I am trying to compute the sample size of a cohort study, pregnant women beyond 20 weeks of gestation with normal blood pressure <130/80 mmhg versus pregnant women beyong 20 weeks of gestation with subclinical elevation of blood pressure 130 - 139/ 80 - 89 mmhg. The outcome of interest is onset (incidence) of pregancy induced hypetenstion as per existing threasholds 140/90 mmhg). Recent cohort study of 2,090 normotensive women, 1318 (63.0%) remained normotensive for their entire antenatal course prior to delivery admission and 772 (37.0%) had new-onset blood pressure elevations between 130-139/80-89 mmHg. The incidence of pregnancy induced hypertension in the normotensive group was 11.6% vs 32% in the blood pressure elevation group. https://www.ajog.org/article/S0002-9378(20)30635-9/pdf
Therefore using 0.116 as the proportion of the outcome in the normotensive group, i would like to calculate the sample size for a cohort study powered at 80%, with type I error of 5%.
will the power twoproportions function in STATA sufice? how would I incoperate loss to follow up (drop out) in the said syntax?
0 Response to estimating sample size for cohort studies
Post a Comment