Extend the Repository of HSI_Events
to include Consumables, Equipment and All HSI_Events
#1368
Projects
HSI_Events
to include Consumables, Equipment and All HSI_Events
#1368
We create and publish the catalogue of all the
HSI_Events
that are used in the model here. This aggregation of information is very useful for sense-checking of assumptions by the modelling team (seeing all the information in one place, seeing how similar declarations are made across different disease modules, etc.) and for discussions with stakeholders and external clinical experts. It's also a key part of the methods documentation which we don't want to ask readers to have to find buried in the code. We also useget_details_of_defined_hsi_events()
function in tests (e.g.test_manipulation_of_service_availability()
) to find (almost*) all theHSI_Events
used in the model. [* see second bullet]This issue proposes two extensions:
The information collected for each HSI should also include:
__init__()
thoughself.add_equipment()
apply()
throughself.add_equipment()
apply()
throughself.get_consumables()
It would be helpful if this repository could also include all permutations of
HSI_Events
, including those with characteristics defined only at run-time. (The current method inget_details_of_defined_hsi_events()
usesinspect
to do a "cold read" of the code, whereas some HSI_Events evaluate properties at runtime: e.g.,HSI_Alri_Treatment
.(In fact, these two requirements are linked: I don't think we'd able to to know what is happening in the
apply()
function of anHSI_Event
(for consumables and equipment definitions) without running it.)In the initial approach to this issue (#334), @matt-graham had proposed a way of extracting the information from a run of the simulation, but we didn't use it. But, now we have a nightly run of the model and all the information for the above is coming out in the logger (
tlo.methods.healthsystem
), it seems like it would be possible for this table to be built from the log-file from the nightly run and then assembled into a table. Though not necessary, it would be very nice if this table could be automatically published to the website, but that is probably a part of the website-updating issue (#433).Closed issues that this issue is now covering:
https://www.tlomodel.org/hsi_events.html
#1339The text was updated successfully, but these errors were encountered: