From Wiki Journal Club
Jump to: navigation, search


It seems tricky to present results from this paper elegantly.

Two thoughts:

  1. Dumping inconsequential outcomes - really, not much from the outcomes is relevant ~25 years later except the primary outcome and maybe the EF
  2. Presenting the data from the "significant" outcomes by comparing the groups directly rather than doing "a vs. b (P<0.001) vs. c" and adopting this format:
Ejection fraction
ISDN/hydralazine vs. placebo at 8 weeks: +2.9% vs. +0.4% (P<0.001)
Prazosin vs. placebo at 8 weeks: +0.7% vs. +0.4% (NS)
ISDN/hydralazine vs. prazosin at 8 weeks: +2.9% vs. +0.7% (P not specified)

Thoughts? --Tim Plante 18:58, 5 December 2012 (PST)

I think it's ok to present the majority if the outcomes here, because the trial was legitimately asking about hemodynamics and survival separately. But I agree that it's difficult to present them. I like what you've done with the EF outcome above. When I'm at a computer and not my iPhone ill see about updating them. --Dave Iberri 07:57, 6 December 2012 (PST)

So Timmay, I saw the work you did and thought I might fill in some of the details. But now I'm convinced we should just stick to your idea #1, namely to dump the inconsequential outcomes. Bleh. --Dave Iberri 14:40, 6 December 2012 (PST)

And I've just tried something totally different by using tables instead. I need to check it out on iOS to make sure it doesn't look too wonky in the app. How does it look in Android? --Dave Iberri 11:17, 7 December 2012 (PST)

Great work on this so far. What do you guys think of dumping the secondary outcome regarding change in SBP. I think the main outcomes of interest are mortality, and maybe EF. The change in SBP is not clinically significant. A change in SBP of a range between <1 mmHg to 4 mmHg is not very notable. --mlam 22:53, 13 December 2012 (PST)

I want to keep the secondary endpoint of systolic blood pressure change, because the investigators were actually looking at this specifically. The earlier studies had been looking at hemodynamic effects, and this was used to justify ISDN/hydralazine in the past. I think we should keep it. --Dave Iberri 23:24, 13 December 2012 (PST)