TY - JOUR

T1 - Assessment of assumptions of statistical analysis methods in randomised clinical trials

T2 - The what and how

AU - Nørskov, Anders Kehlet

AU - Lange, Theis

AU - Nielsen, Emil Eik

AU - Gluud, Christian

AU - Winkel, Per

AU - Beyersmann, Jan

AU - De Uña-Álvarez, Jacobo

AU - Torri, Valter

AU - Billot, Laurent

AU - Putter, Hein

AU - Wetterslev, Jørn

AU - Thabane, Lehana

AU - Jakobsen, Janus Christian

PY - 2021

Y1 - 2021

N2 - When analysing and presenting results ofrandomised clinical trials, trialists rarely report if or how underlying statisticalassumptions were validated. To avoid data-driven biased trial results, it should be common practice to prospectively describe the assessments of underlying assumptions. In existing literature, there is no consensus onhow trialists should assess and report underlying assumptions for the analysesof randomised clinical trials. With this study, we developed suggestions on howto test and validate underlying assumptions behind logistic regression, linearregression, and Cox regression when analysing results of randomised clinicaltrials. Two investigators compiled an initial draft based on a review of the literature. Experienced statisticians and trialists from eight different research centres and trial units then participated in a anonymised consensus process, where we reached agreement on the suggestions presented in this paper. This paper provides detailed suggestions on 1) whichunderlying statistical assumptions behind logistic regression, multiple linear regression and Cox regressioneach should be assessed; 2) how these underlying assumptions may be assessed;and 3) what to do if these assumptions are violated. We believe that the validity of randomised clinical trial results will increase if our recommendations for assessing and dealing with violations of the underlying statistical assumptions are followed.

AB - When analysing and presenting results ofrandomised clinical trials, trialists rarely report if or how underlying statisticalassumptions were validated. To avoid data-driven biased trial results, it should be common practice to prospectively describe the assessments of underlying assumptions. In existing literature, there is no consensus onhow trialists should assess and report underlying assumptions for the analysesof randomised clinical trials. With this study, we developed suggestions on howto test and validate underlying assumptions behind logistic regression, linearregression, and Cox regression when analysing results of randomised clinicaltrials. Two investigators compiled an initial draft based on a review of the literature. Experienced statisticians and trialists from eight different research centres and trial units then participated in a anonymised consensus process, where we reached agreement on the suggestions presented in this paper. This paper provides detailed suggestions on 1) whichunderlying statistical assumptions behind logistic regression, multiple linear regression and Cox regressioneach should be assessed; 2) how these underlying assumptions may be assessed;and 3) what to do if these assumptions are violated. We believe that the validity of randomised clinical trial results will increase if our recommendations for assessing and dealing with violations of the underlying statistical assumptions are followed.

KW - epidemiology

KW - statistics & research methods

U2 - 10.1136/bmjebm-2019-111268

DO - 10.1136/bmjebm-2019-111268

M3 - Journal article

C2 - 31988195

AN - SCOPUS:85079393314

VL - 26

JO - BMJ Evidence-Based Medicine

JF - BMJ Evidence-Based Medicine

SN - 2515-446X

IS - 3

M1 - 111268

ER -