Pneumococcal disease, mainly invasive pneumococcal disease (IPD) and pneumococcal pneumonia (PP), is an major cause of morbidity and mortality. All-cause pneumonia (ACP), 20%–40% of which are PP, causes approximately 8000 deaths here in Spain every year.1

The 23-valent pneumococcal polysaccharide vaccine (VNP23) is recommended for adults. It has been publicly funded in Catalonia since 1999 for people aged 18–64 with risk conditions and for all people aged≥65.

In children, vaccination with pneumococcal conjugate vaccines (PCV7/PCV13) was started later, and was then funded only in high-risk (essentially immunocompromised) children. Universal public funding for PCV13 was implemented in Catalonia from the fourth quarter of 2016 (administered at 2, 4 and 11 months, in accordance with the guidelines of the Spanish National Health System's Interterritorial Council).2

Indirect protective effects of childhood vaccination on the incidence of IPD in the adult population have been reported, but there are limited published data on PP/ACP.3–6

The aim of this study was to analyse and compare the incidence of hospital admission for PP and ACP in the general adult population aged≥50 in Catalonia during the two-year periods 2015–2016 (pre-PCV13 funding) and 2017−2018 (post-PCV13 funding).

We analysed two population cohorts, one for each period, which were made up of all people aged ≥50 assigned to one of the 274 basic health areas managed by the Institut Català de la Salut (ICS) [Catalan Health Institute] throughout Catalonia.

The 2015–2016 cohort consisted of 2,025,730 people aged ≥50 and the 2017−2018 cohort, 2,059,645 people.

The baseline characteristics of both cohorts have been described in previous studies.7,8

The primary data sources were the Sistema de Información para el Desarrollo de la Información en Atención Primaria “SIDIAP” [Information System for the Development of Information in Primary Care]9 of Catalonia (establishment of cohorts) and the Minimum Basic Data Set (MBDS) records (identification of events) of 68 Catalan reference hospitals.

Hospital admissions for PP/ACP occurring among the cohort members were identified using MBDS (ICD-10 codes: J12-J18) in both periods.

The main covariates were age, gender and the presence of risk factors/underlying comorbidity (immunosuppression, chronic lung disease, heart disease, diabetes mellitus and/or smoking). Three baseline risk strata were considered: low (immunocompetent without risk factors); medium (immunocompetent with some risk factors); and high (immunocompromised).

We calculated incidence rates (per 100,000 person-years) in 2015–2016 and 2017−2018, estimating incidence rate ratios (IRR) for PP and ACP between the two periods, overall and specifically for different population subgroups.

During 2015–2016, 24,079 cases of ACP were recorded, 3259 of which were PP, representing an incidence of 617.9 per 100,000 person-years (95% CI: 580.2–657.4) for ACP and 83.6 per 100,000 person-years (95% CI: 78.5–89.0) for PP.

During 2017−2018, 24,136 cases of ACP were detected, 3592 of which were PP, with an incidence of 609.7 per 100,000 person-years (95% CI: 572.5–648.7) for ACP and 90.7 per 100,000 person-years (95% CI: 85.2–96.5) for PP.

Tables 1 and 2 show the absolute number of cases and incidence rates for PP and ACP respectively, as well as incidence rate ratios between 2015–2016 and 2017−2018 according to age groups, gender, comorbidity and baseline risk stratum.

Overall, we found a small increase in the incidence of PP (IRR: 1.09; 95% CI: 1.03−1.14). However, no significant differences were found in the overall incidence of ACP (IRR: 0.99; 95% CI: 0.97−1.00).

We found a slight reduction in the incidence of ACP in people aged 65−79 (IRR: 0.97; 95% CI: 0.94−0.99), while in the population aged≥80, there was a small increase in the incidence of PP (IRR: 1.08; 95% CI: 1.01−1.17).

There was a slight decrease in the incidence of ACP among males (IRR: 0.97; 95% CI: 0.95−0.99), while a significant increase in the incidence of PP was found among females (IRR: 1.15; 95% CI: 1.07−1.24).

A reduction in the incidence of ACP was found in smokers (IRR: 0.92; 95% CI: 0.88−0.97) and in those with chronic lung disease (IRR: 0.96; 95% CI: 0.93−0.99). We found no significant changes in the incidence of PP according to other comorbidities.

