Trends in invasive bacterial diseases during the first 2 years of the COVID-19 pandemic: analyses of prospective surveillance data from 30 countries and territories in the IRIS Consortium

Abstract

Background: The Invasive Respiratory Infection Surveillance (IRIS) Consortium was established to assess the impact of the COVID-19 pandemic on invasive diseases caused by Streptococcus pneumoniae, Haemophilus influenzae, Neisseria meningitidis, and Streptococcus agalactiae. We aimed to analyse the incidence and distribution of these diseases during the first 2 years of the COVID-19 pandemic compared to the 2 years preceding the pandemic.

Methods: For this prospective analysis, laboratories in 30 countries and territories representing five continents submitted surveillance data from Jan 1, 2018, to Jan 2, 2022, to private projects within databases in PubMLST. The impact of COVID-19 containment measures on the overall number of cases was analysed, and changes in disease distributions by patient age and serotype or group were examined. Interrupted time-series analyses were done to quantify the impact of pandemic response measures and their relaxation on disease rates, and autoregressive integrated moving average models were used to estimate effect sizes and forecast counterfactual trends by hemisphere.

Findings: Overall, 116 841 cases were analysed: 76 481 in 2018-19, before the pandemic, and 40 360 in 2020-21, during the pandemic. During the pandemic there was a significant reduction in the risk of disease caused by S pneumoniae (risk ratio 0·47; 95% CI 0·40-0·55), H influenzae (0·51; 0·40-0·66) and N meningitidis (0·26; 0·21-0·31), while no significant changes were observed for S agalactiae (1·02; 0·75-1·40), which is not transmitted via the respiratory route. No major changes in the distribution of cases were observed when stratified by patient age or serotype or group. An estimated 36 289 (95% prediction interval 17 145-55 434) cases of invasive bacterial disease were averted during the first 2 years of the pandemic among IRIS-participating countries and territories.

Interpretation: COVID-19 containment measures were associated with a sustained decrease in the incidence of invasive disease caused by S pneumoniae, H influenzae, and N meningitidis during the first 2 years of the pandemic, but cases began to increase in some countries towards the end of 2021 as pandemic restrictions were lifted. These IRIS data provide a better understanding of microbial transmission, will inform vaccine development and implementation, and can contribute to health-care service planning and provision of policies.

Funding: Wellcome Trust, NIHR Oxford Biomedical Research Centre, Spanish Ministry of Science and Innovation, Korea Disease Control and Prevention Agency, Torsten Söderberg Foundation, Stockholm County Council, Swedish Research Council, German Federal Ministry of Health, Robert Koch Institute, Pfizer, Merck, and the Greek National Public Health Organization.

Figure 1

Streptococcus pneumoniae invasive disease case counts For each country or territory, weekly invasive disease cases from Jan 1, 2018, to Jan 2, 2022 (four complete International Organization for Standardization years), were plotted against the weekly Oxford COVID-19 Government Response Tracker stringency index value in 2020–21. The vertical dashed line indicates the week in which pandemic response measures were initiated in each country. *Many of the Spanish sampling dates were submitted only by month and not day of sampling, so the sampling date was entered as the first day of the month if the actual sampling day was unavailable.

Figure 2

Interrupted time-series analyses of invasive disease data in the northern and southern hemispheres Observed cases of invasive disease for each bacterial species (blue solid lines) were plotted against the counterfactual weekly number of cases predicted by the ARIMA models (red dashed lines) if the COVID-19 pandemic had not occurred. The black vertical dashed line indicates the modal month (ie, the most common month when containment measures were put in place). The grey shading depicts 95% prediction intervals. Streptococcus agalactiae data were only collected in the northern hemisphere, and data are plotted by weeks in the calendar year rather than International Organization for Standardization year. The numbers next to ARIMA in the first set of parentheses indicate which components have been included to generate the counterfactual (p,d,q), while the second set of parentheses provides an indication of the seasonal model used (P,D,Q). The square brackets indicate that the model is generated using monthly data (12 months in a year). Please see text for details. ARIMA=autoregressive integrated moving average.

Figure 3

Risk of invasive disease during the pandemic for each bacterial species by hemisphere Results of the meta-analysis are shown as fixed-effects model estimates for each bacterial species.

Figure 4

Streptococcus pneumoniae invasive disease cases by serotype and patient age (A) Distribution of serotypes responsible for 90% of all reported cases between 2018 and 2021, listed by case count, increasing or decreasing trend year by year, and percentage change of each serotype recovered in 2018–19 compared to 2020–21 (average number of cases each year, pre-pandemic vs during the pandemic). (B) Heat map depicting the number of cases of each serotype recovered per year and by age group. ND=not determined.

Figure 5

Haemophilus influenzae and Neisseria meningitidis invasive disease cases by serotype, capsule group, country, and age Distribution of serotypes between 2018 and 2021 for Haemophilus influenzae (A) and Neisseria meningitidis (B) are shown, listed by case count, increasing or decreasing trend year by year, with percentage change of each serotype recovered in 2018–19 compared to 2020–21 (average number of cases each year, pre-pandemic vs during the pandemic). Heat maps depicting the number of cases of each serotype recovered per year and by age group are shown for H influenzae (C) and N meningitidis (D). Number of cases of H influenzae by country are shown (E), displaying only those countries where at least 100 cases in total had been reported across all four study years. Circles represent the total number of cases each year, and lines indicate the increasing or decreasing trend year by year, with all serotypes depicted by the dashed line and H influenzae serotype b (Hib) depicted by the solid line. ND=not determined.