Review Papers Claim by Claim

1) Endemicity assessment via strain & typing patterns

The supplement describes NICED receiving large numbers of V. cholerae strains annually and using standardized serotyping/phage typing; it then lists multiple states as endemic based on reoccurrence patterns and reports a predominance of Ogawa serotype and specific phage-type patterns. NICED strain typing and endemicity listing

Critical blind spot: “Strain receipt” is not the same as “community incidence.” Receipt depends on referral networks, sampling intensity, and laboratory submission incentives/availability—so endemicity inferred from laboratory sampling can exaggerate geography or miss silent regions without appropriate denominators. The supplement acknowledges missing datasets and surveillance limitations, but the provided text does not show explicit statistical correction for differential strain submission. Potential representativeness gap in strain-based endemicity

2) Case underreporting via literature search vs WHO notifications

The supplement describes a systematic search for cholera/outbreak reports in the biomedical literature for 1997–2006 and compares total cases “observed” to counts “reported to WHO,” reporting a large discrepancy. Literature search vs WHO notifications discrepancy

Counterpoint: Literature-based case counts are themselves subject to publication/reporting bias (e.g., severe or high-visibility outbreaks are more likely to be published). That means both data streams can be biased in different directions; calling the discrepancy “underreporting” is plausible, but not fully falsified by the excerpt alone. Uncertainty about literature-vs-VDP reporting bias

3) Hotspot identification with spatial/risk factor analysis (IDSP + Census)

The supplement describes including 36 administrative entities and 641 districts, classifying 78 districts as hotspots based on reported cases and then using socioeconomic/WASH and other factors to label 111 districts as “at risk.” It associates higher risk with lack of drinking water/drainage, urban amenities, and health access; it links lower risk with better literacy/telecommunications/WASH access. Spatial analysis inputs and risk-factor interpretations

Scientific quality red flag (from the excerpt): The provided text does not specify the statistical model (e.g., thresholds, regression structure, collinearity handling, calibration/validation, or sensitivity analyses). Without those, the hotspot boundary (78 vs 111) is not fully interpretable or reproducible from the excerpt. Missing modeling/validation details in excerpt

• Uncertainty intervals for “>200,000” vs “~37,000,” including how literature search completeness was quantified. Need uncertainty around discrepancy estimate
• Sensitivity analyses: how hotspot/risk classifications change under alternative thresholds or model assumptions. Need sensitivity/validation details
• Representativeness reasoning for NICED strain data: denominator, catchment areas, and submission bias mitigation. Representativeness framework missing in excerpt