Effect of ten-valent pneumococcal conjugate vaccine introduction on pneumonia hospital admissions in Fiji: a time-series analysis.

Background: In October, 2012, Fiji introduced routine infant immunisation with a ten-valent pneumococcal conjugate vaccine (PCV10) using three primary doses and no booster dose (3 + 0 schedule). Data are scarce for the effect of PCV in the Asia and Pacific region. We aimed to evaluate the effect of PCV10 on pneumonia hospital admissions in children younger than 5 years and adults aged 55 years and older in Fiji, 5 years after vaccine introduction.

Methods: We did a time-series analysis assessing changes in pneumonia hospital admissions at three public tertiary hospitals in Fiji. Four pneumonia outcomes were evaluated: all-cause pneumonia, severe or very severe pneumonia, hypoxic pneumonia, and radiological pneumonia. Participants aged younger than 2 months, 2-23 months, 24-59 months, and 55 years and older were included. Data were extracted from the national hospital admission database according to International Classification of Diseases-tenth revision codes J10·0-18·9, J21, and J22 for all-cause pneumonia. Medical records and chest radiographs were reviewed for the main tertiary hospital to reclassify hospital admissions in children aged younger than 2 years as severe or very severe, hypoxic, or radiological pneumonia as per WHO definitions. Time-series analyses were done using the synthetic control method and multiple imputation to adjust for changes in hospital usage and missing data.

Findings: Between Jan 1, 2007, and Dec 31, 2017, the ratio of observed cases to expected cases for all-cause pneumonia was 0·92 (95% CI 0·70-1·36) for children aged younger than 2 months, 0·86 (0·74-1·00) for children aged 2-23 months, 0·74 (0·62-0·87) for children aged 24-59 months, and 1·90 (1·53-2·31) in adults aged 55 years and older, 5 years after PCV10 introduction. These findings indicate a reduction in all-cause pneumonia among children aged 24-59 months and an increase in adults aged 55 years and older, but no change among children aged younger than 2 months. Among children aged 2-23 months, we observed declines of 21% (95% CI 5-35) for severe or very severe pneumonia, 46% (33-56) for hypoxic pneumonia, and 25% (9-38) for radiological pneumonia. Mortality reduced by 39% (95% CI 5-62) for all-cause pneumonia, bronchiolitis, and asthma admissions in children aged 2-23 months.

Interpretation: The introduction of PCV10 was associated with a decrease in pneumonia hospital admissions in children aged 2-59 months. This is the first study in a middle-income country in the Asia and Pacific region to show the effect of PCV on pneumonia, filling gaps in the literature on the effects of PCV10 and 3 + 0 schedules. These data support decision making on PCV introduction for other low-income and middle-income countries in the region.

Funding: Department of Foreign Affairs and Trade of the Australian Government.