- Oral presentation
- Open Access
A novel measure of treatment benefit for an ordinal scale: a case study of the IST-1 and the IST-3 stroke trials
© Thompson et al; licensee BioMed Central Ltd. 2013
- Published: 29 November 2013
- Functional Outcome
- Bootstrap Method
- Ordinal Scale
- Medium Risk
- Treatment Benefit
Relative measures quantify the effect of an intervention but are difficult to translate into practice. Clinicians prefer absolute measures like the number needed to treat (NNT). We demonstrate a novel approach for reporting treatment effect across an ordered outcome where higher scores indicate worse functional outcome. We used the first and the third International Stroke Trials (IST-1 and IST-3) as case studies.
A two by ‘K' table of treatment versus control over an ordinal outcome (with K levels) can be modelled as a multinomial distribution and the probabilities for each cell estimated. We calculated the number of score points gained per 1000 patients treated and estimated the 95% CI using bootstrap methods. Negative values indicate benefit with treatment whilst positive values indicate harm with treatment. We categorised patients into groups of poor functional outcome using prediction models (low (≤35%), medium (35 to 56%), and high (>56%)) and calculated the net gain in functional outcome within each stratum.
The gain in Oxford Handicap Score (OHS) points per 1000 treated in IST-3 for low risk was 14 (95% CI -199 to 240), for medium risk -295 (95%CI -566 to -19) and for high risk -230 (95%CI -396 to -65). The gain in a four level functional outcome score in IST-1 for low risk was -112 (95% CI -214 to -9), for medium risk -42 (95% CI -95 to 12) and for high risk -29 (95% CI -59 to 2).
A ‘net reduction in disability per 1000 patients treated' could be reported alongside the common odds ratio from the proportional odds model.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.