were applied in disabled
populations, results should at least be adjusted for the intrinsic
reduction of baseline quality of life in comparison with non-disabled
people. Using the published data of the QALY model (1), one year at
walking indoors with assistance would equal 2.6 years at unrestricted
ambulation. Consecutively, all cost calculations should at least be
corrected for this factor, yielding definitely lower costs per QALY gained
by interferon beta-1b treatment.
Physicians are trained to guarantee optimal treatment for their patients.
Inappropriate calculations based on unwise methods must not lead us to the
inadequate consequence of withholding the single available drug for
treatment of secondary progressive multiple sclerosis patients. Apart from
that, we agree with the authors that a comparison should be conducted
between interferon beta-1b treatment and supportive care such as
physiotherapy, to prove treatment effects and to allocate resources on a
scientifically sound basis.
E. Fertl and K.Vass,
University of Vienna Medical School,
Universitätsklinik für Neurologie,
Klinische Abteilung Neurologische Rehabilitation
References:
1. Forbes RB, Lees A, Waugh N, Swingler RJ. Population based cost
utility study of interferon beta-1b in secondary progressive multiple
sclerosis. BMJ 1999;319:1529-33
2. Singer P, McKie J, Kuhse H, Richardson J. Double jeopardy and the
use of QALYs in health care allocation. J Medical Ethics 1995;21:144-50
3. Staquet MJ, Hays RD, Fayers PM. Quality of life assessment in
clinical trials. Methods and Practice. Oxford, University Press, 1998, pp
119-41
4. Patrick DL. Quality of life, health status, and vulnerable
populations. Lecture given at 6th ISOQOL, Barcelona, 1999.
5. Fransen M, Edmonds J. Reliability and validity of the EuroQol in
patients with osteoarthritis of the knee. Rheumatology Oxford 1999;38:807-
13
Competing interests: No competing interests