Hello.
I plan to calculate potentially avoidable mortality rates in four-year periods, adjust them for underregistration, smooth them down by small areas and compare standardized rates. However, someone told me that there were problems when calculating mortality rates for periods longer than three years, which I have not found in other publications. Could you guide me? Thanks in advance
Maylen,
I can not think of an intrinsic problem in the estimation of mortality rates for periods longer than 3 years (besides those you are going to address).
I would be more concern in hiding period or cohort effects because of the size of analytical periods (a methodological decision). You could verify this with sensibility analyses with different period sizes.
At the end it depends on the temporal trends of the outcome.
Yes, that is something I was thinking about! However, those intervals match with municipalities administrative periods, which I want to analyze forward.
I could do the sensibility analysis, as well.
Thank you, Diego
The error between your estimated mortality rates and the actual rates will increase with the time frame. But there is no exact error allowed and therefore no precise cut-off date for predictions.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publications