I want to compare the mortality of cases versus controls for a cohort of patients with a complication of a viral illness. We have just been notified that we have only 5 case fatalities in our sample. We initially aimed for a case: control sample ratio of 1:2, but given the number of dead patients, I can't decide whether I should give up the idea or look for another statistical solution to the limited number of deaths in my cohort. Do you have suggestions for an acceptable sample size given the above terms? Should I go with a total cohort of 15 patients? Do you have other solutions?
It seems like you have a number of options. If you publish an article with your 5 cases, it should alert others in your field to do similar studies. Right now you could reach out to other centers and look for collaborators who might be interested in working on a similar project to increase your numbers.
In general, my experiences are, rare diseases are often studied using multiple sites and groups. If you have any contact with pharmaceutical companies, they might be willing to share some research methodologies that are helpful. The have experiences studying rare diseases as well as rare side-effect.
good luck to you.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size