analysis

name: analysis description: Skill for analysing outbreaks: estimating severity (CFR), transmissibility (Rt), final size, superspreading, and vaccine effectiveness.

Epiverse-TRACE Analysis

[!IMPORTANT] Use the Introspection Protocol: See epiverse-overview skill for the protocol. Before generating code, verify package APIs and functions using R introspection commands.

[!NOTE] Workflow Context: This skill's code should be included in Quarto document code chunks (not separate R scripts). See the reporting skill for the complete workflow structure.

This skill provides tools for analyzing outbreak data to estimate key epidemiological metrics.

Packages

cfr

Purpose: Estimate disease severity and case ascertainment with delay correction.

Key Functions:

• cfr_static(): Estimate overall CFR for an outbreak
• cfr_rolling(): Estimate time-varying CFR
• cfr_time_varying(): Estimate CFR with temporal changes

Key Features: Accounts for delay from onset to death, provides corrected and naive estimates

finalsize

Purpose: Calculate the final size of an epidemic in heterogeneous populations.

Key Functions:

• final_size(): Calculate epidemic final size given R0
• r_eff(): Calculate effective reproduction number

Key Features: Handles age-structured populations, contact matrices, heterogeneous susceptibility

superspreading

Purpose: Estimate individual-level variation in transmission.

Key Functions:

• probability_epidemic(): Calculate probability outbreak becomes epidemic
• probability_extinct(): Calculate probability of extinction
• proportion_cluster_size(): Proportion of transmission in clusters
• proportion_transmission(): Proportion causing X% of transmission

Key Features: Fits offspring distributions (negative binomial, Poisson-lognormal), estimates R and k

epidemics

Purpose: Composable epidemic scenario modelling.

Key Functions:

• model_default(): SEIR-V model with interventions
• population(): Create population structure
• intervention(): Define contact reduction or vaccination
• model_ebola(): Stochastic Ebola model
• model_vacamole(): COVID-19 vaccination model

Key Features: Contact matrices, age-stratified risks, time-varying parameters, interventions

vaccineff

Purpose: Estimate vaccine effectiveness from cohort and case-control studies.

Key Functions:

• make_vaccineff_data(): Prepare data for VE estimation
• estimate_vaccineff(): Estimate VE using Cox regression
• plot(): Visualize VE results (log-log plots, survival curves)
• summary(): Get VE estimates with confidence intervals

Key Features: Handles cohort design, matching, immunization delays, time-varying effects

Typical Workflow

Severity Analysis

1. Prepare line list with linelist package
2. Get onset-to-death delay from epiparameter
3. Estimate CFR with cfr::cfr_static() or cfr::cfr_rolling()

Transmission Analysis

1. Get generation time from epiparameter
2. Estimate R and k using superspreading or fit offspring distribution
3. Calculate epidemic probability with probability_epidemic()

Scenario Modeling

1. Define population with epidemics::population()
2. Set up interventions with epidemics::intervention()
3. Run model with epidemics::model_default() or disease-specific model

Final Size

1. Get contact matrix (e.g., from socialmixr)
2. Define susceptibility structure
3. Calculate with finalsize::final_size()

Best Practices

• Always account for delays when estimating severity
• Use delay distributions from literature (epiparameter)
• Report estimates with uncertainty (confidence intervals)
• Visualize time-varying estimates to show temporal changes
• Check model assumptions (proportional hazards for CFR)
• Document parameter sources and justifications

Common Pitfalls

• CFR without delay correction: Underestimates severity during active outbreak
• Ignoring heterogeneity: Leads to incorrect final size estimates
• Wrong offspring distribution: Affects superspreading inference
• Misspecified contact matrices: Impacts intervention effectiveness

Integration with Other Skills

• data-intake: Provides cleaned line list data
• parameters: Provides delay distributions for correction
• visualisation: Creates plots of rolling estimates, epidemic curves
• reporting: Synthesizes results into publication-ready document