Association of Nonpharmaceutical Interventions During the COVID-19 Pandemic With Invasive Pneumococcal Disease, Pneumococcal Carriage, and Respiratory Viral Infections Among Children in France

IMPORTANCE: An association between pneumococcal nasopharyngeal carriage and invasive pneumococcal disease (IPD) has been previously established. However, it is unclear whether the decrease in IPD incidence observed after implementation of nonpharmaceutical interventions (NPIs) during the COVID-19 pandemic was associated with concomitant changes in pneumococcal carriage and respiratory viral infections. OBJECTIVE: To assess changes in IPD incidence after the implementation of NPIs during the COVID-19 pandemic and examine their temporal association with changes in pneumococcal carriage rate and respiratory viral infections (specifically respiratory syncytial virus [RSV] and influenza cases) among children in France. DESIGN, SETTING, AND PARTICIPANTS: This cohort study used interrupted time series analysis of data from ambulatory and hospital-based national continuous surveillance systems of pneumococcal carriage, RSV and influenza-related diseases, and IPD between January 1, 2007, and March 31, 2021. Participants included 11 944 children younger than 15 years in France. EXPOSURES: Implementation of NPIs during the COVID-19 pandemic. MAIN OUTCOMES AND MEASURES: The estimated fraction of IPD change after implementation of NPIs and the association of this change with concomitant changes in pneumococcal carriage rate and RSV and influenza cases among children younger than 15 years. The estimated fraction of change was analyzed using a quasi-Poisson regression model. RESULTS: During the study period, 5113 children (median [IQR] age, 1.0 [0.6-4.0] years; 2959 boys [57.9%]) had IPD, and 6831 healthy children (median [IQR] age, 1.5 [0.9-3.9] years; 3534 boys [51.7%]) received a swab test. Data on race and ethnicity were not collected. After NPI implementation, IPD incidence decreased by 63% (95% CI, −82% to −43%; P < .001) and was similar for non–13-valent pneumococcal conjugate vaccine serotypes with both high disease potential (−63%; 95% CI, −77% to −48%; P < .001) and low disease potential (−53%; 95% CI, −70% to −35%; P < .001). The overall pneumococcal carriage rate did not significantly change after NPI implementation (−12%; 95% CI, −37% to 12%; P = .32), nor did the carriage rate for non-PCV13 serotypes with high disease potential (−26%; 95% CI, −100% to 52%; P = .50) or low disease potential (−7%; 95% CI, −34% to 20%; P = .61). After NPI implementation, the estimated number of influenza cases decreased by 91% (95% CI, −74% to −97%; P < .001), and the estimated number of RSV cases decreased by 74% (95% CI, −55% to −85%; P < .001). Overall, the decrease in influenza and RSV cases accounted for 53% (95% CI, −28% to −78%; P < .001) and 40% (95% CI, −15% to −65%; P = .002) of the decrease in IPD incidence during the NPI period, respectively. The decrease in IPD incidence was not associated with pneumococcal carriage, with carriage accounting for only 4% (95% CI, −7% to 15%; P = .49) of the decrease. CONCLUSIONS AND RELEVANCE: In this cohort study of data from multiple national continuous surveillance systems, a decrease in pediatric IPD incidence occurred after the implementation of NPIs in France; this decrease was associated with a decrease in viral infection cases rather than pneumococcal carriage rate. The association between pneumococcal carriage and IPD was potentially modified by changes in the number of RSV and influenza cases, suggesting that interventions targeting respiratory viruses, such as immunoprophylaxis or vaccines for RSV and influenza, may be able to prevent a large proportion of pediatric IPD cases.