Clustered lot quality assurance sampling to assess immunisation coverage: increasing rapidity and maintaining...

Area of Science:

• Public Health
• Epidemiology
• Biostatistics

Background:

• Vaccination programs aiming for disease elimination require very high coverage levels, often exceeding 95%.
• Accurate and rapid assessment of vaccination coverage is crucial for program managers to ensure targets are met.
• Lot Quality Assurance Sampling (LQAS) is a method used to assess program performance at a local level.

Purpose of the Study:

• To calculate the precision of different clustered lot quality assurance sampling (LQAS) designs using computer simulations.
• To provide field health officers with preset LQAS plans for rapid assessment of high vaccination coverage targets.
• To determine optimal sample sizes and decision values for LQAS plans under various coverage scenarios.

Main Methods:

• Conducted 10,000 simulations to calculate sample size (N), decision value (d), and misclassification errors (alpha and beta) for various LQAS plans.
• Varied upper (UT) and lower (LT) coverage thresholds, with UT at 90% or 95% and LT decreasing by 5% increments.
• Evaluated clustered designs by dividing lots into 5-10 clusters, simulating coverage variation within clusters.

Main Results:

• Precise LQAS plans were identified for 95% coverage targets using N=50 (5 clusters x 10 individuals) with d=4, and for 90% targets with d=7.
• These plans were feasible when the interval between UT and LT was greater than 10% (e.g., 15%).
• The clustered LQAS designs met specified error criteria (alpha < 10%, beta < 25%) when lots were divided into five clusters.

Conclusions:

• The developed preset LQAS plans significantly enhance the feasibility and speed of conducting field assessments.
• These findings support the use of clustered LQAS for efficient monitoring of high-coverage vaccination programs.
• The study provides practical tools for public health officials to rapidly evaluate program effectiveness.