longitudinal data

iuri_leite — Sat, 03 Jun 2023 20:58:32 GMT

Dear all,

I am analyzing a dataset from a cohort of participants in which we observe if they develop specific infection at each quarterly visit. In this dataset a significant number of participants developed the disease more than once. Initially, we used a hazard model considering the time until the first occurrence of the infection.

However, we are now interested in identifying factors associated with more than one occurrence of the disease. Specifically, in the case of two infections, the individual is identified with an infection at a specific visit, receives treatment and then in another visit, the disease is identified again. Since time is an important element, is it possible to conduct a survival analysis in this case? Is it possible to conduct a survival considering an ordinal response?

Thank you in advance.

Best regards,

Iuri Leite

Re: longitudinal data

mkeintz — Sun, 04 Jun 2023 23:51:54 GMT

Why not do the same as you did for the first infection? i.e. do a hazard model to the 2nd infection, by treating records for the first infection as you previously treated records reporting no infection.

Re: longitudinal data

iuri_leite — Thu, 08 Jun 2023 12:19:43 GMT

Dear Mkeintz,

it seems to be a good approach.

Regards,

Iuri

topic Re: longitudinal data in Statistical Procedures

longitudinal data

Re: longitudinal data

Re: longitudinal data