Toward Optimizing Pertussis Detection During Outbreaks: A Comparison of National and High Epidemic Area Case Definitions in Narathiwat, Thailand, 2024

Abstract

Pertussis remains a significant public health challenge in low-vaccination areas. This study described the characteristics of cases meeting the Narathiwat and national definitions and cases confirmed by reverse transcription polymerase chain reaction (RT-PCR) during the epidemic. It also assessed diagnostic accuracy of the definitions. A cross-sectional analysis utilized data from the Narathiwat Emergency Operations Center during September 2023 to May 2024. For the Narathiwat definition, a suspected case was a person with a cough lasting ≥1 week plus at least one symptom (i.e., paroxysmal cough, post-tussive vomiting, inspiratory whooping, or apnea), while the national definition required cough lasting ≥2 weeks. Among 486 cases, 171 met the Narathiwat definition, 107 met the national definition, and 208 met neither. RT-PCR confirmed cases were 47.9%, 57.9% and 30.8% among cases meeting the Narathiwat, national, and neither definitions, respectively. Timeliness of pertussis detection had a median of 9 days (interquartile range (IQR) 7.8–10 days) for the Narathiwat definition and 16 days (IQR 15–21 days) for the national definition. Paroxysmal cough had the highest detection rate by RT-PCR (84.6%) and was more common among RT-PCR positive cases compared to RT-PCR negative cases. The Narathiwat definition had a sensitivity of 39.4% and a positive predictive value (PPV) of 48.0%. The national definition had a lower sensitivity, 29.8%, but a higher PPV, 57.9%. A proposed alternative definition (cough ≥1 week, accompanied by paroxysmal cough) achieved a sensitivity of 37.5% and a PPV of 49.1%. Modifying the definition improved early outbreak detection during epidemics.