Good day all. Am doing a research on risk factors of COVID-19 among health care workers. My study design is a case-ascertained prospective study.
Can I be able to run a logistic regression with the type of study design? Also how can I handle skip questions for regression analysis? If I want to estimate relative risk using logistic regression, which command should I use in Stata?
If you have time-to-event data (COVID-19), which is right censored (leaving the workplace, death from other causes etc.), then I would recommend using a cox proportional hazard model. If you have full follow-up (no right censoring), then a logistic regression would be more appropiate.
For logistic regression use the command:
Logit "Covid-19" "predictors"..., or (remember to check assumptions).
A logistic regression presents the odds ratio and not the relative risk. However, the odds ratio may be similar when the event (here Covid-19 ) is rare.