ACP decreased in immunocompetent individuals with risk factors (IRR: 0.94; 95% CI: 0.92−0.96). At the same time, the incidence of PP increased in immunocompetent individuals without risk factors (IRR: 1.26; 95% CI: 1.14−1.39).

This study evaluated the incidence of PP/ACP in adults aged≥50 in Catalonia before and after the implementation of universal pneumococcal vaccination in the paediatric population. As main findings, in the population studied as a whole, no significant changes were detected in the overall incidence of ACP. Unexpectedly, however, we identified an increase of approximately 9% in the overall incidence of PP. In the subgroup analysis, significant decreases in ACP were found in people aged 65−79 (–3%), males (–3%), immunocompetent people with risk factors (–6%), people with chronic lung disease (–4%) and smokers (–8%). Analysing the incidence of PP, significant increases were found in people aged≥80 (8%) and in immunocompetent people without risk factors (26%).

Although difficult to interpret, given the variability in the analysis by subgroups, the results show that in the first two years after the implementation of universal PCV13 in paediatrics, there was no early indirect protective effect in the general population aged≥50 in Catalonia (either in terms of reducing the risk of hospital admission for PP or for ACP overall). In fact, there was a slight increase in PP and the incidence rates of ACP were similar in the two periods and we found no statistically significant differences when analysing the entire population.

However, we should underline that the implementation of universal PCV13 vaccination in paediatrics was carried out in those born after 2016 (and not in the entire paediatric population), which could explain the low protective effect found in the adult population in the first two-year period.

In recent years, changes have been observed in the epidemiology of pneumococcal disease (especially in IPD, both in children and adults),10–12 although data on a possible reduction in the total incidence of PP and/or ACP are inconclusive with regard to the adult population.13,14 Unlike in IPD, in Spain there are very limited published epidemiological data assessing the incidence of PP/ACP in both adults in general and the population with risk factors.

The main strengths of this study include the large size and representativeness of the cohorts studied (which included more than two million people aged≥50 and represented almost 75% of the inhabitants of Catalonia in that age group),15 as well as its ability to estimate specific incidence rates of PP and ACP by age subgroups, gender and baseline risk level of the population studied. Among the limitations, mainly linked to its retrospective design, we should point out that out-of-hospital pneumonias were not included (which could mean an underestimation of up to 25% in the estimate of the real incidence of ACP in the population studied). Not all hospitalised patients with pneumonia had microbiological tests (a presumptive diagnosis being established based on clinical suspicion), so possible classification biases cannot be excluded. Some cases considered PP at hospital discharge were not, and vice versa. Among the limitations, we highlight the absence of data on serotypes causing PP (which are not reflected in the MBDS system) and the lack of data on vaccination status in the study subjects.

In conclusion, the incidence of ACP in adults remained stable in the first two-year period after the implementation of public funding of the PCV13 for children in Catalonia. Unexpectedly, a small but significant increase in PP cases was found in the general population aged≥50, suggesting that there was no early indirect protective effect of childhood vaccination on adults in Catalonia.

However, after stratified analysis of the data, a possible protective effect cannot be excluded, either indirectly due to childhood vaccination with PCV13 or directly due to vaccination with VNP23 in people aged 65−79, people with chronic lung disease and people who are immunocompetent with some risk factor.

VTV and CDC wrote and edited the manuscript; ESG and MJF obtained and reviewed the data; VTV and ESG performed the statistical analyses; and OOG and AVR designed and coordinated the study.

This work was funded by a grant from the Health Research Fund of the Instituto de Salud Carlos III [Carlos III Health Institute] (2020 call) for Strategic Action in Health 2020/2030 (file No. PI20/01223), co-financed by the European Union through the European Regional Development Fund (ERDF). This work received funding from the private semFYC Foundation after winning an "Isabel Fernández 2023" grant for the completion of doctoral theses. Institut Universitari per a la recerca a l'Atenció Primària de Salut Jordi Gol (IDIAPJGol) [Jordi Gol University Institute for Research in Primary Health Care] will contribute to publication costs through “Ajuts PhDay 2024”.