Question
A double-blind study was conducted at 33 centers in seven countries to determine whether supplementation with folic acid or a mixture of seven other vitamins around the time of conception can prevent neural tube defects (NTDs). 1817 women were randomized to one of four groups: folic acid, other vitamins, both, or neither. There were no significant differences among the groups in age and outcome of previous pregnancies. The trial was stopped after one year because a significant protective effect of folic acid was seen. When all women allocated to receive folic acid were compared to all women not receiving folic acid, the measure of association was 0.28 (95% CI: 0.12-0.71). The analysis was repeated excluding women who reported that they had stopped taking their folic acid capsules and investigators found a measure of association of 0.21 (95% CI: 0.07; 0.62). The measure of association for NTDs among women allocated to the “other vitamin” groups compared with those not receiving other vitamins nor folic acid? was 0.80 (95% CI: 0.37-1.72). The measure of association for the “other vitamin” groups, excluding those who stopped taking capsules, was 0.93 (95% CI: 0.41-2.12) [Adapted from: MRC Vitamin Study Research Group, The Lancet 1991;338:131-7].
What type of study design is this? (Circle all that apply)
cluster randomized design
prospective cohort
Cross-over
Factorial randomized trials
What type of measure of association was most appropriate for this study? (Circle one)
POR (prevalence odds ratio)
RR
OR
SMR
What is the purpose of masking in randomized studies? What do you think the authors mean by “double blind” in this study?
What are the authors implying when they state that “there were no significant differences among the groups in age and outcome of previous pregnancies?”
What is the difference between efficacy and effectiveness? Do the authors analyses aim to estimate the efficacy or effectiveness of folic acid?
Comment on whether you think it was ethical to stop the trial after one year.
