List differences between Kenya and KZN model structures

[carajbro]

Demography

• Different changes in fertility and mortality over time

Sexual behavior

• Different assumptions about baseline partnerships, acts, and risk distribution by age. Also calibrate this matrix in different ways.
• Kenya model assumes partnerships and risk distribution changes over time, while KZN model does not.

HPV/CIN/CC natural history

• Kenya model assumes non-9v HPV transmission is 1.6x 9v HPV transmission
• Kenya model assumes an additional 1.6x multiplier on HPV acquisition with HIV that the KZN model does not
• Different age multipliers on progression/regression rates (HPV -> CIN1 progression increases with age in the KZN model and decreases with age in the Kenya model; CIN3 -> CC progression increases to a much greater degree with age in the Kenya model compared to the KZN model; other smaller differences in age multipliers between models but with trends in the same direction)

Interventions

• Kenya model assumes historically different screening in HIV-negative and WLWHIV, while the KZN model does not.
• KZN model assumes VMMC protects males from HPV acquisition, while the Kenya model does not.
• KZN model assumes condoms protect both men and women from HPV acquisition, while the Kenya model does not.
• Different scale-up of ART and VMMC

[guiliu53]

An RCT in Uganda suggested that VMMC reduced HPV acquisition in female partners of circumcised men. Since VMMC does not directly protect women from HPV acquisition in the model, we will approximate this protection by reducing HPV acquisition in circumcised men. With this change, the Kenya model will be consistent with the KZN model regarding VMMC protection against HPV acquisition.

[carajbro]

Now calibrating 9v-HPV and non-9v-HPV transmission separately without 1.6x multiplier