I want to see if there seasonality to a disease

Im looking at an autoimmune disease that has several subtypes, characterized by different autoantibodies (lets call them autoab1, autoab2, autoab3).

I want to see if there is a seasonal pattern to their incidence. (e.g. is autoab1 more frequent during the autumn months?)

I have data testing autoab1, 2 and 3 (negative or positive) spanning several years in the following format:

pt_id   date    autoab  result
001 12/12/2018 autoab1 positive
001 12/12/2018 autoab2 negative
001 12/12/2018 autoab3 negative
002 29/04/2019 autoab1 negative
002 29/04/2019 autoab1 positive
002 29/04/2019 autoab1 negative
003 ...

I tried grouping them by autoantibody and by month and then doing a chi squared. But I don't think it's correct as the dates are actually a continous variable that's binned, and its kind of ordered in a circular way, because the months come one after the other (and jan follow dec!).

In other words i'm trying to answer the following questions "are there peaks of incidence? do they follow a seasonal pattern? (spring-autumn, or summer for e.g.)"

I suspect from watching the plotted data, that autoab1 peaks in the autumn, autoab2 has no clear seasonal pattern and autoab3 peaks in the summer.

Finally, the number of positives is quite low because its a rare disease. I have around 1500 samples, spanning 4 years, every autoab group has between 10 and 50 positive samples overall...


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.