The assessment of the association between the absence of medical statistics experts and the use of inappropriate/desirable algorithms in multivariate analysis with adjustment for potential confounders
| Factor | Using only significant variables in univariate analysis | Fitting several models for the same outcome and selected factors | ||||
| OR | 95% CI | OR | 95% CI | |||
| Lower | Upper | Lower | Upper | |||
| Medical statistics experts included as first author or coauthor (vs no experts) | 0.28 | 0.15 | 0.53 | 3.51 | 1.88 | 6.58 |
| Medical statistics experts included as first author or coauthor (vs no experts) when first author is clinician or other | 0.42 | 0.19 | 0.97 | 2.36 | 1.03 | 5.38 |
All models were adjusted for impact factor and the number of events.