Abstract
INTRODUCTION: Pertussis, a contagious respiratory disease, is underreported in adults. The study objective was to quantify underestimation of pertussis cases in adults aged ≥ 50 years in five Latin American countries (Argentina, Brazil, Chile, Mexico, Peru). METHODS: A previously published probabilistic model was adapted to adjust the number of pertussis cases reported to national surveillance systems by successive multiplication steps (proportion of pertussis cases seeking healthcare; proportion with a specimen collected; proportion sent for confirmatory testing; proportion positive for pertussis; proportion reported to passive surveillance). The proportions at each step were added in a random effects model to produce a pooled overall proportion, and a final multiplier was calculated as the simple inverse of this proportion. This multiplier was applied to the number of cases reported to surveillance to estimate the number of pertussis cases. Monte Carlo simulation with 10,000 iterations estimated median as well as upper and lower 90% values. Input data were obtained from surveillance systems and published sources. RESULTS: The estimated median underestimation factor for pertussis cases in adults ranged from 104 (90% limits 40, 451) in Chile to 114 (90% limits 39, 419) in Argentina. In all five countries, the largest estimated number of cases was in the group aged 50-59 years. The highest number per 100,000 population was in the group aged ≥ 90 years in most countries. The estimated median underestimation factor for pertussis hospitalizations was 2.3 (90% limits 1.8, 3.3) in Brazil and 2.4 (90% limits 1.8, 3.2) in Chile (data not available for other countries). CONCLUSION: This analysis indicates that the number of pertussis cases in adults aged ≥ 50 years in five Latin American countries is approximately 100 times higher than the number captured in surveillance data. These results could support decision-making in the diagnosis, management, and prevention of pertussis disease in adults.