Abstract
BACKGROUND: Mozambique introduced the 10-valent pneumococcal conjugate vaccine (PCV10) in 2013. We aimed to assess PCV10 effectiveness against radiologically confirmed pneumonia (RCP) and invasive pneumococcal disease (IPD) among young children in southern Mozambique. METHODS: We conducted a case-control vaccine effectiveness (VE) study at 2 sites (Mavalane and Manhiça) from January 2014 to December 2016. Children age-eligible to have received PCV10 admitted with pneumonia underwent chest radiograph with standardized interpretation as per the World Health Organization and had nasopharyngeal swabs collected and cultured. IPD cases were recruited in Manhiça only, leveraging existing surveillance; IPD was defined as pneumococcus detected through blood culture or cerebrospinal fluid culture and/or PCR with serotyping by Quellung or PCR. We enrolled age-matched community controls (date of birth ±1 month for cases aged <6 months and +/- 2 months for cases aged ≥6 months). VE was estimated using multivariate logistic regression with 0 doses as a comparator. RESULTS: Among 812 enrolled RCP cases, 780 (96.1%) had matched controls. The median age of included cases was 11.6 months and 51.8% were male; 94.4% of cases and 97.0% of controls had ≥1 PCV10 dose. Adjusted VE against RCP was 23.5% (95% confidence interval [95%CI]: -31.9, 55.6) for 2 doses and 47.2% for 3 doses (95%CI: 13.7, 67.7). Adjusted VE of ≥2 doses and exactly 3 doses was 36.8% (95CI%: -11.9, 64.3) and 41.6% (95CI%: -5.7, 67.7), respectively. Models for VE against RCP with PCV10-type carriage yielded negative point estimates with wide confidence intervals. Among 26 enrolled IPD cases, 24 (92.3%) had matched controls (including 8 PCV10-type cases); 91.7% of IPD cases and 100% of controls had ≥1 PCV10 dose. IPD VE models did not converge. CONCLUSION: PCV10 offers substantial protection against RCP in young children in a high burden setting, highlighting its importance for reducing child pneumonia globally.