Should we use logistic mixed model analysis for the effect estimation in a longitudinal RCT with a dichotomous outcome variable?
| Autor: | Adri T. Apeldoorn, Wieke de Vente, Jos W. R. Twisk, Michiel R. de Boer |
|---|---|
| Přispěvatelé: | Epidemiology and Data Science, APH - Methodology, Developmental Psychopathology (RICDE, FMG), Brain and Cognition |
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Předmět: |
0301 basic medicine
Mixed model lcsh:R5-920 030109 nutrition & dietetics lcsh:Public aspects of medicine lcsh:RA1-1270 Estimating equations Missing data Logistic regression Gee 03 medical and health sciences 0302 clinical medicine Standard error Linear regression Statistics Epidemiology RCT Statistical methods 030212 general & internal medicine lcsh:Medicine (General) Mathematics Multinomial logistic regression |
| Zdroj: | Epidemiology Biostatistics and Public Health, 14(3). Prex Epidemiology Biostatistics and Public Health, Vol 14, Iss 3 (2017) Epidemiology, Biostatistics and Public Health, 14(3):e12613. Prex Epidemiology, Biostatistics and Public Health; Vol 14, No 3 (2017) Epidemiology, Biostatistics, and Public Health; V. 14 N. 3 (2017) Epidemiology, Biostatistics, and Public Health; Vol. 14 No. 3 (2017) Twisk, J W R, de Vente, W, Apeldoorn, A T & de Boer, M R 2017, ' Should we use logistic mixed model analysis for the effect estimation in a longitudinal RCT with a dichotomous outcome variable? ', Epidemiology Biostatistics and Public Health, vol. 14, no. 3, pp. e12613-1-e12613-8 . https://doi.org/10.2427/12613 |
| ISSN: | 2282-0930 |
| DOI: | 10.2427/12613 |
| Popis: | Background: Within epidemiology both mixed model analysis and Generalised Estimating Equations (GEE analysis) are frequently used to analyse longitudinal RCT data. With a continuous outcome, both methods lead to more or less the same results, but with a dichotomous outcome the results are totally different. The purpose of the present study is to evaluate the performance of a logistic mixed model analysis and a logistic GEE analysis and to give an advice which of the two methods should be used. Methods: Two real life RCT datasets with and without missing data were used to perform this evaluation. Regarding the logistic mixed model analysis also two different estimation procedures were compared to each other. Results: The regression coefficients obtained from the two logistic mixed model analyses were different from each other, but were always higher then the regression coefficients derived from a logistic GEE analysis. Because this also holds for the standard errors, the corresponding p-values were more or less the same. It was further shown that the effect estimates derived from a logistic mixed model analysis were an overestimation of the ‘real’ effect estimates. Conclusions: Although logistic mixed model analysis is widely used for the analysis of longitudinal RCT data, this article shows that logistic mixed model analysis should not be used when one is interested in the magnitude of the regression coefficients (i.e. effect estimates). |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|