We are testing whether screening and management of depression during the pregnancy would improve the birth outcome. We will randomize the cluster (health facility) but at the baseline we will only screen the depression in the intervention group assuming similar prevalence will be in the control group. However, we will measure the birth outcome for both in order to make the comparison. Would this be considered as Cluster RCT? Also, any further suggestion to improve?
You can look at our paper where the end of the possibility to bear children is depicted but depression, anxiety and hostility in the women are measured and can give you some ideas:
Is everyone in the cluster(health facility) going to get the same treatment, and the determination of which health facility gets the treatment is random?. If so, that is a cluster rct. You need as many clusters as possible.
Thank you Kelvyn Jones ! All the pregnant women in the intervention arm will be screened for the depression and will be provided the treatment care services accordingly. However, in the control arm, no such intervention will be done neither the depression will be screened but only the birth outcome as the primary outcome will be measured. Would this outcome be comparable between two arms?
Kelvyn Jones ' point on getting lots of clusters, that number will be important for power. You should look at http://www.bristol.ac.uk/cmm/learning/multilevel-models/samples.html, lots of info from Kelvyn's colleagues. Also, this is unrelated to your question (disclosure, I'm an author), but have a look at Article Development of a Measure of Postpartum PTSD: The City Birth ...
. If you are interested in this we'd be happy to discuss this.ps. it has been translated into 8 languages (https://blogs.city.ac.uk/citybirthtraumascale/translations/), so there is some experience in translating it.
- Badges
- Science topic
- Similar topics
- Computer Science
- Research Topics