05-07-2020, 03:52 PM

Yes, they are the 95% confidence intervals. Here's the table of numeric results:

You can see the effect of confounding by indication; unadjusted hydroxychloroquine patients were over twice as likely to require mechanical ventilation or die. However, using a few methods of adjustment (multivariable analysis, propensity adjustment via inverse probability weighting, matching, and adjustment for propensity scoring) basically gives you the same answer. There is always the possibility of some residual confounding, but again, it is clear that it is not a game changer.

BC

You can see the effect of confounding by indication; unadjusted hydroxychloroquine patients were over twice as likely to require mechanical ventilation or die. However, using a few methods of adjustment (multivariable analysis, propensity adjustment via inverse probability weighting, matching, and adjustment for propensity scoring) basically gives you the same answer. There is always the possibility of some residual confounding, but again, it is clear that it is not a game changer.

BC