search

ICRC-Models / HHCoM

Compartmental model of HIV and HPV heterosexual transmission, development of AIDS and cervical cancer, and interventions
3 stars 1 forks source link

Relative duration of CIN2 compared to CIN1 and CIN3 #92

Closed carajbro closed 4 years ago

carajbro commented 4 years ago

In the SA model after initial 24Feb20 calib followed by hand-calibration, the time spent in CIN2 is relatively short due to the progression and regression rates, such that there is a lower CIN2 than CIN3 prevalence.

carajbro commented 4 years ago

Gui: For the Kenya model, I had a hard time finding studies that reported separate prevalence estimates for CIN1, CIN2, and CIN3 - most studies combined CIN2 and CIN3 or use the LSIL/HSIL categorization. In the two studies that I found that reported prevalence separately for CIN1, CIN2, and CIN3, CIN2 prevalence was higher than CIN3 in one and lower than CIN3 in the other. And CIN1 prevalence was the highest in both studies. But because there is a lot of uncertainty around CIN2 diagnoses, I have not really tried very hard to match the observed prevalence for CIN2. My approach had been getting the model to sort of match the observed prevalence of CIN1 and CIN3 and putting more emphasis on getting more plausible cancer incidence rates.

I just did a quick search for papers on the duration of CIN1/2/3 and did not find any that estimated time in each state separately.

Darcy: I agree that CIN2 is not really a well-defined state. We decided to keep that state in the model mostly to control the time for people to progress through.

Consensus: Not going to worry too much about this