Param Revamp - Pregnancy Supervisor #1733
Conversation
| sim.schedule_event( | ||
| PregnancySupervisorEvent(self), | ||
| sim.date + DateOffset(days=params['initial_event_scheduling_offset_days']) | ||
| ) |
There was a problem hiding this comment.
wonder if we should just delete this as its 0 days...
| # the next week | ||
| if anc_month == 2: | ||
| days_until_anc = self.rng.randint(0, 6) | ||
| if anc_month == params['anc_care_seeking_initial_weeks']/4: |
There was a problem hiding this comment.
i can see what youve done here but maybe the parameter names could be a bit more descriptive...? the code is a little less readible without seeing the values (which obviously is not your fault!!)
There was a problem hiding this comment.
could it be "gestational_age_at_which_anc_is_sought" or something like that?
| poss_day_onset = (27 - 22) * 7 | ||
| current_gestation = df.at[person, 'ps_gestational_age_in_weeks'] | ||
|
|
||
| if current_gestation == 22: |
There was a problem hiding this comment.
not following why these 'floating' values can remain?
| poss_day_onset = (31 - 27) * 7 | ||
| onset_day = self.rng.randint((params['preterm_labour_min_weeks'] - 22) * 7, poss_day_onset) | ||
| elif current_gestation == 27: | ||
| poss_day_onset = (params['preterm_labour_gestation_27_max'] - 27) * 7 |
There was a problem hiding this comment.
i like the numbers in these parameter names - could we do similar for the ones above where anc is sheduled??
| new_pregnancy = df.is_alive & df.is_pregnant & (df.ps_gestational_age_in_weeks == 3) | ||
| new_pregnancy = ( | ||
| df.is_alive & df.is_pregnant & | ||
| (df.ps_gestational_age_in_weeks == params['gestation_week_for_risk_application']) |
There was a problem hiding this comment.
"gestation_week_for_first_risk_application"
| 'anc_care_seeking_initial_weeks': Parameter( | ||
| Types.INT, 'Weeks of gestation at which initial care seeking is applied'), | ||
| 'anc_scheduling_window': Parameter( | ||
| Types.INT, 'Window in days for scheduling of ANC event'), |
|
On you 'general' point above - we can delete any values which arent in the 'value' column. we can keep the references to specific studies however. Just going to look at your labelling of the parameters now |
|
@joehcollins thanks so much for taking the time to review this modifications. Super helpful, and really do agree with your recommendations and the push to rethink some of these parameter names to add clarity to the code. Going to go through your comments/recommendations in more detail. |
no worries! Could you share the criteria for labelling the parameters? I wanted to use that to review but couldnt find it - will do next week |
Yes, for the latest guidance, you can reference the tab 'Guide for Modelers' in this document HERE. Also, I'll be sure to tag you in the other labour modules that I've worked through that would also be helpful to start getting your eyes on. thanks! |
2 participants
Parameters
Assumed
Variables that have been defined as 'assumed' in the write-up because no reliable data source identified for the parameter:
prob_ectopic_pregnancyprob_mod_sev_aphprob_glycaemic_control_diet_exerciseGeneral
prob_ectopic_pregnancy, the value column is populated with [0.01, 0.01] and the Primary Source (or calibration data) value is populated with [0.004937, 0.00366]. It seems that the values from Primary Source (or calibration data) are not used in the module?Local v. Universal
problabeled as local, and those with prefix rr, odds, and aor labeled as localtreatment_effect_ectopic_pregnancy_treatment,treatment_effect_post_abortion_care,prob_glycaemic_control_diet_exercise,treatment_effect_gdm_case_management,treatment_effect_still_birth_food_sups), the rest (related to supplementation and medication) labeled as universal . Currently havetreatment_effect_still_birth_food_supslabeled as local as I believe the type of food supplement may be location dependent.min_age_reproductive&max_age_reproductive: Defined as universaldeath_delay_days_after_abortion&death_delay_days_after_ectopic_rupture: both set to local because has to do with duration of time that treatment/case management can take into effect to prevent deathScenario
alternative_anc_coverageandanc_availability_oddswhich is the target that is set whenalternative_anc_coverageis TRUE)Undetermined, Design Decision
hsi_event_window_days: this parameter is used across multiple HSI events as it is it is the 'buffer' that is used for scheduling. To be discussed if helpful to have as this widely applicable parameter, or if each HSI event should have its own parameter (e.g.hsi_event_first_anc_appt_window_days). I have opted to have them all fall under this singular parameter for readability + simplicityUndetermined, Calibration
prob_still_birth_per_month: Is 'scaled at initiation' the same as calibrated?