@Duke77

J’aime bien l’étude Fiolet, moi !

On y voit par exemple (en autre) ceci :
In sensitivity analysis, after inclusion of studies with critical risk of bias, the global RR was marginally not significant 0.80 (95% CI 0.65–1.00) (see Supplementary material, Table S6).

En gros, Hydroxychloroquine est efficace entre 35% à 0% et est dans tous les cas supérieur à zéro... Et il va conclure l’inverse !

C’est fort, il est trop fort !

Autre exemple :
After excluding studies with critical bias, the pooled RR for COVID-19 mortality was 0.83 (95% CI 0.65–1.06, n = 17 studies) indicating no significant association between hydroxychloroquine and COVID-19 mortality (Fig. 2).

There was a significant subgroup difference between RCT and non-randomized studies (Pheterogeneity between = 0.03) with respectively RRRCT = 1.09 (95% CI 0.97–1.24) and RRnon-randomized = 0.79 (95% CI 0.60–1.04) (Fig. 2).