necessary updates for pem mortality model

Author: Ali Bowman

docs/source/models/concept_models/vivarium_nutrition_optimization_extension/concept_model.rst

@@ -51,14 +51,9 @@ Components that need to be altered to support new wasting state (more details to
 
 Components with substantial updates (links/updates to come)
 
-- :ref:`Child wasting exposure <2021_risk_exposure_wasting_with_complicated_sam>`, including changes to:
-
-  - Exposure initialization, see relevant changes in the third (and any later) commits of `this pull request <https://github.com/ihmeuw/vivarium_research/pull/1922>`__
-  - Transition rates, see `relevant changes to modeling strategy in this PR <https://github.com/ihmeuw/vivarium_research/pull/1926>`__. Data values for transition rates still to come as of 4/15/2026
-
+- :ref:`Child wasting exposure <2021_risk_exposure_wasting_with_complicated_sam>`
 - SAM treatment for complicated SAM 
 - :ref:`Child wasting state-specific mortality and morbidity <2021_pem_inpatient_sam_extension>`
-- Mortality due to other causes: set to zero for this simulation (all mortality will be captured in the :ref:`Child wasting state-specific mortality and morbidity <2021_pem_inpatient_sam_extension>` modeling strategy)
 
 .. _nutrition_optimization_extension2.2:
 

docs/source/models/risk_attributable_causes/protein_energy_malnutrition/gbd_2021_inpatient_sam_extension/index.rst

@@ -63,72 +63,111 @@ The following disability weights and excess mortality weights should apply to ea
   * - cat4 (susceptible to child wasting), cat3 (mild childwasting), cat2_better (better MAM), cat2_worse (worse MAM)
     - other_protein_energy_malnutrition
     - 0
-    - daily_probability_to_annual_rate(other_causes_mortality_daily_probability)
+    - mort_rate_other_causes
     - 
   * - cat1_uncomplicated
     - uncomplicated_severe_protein_energy_malnutrition
-    - DW_uncomplicated_sam
-    - daily_probability_to_annual_rate(d1_uncomplicated)
+    - * First pass value: 0
+      * Eventual value if time allows: DW_uncomplicated_sam
+    - mort_rate_uncomplicated_sam
     - 
   * - cat1_complicated
     - complicated_severe_protein_energy_malnutrition
-    - DW_severe_wasting_and_oedema
-    - daily_probability_to_annual_rate(d1_complicated)
+    - * First pass value: 0
+      * Eventual value if time allows: severe_wasting_and_oedema
+    - mort_rate_complicated_sam
     - For simplicity, assumed all complicated SAM cases have DW corresponding with severe wasting and oedema
 
+.. list-table:: Restrictions
+  :header-rows: 1
+
+  * - Restriction
+    - Value
+    - Note
+  * - Age start
+    - 1 month 
+    - 
+  * - Age end
+    - 5 years
+    - 
+  * - Male only
+    - False
+    - 
+  * - Female only
+    - False
+    -
+  * - YLD only
+    - False
+    - 
+  * - YLL only
+    - * First pass: True (set all disability weights equal to zero for simplicity)
+      * Final pass (as time allows): False
+    - Note that simulation implementation should ideally set this value to False to preseve future functionality for non-zero disability weights
+
 .. list-table:: Parameter values
   :header-rows: 1
 
