Association between village coverage of 13-valent pneumococcal conjugate vaccine and vaccine-type pneumococcal carriage among children aged 0–59 months with pneumonia in Papua New Guinea: an observational study

Pneumococcal conjugate vaccine (PCV) prevents pneumococcal disease and pneumonia, but indirect effects are poorly understood in low-coverage, high-burden settings like Papua New Guinea (PNG). PNG introduced 13-valent PCV (PCV13) in 2014. We aimed to assess direct and indirect effectiveness of PCV13 against vaccine-type pneumococcal carriage among children with pneumonia or suspected meningitis in PNG. Children aged ≤59 months with pneumonia or suspected meningitis (cases) were enrolled at Eastern Highlands Provincial Hospital and nearby clinics (April 2016–December 2019). Clinical data, PCV13 status, and nasopharyngeal swabs were collected. Pneumococci were detected by lytA real-time PCR and serotyped by microarray. PCV13 coverage among village children ≤59 months was estimated via household surveys. Vaccine-type carriage by case-level PCV13 status was estimated. Associations between carriage and case PCV13 status (direct effects), and village-level coverage (indirect effects) were estimated using GEE logistic regression with village clustering. Among 1214 cases, vaccine-type carriage was lower in vaccinated (24.8%) than under-vaccinated (41.3%) cases. Adjusted PCV13 effectiveness against vaccine-type carriage was 46.6% (95% CI 27.0–60.9). Coverage surveys occurred in 148 villages (5978 children). Median village-level coverage increased from 2016 (12.5% [IQR 0–37.5]) to 2019 (28.6% [IQR 0–53.8]). Each percent increase in village coverage corresponded to a 0.5% (95% CI 0.0–1.0) decrease in vaccine-type carriage. PCV13 vaccination reduced vaccine-type carriage in PNG. Despite low coverage, indirect effects were evident, consistent with emerging herd protection. These findings highlight the importance of strengthening PCV13 programs to optimise individual and community protection. • PCV13 reduced vaccine-type pneumococcal carriage among children in PNG. • Higher village PCV13 coverage was linked to lower vaccine-type carriage despite low/uneven uptake. • Vaccine-type carriage persisted even as PCV13 coverage increased. • Overall carriage stayed high; non-vaccine-type carriage was higher among vaccinated children.