search

optimamodel / optima

Optima HIV software tool
http://optimamodel.com
GNU Lesser General Public License v3.0
7 stars 1 forks source link

Age varying mortality #1754

Closed robynstuart closed 6 years ago

robynstuart commented 6 years ago

@cliffckerr please review

Adds a parameter to control HIV death rates by population, as requested by WB. (see https://trello.com/c/KEM2koVP/998-age-specific-mortality-rates).

cliffckerr commented 6 years ago

@robynstuart another way to handle this would be to include it as a Timepar, which would allow variation over both age and time (it could be entered in the databook just under the background mortality rate table). it's a bit messy either way, because people aren't used to entering metaparameters in the databook...

another option would be to introduce additional constants for defined age bands. then it should be fairly straightforward to decide which mortality rate to use for each population the user has entered.

i don't have a strong preference, and probably have a slight preference for the way you've implemented it in this PR, but it might be worth considering these options...

robynstuart commented 6 years ago

@cliffckerr I considered both options:

  1. I started out doing it that way, but decided against it because HIV death rates are currently constant & therefore applied outside the time-loop, and I was concerned it'd slow things down to bring it inside the time loop. Also, I don't know it there's a great deal of evidence for time variation in HIV mortality (other than what's already captured via treatment, obviously)

  2. I considered this too, but preferred to do things this way as it means people can also include higher mortality for groups like PWID, who may be at greater risk of opportunistic infections.

cliffckerr commented 6 years ago

@robynstuart ok, sounds good, will review & merge as-is