Intended for healthcare professionals

CCBY Open access

Rapid response to:

Research

Prediction models for diagnosis and prognosis of covid-19 infection: systematic review and critical appraisal

BMJ 2020; 369 doi: https://doi.org/10.1136/bmj.m1328 (Published 07 April 2020) Cite this as: BMJ 2020;369:m1328

Linked Editorial

Prediction models for diagnosis and prognosis in covid-19

Read our latest coverage of the coronavirus outbreak

Rapid Response:

Systematic review of prediction models for diagnosis and prognosis of covid-19 infection: applicability?

Dear Editor

I read Wynants et al's systematic review of prediction models for diagnosis and prognosis of covid-19 with interest [1]. The group employ the prediction model risk of bias assessment tool (PROBAST) to scrutinise the models [2]. In essence, this asks 1) if the models are robust (ie. are they at high risk of bias?) and 2) if they are applicable to the population in question (in this case, they highlight various populations to which the models might be applied).

In terms of reporting the authors seem only to cover overall risk of bias. Assessment of applicability is not explicitly reported as per PROBAST. The authors state: "However, the included studies in our systematic review often lacked an adequate description of the study population, which leaves users of these models in doubt about the models’ applicability". A further elaboration of applicability of predictors would be helpful. Without this it is difficult to assess in whom and indeed whether at all the models could be of use in clinical practice. As an example, availability and use of CT scanning will be different between countries and even centres within countries. An excellent model built with CT findings as a predictor variable is unlikely to be generalisable internationally. Even a model with reliable performance is rendered useless if ported to a population in whom data on the predictors cannot be obtained.

I understand that this publication has gone through a Fast Track process and is part of a living review process. I hope the authors will supplement their analysis with information on applicability when this information is available.

References
1. Wynants L, Van Calster B, Bonten MMJ, Collins GS, Debray TPA, De Vos M, et al. Prediction models for diagnosis and prognosis of covid-19 infection: systematic review and critical appraisal. BMJ (Clinical research ed). 2020;369:m1328.

2. Moons KGM, Altman DG, Reitsma JB, et al. Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): explanation and elaboration. Ann Intern Med2015;162:W1-73. doi:10.7326/M14-0698. pmid:25560730

Competing interests: No competing interests

10 April 2020
Fraser Brown
Clinical Fellow
University of Edinburgh
Edinburgh, Scotland