Looking forward to discussing!
Thanks and nice thinking. I've struggled with these myself. For eligibility, I think the authors often are not totally explicit about who they would target in usual care and therefore the eligibility dimension is something one ends up having to guess a little bit on. For example, Rosen might argue that their intervention for same day start isn't meant to be used in persons who are "psychologically unstable" etc., I suppose the question that follows is whether they say so (this our findings only apply or are meant to apply to "stable" patients, as well as whether in routine care settings "stability" is something that can be ascertained. For the last dimension - I think we dinged the Rosen study for excluding people who could be eligible for the intervention (those who could not complete study procedures in one day) in the eligibility category. I think (and i could be wrong) the analysis dimension is restricted to what they do with the data they have on hand. In this case i think they analyzed all of it...
Hi Asa
We mostly gave similar scores to the PRECIS-2 domains in the two studies, and the majority of our arguments were also similar. The largest difference in our scores relate to the primary outcome in the Amanyire study. I also looked at Sanithia’s evaluation of the studies, and she argued similarly to you, so I might be on the wrong track with my justification. By giving just 1 point, I argued that ART initiation was not an outcome that was of significant importance to the patients (because overall, I think that most of the people don’t like to take medications, but they take them because they’re interested in preventing an outcome (such as AIDS, death) that is of primary interest to them). I’m looking forward to discussing this further in class.
We also scored quite differently on the primary analysis domain of the Rosen study. I did not think of it as being a problem that non-eligible patients that were not randomized in the first place were excluded from the analysis, because it seems that this is what happens in most studies, but I’m curious to hear more about this, and I might have missed something there!