topic Re: study design looking at efficacy of vaccine trial in SAS Programming

study design looking at efficacy of vaccine trial

CathyVI — Fri, 29 Oct 2021 18:06:10 GMT

Hi,

Am new to this topic. Suppose I have a vaccine data and am asked to analyze the efficacy. The data was initially analyze for efficacy using Kaplan–Meier estimator and I want to reanalyze the data using another method e.g. cox regression. Would cox regression be appropriate? what other methods can I attempt?

I aim to demonstrate that new treatment is better (superior) than control, my control is placebo. So i think this should be a superiority trial of vaccine vs placebo not active treatment.

How can i get helpful sas code as resources? Thanks

Re: study design looking at efficacy of vaccine trial

Ksharp — Sat, 30 Oct 2021 10:45:31 GMT

Is your data censored ?
proc lifereg or proc phreg do survival analysis, that need your data is censored .
If not, you could try other normal model like : proc freq or proc logistic or proc ttest or proc npar1way or proc glm .....

Re: study design looking at efficacy of vaccine trial

CathyVI — Tue, 02 Nov 2021 19:44:42 GMT

Yes, the data is censored. What other analytic method apart from cox regression can I attempt?

Re: study design looking at efficacy of vaccine trial

Ksharp — Wed, 03 Nov 2021 11:58:12 GMT

The basic method is Wiloncox test ,Check PROC NPAR1WAY .

or could check parameter method PROC LIFETEST .

or calling @StatDave @SteveDenham

Re: study design looking at efficacy of vaccine trial

SteveDenham — Wed, 03 Nov 2021 12:04:43 GMT

If you have access to SAS/ETS, you might look at PROC QLIM for a variety of tobit analyses.

Also, you may wish to look at this paper, as for a vaccine trial you will likely have competing events:

https://communities.sas.com/t5/SAS-Global-Forum-Proceedings/A-SAS-Macro-to-Perform-Kaplan-Meier-Multiple-Imputation-for/ta-p/726336

SteveDenham

Re: study design looking at efficacy of vaccine trial

CathyVI — Sat, 06 Nov 2021 13:02:01 GMT

Thank you @Ksharp @SteveDenham @StatDave. Let me correct a mistake I said earlier. My data is not censored. I do not have any loss to follow up. No participants dropped out. I randomized 31 participants to the vaccine and 45 control, all participants completed the study except 1 control.
Can all the suggested analysis methods will still works in this case?