Clinical Scorecard: Addressing Methodological Concerns Regarding the 2025 Measles Outbreak Study in Mexico

At a Glance

Key highlights

• The reported interval is a posterior credibility interval of 64.45% to 100.00%, indicating substantial uncertainty regarding immunization coverage recovery.

• The out-of-sample R² of −0.36 indicates poor predictive fit for the 2025 data point and performance worse than a horizontal mean line.

• The 2025 outbreak represented a drastic departure from the preceding 10-year trend.

• Statistical models based on historical data assume continuity and cannot readily account for abrupt, targeted public health interventions.

• The predictive model establishes a baseline risk scenario in the absence of extraordinary programmatic or geographically targeted interventions.

• Integrating predictive assessments with programmatic realities can support more effective policies addressing immunization gaps in Mexico.

Methodological and programmatic implications

Statistical terminology

• The Bayesian uncertainty estimate should be described as a posterior credibility interval rather than a confidence interval.

Model interpretation

• The model’s poor fit for the 2025 data point and exceptionally wide credibility interval limit its reliability for forecasting 2026.

• The projection represents a baseline trajectory rather than the expected effects of extraordinary public health interventions.

Public health interventions

• Complete immunization coverage recovery remains mathematically possible because the credibility interval includes the World Health Organization’s 95% threshold.

• Achieving that recovery would require exceptional measures, potentially including emergency catch-up campaigns, international collaborations, and geographically targeted interventions.

Monitoring

• Model projections should be interpreted alongside programmatic realities, including implemented interventions and geographic patterns.

Data context

• The letter discusses population-level immunization coverage and previously published data from Mexico.

• It presents no new patient-level, diagnostic, treatment, or prescribing data.

• Ethical approval was not required because the letter did not involve new research with human participants or animals.

Key considerations

• Retrospective smoothing cannot adequately capture the 2025 outbreak’s departure from the preceding 10-year trend.

• Predictive models should be integrated with programmatic realities when informing public health policy.

• The baseline projection may change if extraordinary interventions disrupt outbreak progression.

Related resources and content

1. Response to methodological issues regarding the 2025 measles outbreak study in Mexico

2. Original study: Deterioration of Immunization Resilience: A Bayesian Machine Learning Assessment of the 2025-2026 Measles Outbreak in Mexico

3. Methodological commentary on the original study