   * - Parameter
     - Definition
     - Value
     - Note
+  * - mort_rate_complicated_sam
+    - Annual mortality rate due to all causes among complicated SAM population
+    - Use value in custom data file linked on the :ref:`wasting exposure document <2021_risk_exposure_wasting_with_complicated_sam>`
+    - Represents mortality due to all causes. Age start for this cause should be 1-5 month age group, but if data is needed for neonatal age groups, fill with zeros
+  * - mort_rate_uncomplicated_sam
+    - Annual mortality rate due to all causes among uncomplicated SAM population
+    - Use value in custom data file linked on the :ref:`wasting exposure document <2021_risk_exposure_wasting_with_complicated_sam>`
+    - Represents mortality due to all causes. Age start for this cause should be 1-5 month age group, but if data is needed for neonatal age groups, fill with zeros
+  * - mort_rate_other_causes
+    - Annual mortality rate due to causes other than diarrheal diseases, lower respiratory infections, malaria, measles, or malaria among non-SAM states
+    - Use value in custom data file linked on the :ref:`wasting exposure document <2021_risk_exposure_wasting_with_complicated_sam>`
+    - Represents mortality due to causes other than modeled infectious diseases in non-SAM states. Age start for this cause should be 1-5 month age group, but if data is needed for neonatal age groups, fill with zeros
+  * - prevalence_{wasting state}
+    - Prevalence of a given wasting state
+    - Defined on the :ref:`wasting exposure document <2021_risk_exposure_wasting_with_complicated_sam>`
+    - 
+  * - csmr_pem
+    - Population-level cause-specific mortality rate for PEM/wasting risk-attributable cause
+    - mort_rate_complicated_sam * prevalence_cat1_complicated + mort_rate_uncomplicated_sam * prevalence_cat1_uncomplicated + mort_rate_other_causes * (prevalence_cat2.5_better + prevalence_cat2.0_worse + prevalence_cat3 + prevalence_cat4)
+    - Age start for this cause should be 1-5 month age group, but if data is needed for neonatal age groups, fill with zeros
   * - DW_severe_wasting
     - Disability weight of severe wasting
     - 0.128 (0.082–0.183)
-    - 
+    - Not needed for first pass 
   * - DW_oedema
     - Disability weight of oedema
     - 0.051 (0.031–0.079)
-    - 
+    - Not needed for first pass 
   * - prev_s199
     - Prevalence of severe wasting without oedema
     - ``get_draws(release_id=9, year_id=2021, gbd_id_type='sequela_id', gbd_id=199, measure_id=5, source='como')``
-    - 
+    - Not needed for first pass 
   * - prev_s198
     - Prevalence of moderate wasting with oedema
     - ``get_draws(release_id=9, year_id=2021, gbd_id_type='sequela_id', gbd_id=198, measure_id=5, source='como')``
-    - 
+    - Not needed for first pass 
   * - DW_uncomplicated_sam
     - Overall disability weight of uncomplicated SAM
     - DW_severe_wasting * prev_s199 / (prev_s199 + prev_s198) + DW_oedema * prev_s198 / (prev_s199 + prev_s198)
-    - We make the simplifying assumption that uncomplicated SAM has no cases with concurrent severe wasting and oedema (these cases are assumed to be complicated SAM) and that the ratio of cases with moderate wasting and oedema to cases of severe wasting without oedema within uncomplicated SAM is equal to the ratio of these two sequelae in GBD.
+    - Not needed for first pass. 
+
+      We make the simplifying assumption that uncomplicated SAM has no cases with concurrent severe wasting and oedema (these cases are assumed to be complicated SAM) and that the ratio of cases with moderate wasting and oedema to cases of severe wasting without oedema within uncomplicated SAM is equal to the ratio of these two sequelae in GBD.
   * - DW_severe_wasting_and_oedema
     - Disability weight of concurrent severe wasting and oedema
     - 1 - (1 - DW_severe_wasting) * (1 - DW_oedema)
-    - 
-  * - d1_complicated
-    - Daily probability of mortality in the complicated SAM state
-    - `See placeholder values here <https://github.com/ihmeuw/vivarium_research_nutrition_optimization/blob/0acb49fac8[…]lation/nigeria/kebbi-copied-dummy-transitions-all-locations.csv>`__ (parameter == 'd1_complicated')
-    - Represents mortality due to all causes 
-  * - d1_uncomplicated
-    - Daily probability of mortality in the uncomplicated SAM state
-    - `See placeholder values here <https://github.com/ihmeuw/vivarium_research_nutrition_optimization/blob/0acb49fac8[…]lation/nigeria/kebbi-copied-dummy-transitions-all-locations.csv>`__ (parameter == 'd1_complicated')
-    - Represents mortality due to all causes 
-  * - other_causes_mortality_daily_probability
-    - Daily probability of mortality due to causes other than diarrheal diseases, lower respiratory infections, malaria, or measles
-    - `See placeholder values here <https://github.com/ihmeuw/vivarium_research_nutrition_optimization/blob/0acb49fac8[…]lation/nigeria/kebbi-copied-dummy-transitions-all-locations.csv>`__ (parameter == 'other_causes_mortality_daily_probability')
-    - Represents mortality due to causes other than modeled infectious diseases in non-SAM states and mortality experienced in SAM states
+    - Not needed for first pass 
+
 
 Validation 
 ++++++++++
 
-As a validation, YLDs due to the wasting-attributable cause model should be similar in magnitude to YLDs due to PEM.
-
 As a verification, EMRs of for the wasting-attributable cause model should match input data
 
 As a verification, ACMR at the population level should match GBD expectation.
 
+Once YLDs due to PEM are included: As a validation, YLDs due to the wasting-attributable cause model should be similar in magnitude to YLDs due to PEM.
+
 References
 ++++++++++