08. Analyzers

Analyzers are modules that track simulation outcomes during runtime. They collect, aggregate, and store information about agents and their health, service use, intervention exposure, and social determinants over time — enabling detailed post-simulation analysis, plotting, and cost-effectiveness evaluation.

This page provides an overview of all currently implemented analyzers in MIGHTI, their core functionality, and example use cases.

---

Overview Table

Analyzer Name	Tracks	Stored Data	Source File
DeathsByAgeSexAnalyzer	Age- and sex-specific deaths	- Deaths by age and sex- Cumulative infant deaths	analyzer_core.py
SurvivorshipAnalyzer	Survivorship by age and sex	- Age-specific population counts (alive)	analyzer_core.py
ConditionAtDeathAnalyzer	Conditions present at death	- Agent-level death records with condition flags	analyzer_core.py
PrevalenceAnalyzer_HIV	Disease prevalence by HIV	- Prevalence by HIV status, age, sex	analyzer_prevalence.py
PrevalenceAnalyzer_SDoH	Disease prevalence by social determinants	- Stratified prevalence by housing, education, etc.	analyzer_prevalence.py
InterventionAnalyzer	Intervention delivery	- Yearly records of who received what	analyzer_intervention.py
AdherenceAnalyzer	Adherence behavior	- Agent-level adherence state logs	analyzer_intervention.py
HousingStabilityAnalyzer	Housing exposure	- % stably housed by year	analyzer_intervention.py
HospitalizationAnalyzer	Hospitalizations	- Annual hospitalization counts (by sex/intervention)	analyzer_serviceuse.py
OutpatientVisitAnalyzer (planned)	Clinic visits and outpatient care	- Outpatient visit counts (future)	analyzer_serviceuse.py
PreventiveServiceAnalyzer (planned)	Preventive services (e.g., screenings)	- Service usage logs	analyzer_serviceuse.py
MicrocostingAnalyzer	Cost and YLDs	- Total cost and YLD per agent- Condition-specific breakdowns	analyzer_cost.py

---

Adding a New Analyzer

To implement a new analyzer:

1. Subclass starsim.Analyzer
2. Define init_results() to register output data structures.
3. Override step() to collect data at each timestep.
4. Optionally implement to_df() or custom summary/plot functions.

---

Integration

To use analyzers in a simulation:

analyzers = [
    mi.DeathsByAgeSexAnalyzer(),
    mi.SurvivorshipAnalyzer(),
    mi.ConditionAtDeathAnalyzer(conditions=['t2d', 'hiv']),
    mi.InterventionAnalyzer(),
]
sim = ss.Sim(..., analyzers=analyzers)