There are two groups in the study such as group 1 and group 2. One of the groups received treatment, but the other did not. When the mortality of the groups is compared, it seems that there is no statistical difference. However, the expected mortality rate (calculated based on PRISM3 score) in first group ( treatment group) was significantly higher than the other. I think the treatment is successful by lowering the high mortality expectation. However, I could not find how to show this statistically or how can I equalize this imbalance (mortality expectation) between groups at the beginning.
Thanks
.
your expected mortality rate is a confounding variable in your analysis
the adjustment method depends a lot of the data but you can have a look at the following thread
https://www.researchgate.net/post/What_is_the_best_method_to_deal_with_confounding_Stratification_or_matching
if you have a good overlap between the score distribution in both groups (despite different means), you could go for stratification, although it may rule out too small samples
.
Dear Hasan,
Your main analysis should be on an intention to treat basis and therefore should use the outcome as you have planned for.
I hope this helps
.
Mona Kanaan , i'm just curious here since i am compltely out of my field : isn't intention to treat about interpreting randomized controlled trials disregarding the imperfect behaviours (non adherence etc) in the arms ?
If this is right (which is quite a bold hypothesis here, as i may have completely misunderstood what "intention to treat" is !), how does it fit with Hasan Serdar Kihtir 's case study where control and treatment groups indeed seem different from the start (having different expected mortality scores) and therefore should not be the result of a properly conducted randomized controlled trial allocation ?
.
Hasan Serdar Kihtir As far as I understand and guess from your words, you did two analyses, first is a comparison of two KM curves. And second is the comparison of mortality rates. If this is the case, I would say that these two are very different and the results are not surprising.
I recommend you trying a Cox regression (by blind guessing).
I don't think that those two populations have different mortality rates, so an adjustment I believe is not necessary.
- Badges
- Science topic