Rheumatology Advance Access originally published online on March 7, 2008
Rheumatology 2008 47(7):1106-1107; doi:10.1093/rheumatology/ken105
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Comment on: Modelling the cost effectiveness of TNF antagonists in the management of rheumatoid arthritis: results from the British Society for Rheumatology Biologics Registry
Rheumatology Unit, Karolinska Institute, Stockholm, Sweden
Correspondence to: R. F. van Vollenhoven, Department of Rheumatology, The Karolinska University Hospital and Karolinska Institute Reumatologen D2-1, 17176, Stockholm, Sweden. E-mail: ronald.van.Vollenhoven{at}ki.se
SIR, I would like to compliment Brennan et al. [1] on their very interesting paper regarding the cost effectiveness of biologic anti-rheumatic therapy. This study, a methodological tour de force, is a tribute both to the outstanding work by many in establishing the BSRBR, and the possibilities of research based on large and well-designed registers.
Having said that, after reading the article I am left with two questions. The first revolves around a subtle weakness that may have wide repercussions. It is briefly alluded to in the footnote to Table 2, but not otherwise discussed: ‘not all patients in the control group were starting a new DMARD at the time they were recruited ...’. Depending on the degree to which this is the case, it would appear that this fact by itself might explain a significant part of the differences in EULAR responses, differences that are used subsequently to inform the model in terms of the differential increase in utility achieved with biological vs non-biological treatments, and then balanced against the cost. Thus, it would be of interest to know whether the ultimate cost effectiveness results would hold if choosing only those control patients who in fact did receive another treatment at the time of enrolment.
My second question is more conceptual. Brennan et al. estimate cost effectiveness using a model that is (largely) populated with data derived from the register. But why do it this way, when it would be possible, and in fact far easier, to use the data themselves for the calculation—without the need for a model? It would simply entail tabulating the known prices of the treatments given, perhaps adding other measured direct and indirect costs as they are available from the register, adding them up and dividing the sum total by the actual change in utilities as observed (using the SF-6D values from the SF36 and/or the EQ5D as estimated from the HAQ) for the anti-TNF patients and the controls. (Of course, this calculation would only provide information for the cost effectiveness of the first 3 yrs of treatment, but that would be a good start.) One might miss out on some of the refinements in the model, but on the other hand, such a calculation would be much more robust to a range of assumptions that were made by necessity in the model.
Of course, there are only three possible results of such an analysis. It could confirm the findings of the model, which would be nice (no pun intended). It could give a better cost-effectiveness estimate than the model, in which case one might be concerned that the real-time measurements are to some extent biased, but all in all, that would not be a bad result. But most importantly, the actually measured cost effectiveness might be worse than the modelled one, an outcome that would, of course, give us some pause for reflection.
Disclosure statement: The authors have declared no conflicts of interest.
 |
References |
|---|
 Top References |
|---|
- Brennan A, Bansback N, Nixon R, et al. Modelling the cost effectiveness of TNF-a antagonists in the management of rheumatoid arthritis: results from the British Society for Rheumatology Biologics Registry. Rheumatology (2007) 46:1345–54.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  N. Bansback, A. Brennan, D. Symmons, R. Nixon, J. Madan, M. Harrison, and K. Watson Comment on: Modelling the cost effectiveness of TNF-{alpha} antagonists in the management of rheumatoid arthritis: results from the British Society for Rheumatology Biologics Registry: reply Rheumatology, July 1, 2008; 47(7): 1107 - 1107. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||