Burden of Invasive Pneumococcal Disease and Community-Acquired Pneumonia in Adults: Significant Adverse Outcomes and Costs beyond Direct Treatment Authors: Jenni Kononoff, Veli-Jukka Anttila, Peter Klemets, Marita Kalaoja, Mari Pölkki, Maarit J Korhonen Published: 2026 ABSTRACT AbstractIntroductionInfectious diseases, including invasive pneumococcal disease (IPD) and community‑acquired pneumonia (CAP), are associated with severe burden and adverse outcomes beyond the acute episode. The comprehensive burden has not been previously assessed in Finland. Here, we characterized adult patients (≥ 18 years) with IPD and CAP in Finland, followed them up for adverse outcomes, and estimated the economic burden.MethodsThis is a nationwide registry-based cohort study, including 2894 adults with laboratory-confirmed IPD; 129,203 adults with inpatient CAP; 126,314 adults with outpatient CAP; and their 760,748 age-sex-region-matched controls.ResultsOverall, 40–60% of the patients had at least one predefined medical risk factor; however, a substantial proportion (34–40%) of patients aged ≥ 65 years did not have any predefined risk factors. Patients aged ≥ 65 years with IPD and inpatient CAP had the highest rates of adverse outcomes, including recurrence, admission to institutional care, and acute myocardial infarction. IPD and CAP were associated with considerable attributable healthcare costs, including direct disease-specific costs, increased all-cause costs above them, and indirect costs due to work absenteeism. Serotypes included in the pneumococcal conjugate vaccines (PCVs) PCV13, PCV20, and PCV21 covered 43.0%, 61.1%, and 77.7% of IPD cases, respectively, in the age group ≥ 65 years in 2023–2024.ConclusionsIPD and CAP are associated with substantial morbidity, mortality, and economic burden, particularly among adults aged ≥ 65 years. In this age group, pneumococcal disease alone accounted for ~€39 million annually in total costs (inflated to 2023 values). As many older patients had no predefined medical risk factors, age‑based pneumococcal vaccination and broader use of higher‑valency conjugate vaccines seem warranted to reduce disease burden and societal costs.Supplementary InformationThe online version contains supplementary material available at 10.1007/s40121-026-01357-w.Keywords: Streptococcus pneumoniae, Acute myocardial infarction, Healthcare resource use, Pneumococcal conjugate vaccine, Serotype, Register study Key Summary Points Why carry out this study?The study evaluates important long‑term adverse outcomes, including disease recurrence, admission to institutional care, incidence of acute myocardial infarction (AMI), and mortality and quantifies the direct disease-specific costs and all-cause healthcare costs attributable to invasive pneumococcal disease (IPD) and community-acquired pneumonia (CAP).What was learned from the study?Overall, 40–60% of patients with IPD, inpatient CAP, and outpatient CAP had at least one predefined medical risk factor; however, a substantial proportion (34–40%) of adults aged ≥ 65 years did not have any predefined risk factors.Adults aged ≥ 65 years with IPD and inpatient CAP had the highest rates of adverse outcomes, including recurrence, admission to institutional care, and AMI.At the population level, IPD was associated with approximately €7.5 million and CAP with €270 million in disease-specific costs annually, with Streptococcus pneumoniae accounting for an estimated €39.2 million among individuals aged ≥ 65 years (€4.2 million due to IPD and €35.0 million due to estimated pneumococcal CAP).Serotypes included in the following pneumococcal conjugate vaccines (PCVs): PCV13, PCV20, and PCV21 covered 43%, 61%, and 78% of IPD cases, respectively, in the age group ≥ 65 years in 2023–2024, suggesting that broader use of higher‑valency vaccines may reduce the clinical and economic burden of pneumococcal disease. Introduction Streptococcus pneumoniae is a leading cause of severe respiratory infections, such as community-acquired pneumonia (CAP), as well as invasive infections, including invasive pneumococcal disease (IPD). Both conditions are associated with substantial morbidity, mortality, and healthcare burden [1–4]. Risk factors for IPD and pneumonia are well characterized, including chronic medical conditions, immunocompromised states, and age [5]. Many European countries recommend vaccination universally for all adults aged ≥ 65 years in addition to younger adults with underlying medical conditions [4, 6]. Furthermore, recent European initiatives advocate for achieving a vaccination coverage rate of ≥ 75% among adults aged ≥ 65 years [7, 8]. In Finland, pneumococcal vaccination is currently recommended only for adults with underlying medical conditions by the public health authority Finnish Institute for Health and Welfare [9] but not for all people aged 65 years and older. IPD and CAP are associated with adverse outcomes that extend beyond the initial episode. Pneumonia may recur; the recurrence rates have been reported to be 9–12% over 3–5 years, age being a significant risk factor [10]. Pneumonia may also precipitate cardiovascular complications, including acute myocardial infarction (AMI), through inflammatory and prothrombotic pathways [11, 12]. Laboratory-confirmed respiratory infections have been identified as triggers for AMI and stroke in large population-based studies [13, 14]. Furthermore, recent evidence suggests that certain pneumococcal serotypes may confer a higher risk of major adverse cardiovascular events during IPD episodes [15]. While the highest relative rates of AMI are found closest to the time of the clinical disease, increased incidence of AMI up to 10 years after it has been reported [14]. IPD and CAP have been reported to be associated with increased economic burden, encompassing direct medical costs and indirect societal costs such as productivity loss [16–19]. Pneumonia may exacerbate underlying comorbidities and increase frailty in older adults, further increasing healthcare use and associated costs beyond the acute episode [17]. A recent Finnish study quantified disease-specific healthcare costs for episodes of laboratory confirmed IPD [3]. However, data on increase in all-cause healthcare costs attributable to IPD and CAP and indirect costs related to work absenteeism are scarce. CAP, frequently caused by S. pneumoniae—estimated at 17% of all noninvasive pneumonia cases [20, 21]—also imposes a significant economic impact, yet comprehensive cost analyses in the Finnish setting are not available. The aim of this study was to provide a comprehensive view of IPD and CAP in the adult population of Finland. For addressing the aim, (1) the population aged ≥ 18 years diagnosed with IPD or CAP was characterized in terms of demographics and IPD/CAP risk factors, overall and in the age groups 18–49 years, 50–64 years, and ≥ 65 years; (2) patients were followed up with for recurrence of IPD/CAP, incidence of admission to institutional care and AMI, and mortality in comparison with population controls; (3) direct disease-specific and all-cause healthcare costs and indirect costs attributable to IPD and CAP were estimated; and (4) pneumococcal conjugate vaccine (PCV) serotype coverages of PVC13, PCV20, and PCV21 among recent IPD cases were determined. Methods Study Design and Data Sources This retrospective, noninterventional cohort study used secondary data generated through the universal healthcare and social insurance systems covering the whole population of Finland (5.6 million in 2024). Unique personal identity codes assigned to each resident of the country were used for linkage across data sources. Prior to release, data were pseudonymized by a data permission provider, Findata, the Finnish Social and Health Data Permit Authority. Data contained only structured secondary data. Data on laboratory confirmed episodes of IPD were sourced from the National Infectious Disease Registry (NIDR), which is a national laboratory surveillance system maintained by the Finnish Institute for Health and Welfare. A detailed description of the NIDR data has been published before [3, 22]. Briefly, all clinical microbiology laboratories in Finland report isolations of S. pneumoniae from blood or cerebrospinal fluid to NIDR and, as required by law, submit pneumococcal isolates from the reported cases to the reference laboratory of the Finnish Institute for Health and Welfare. Therein, all strains of S. pneumoniae isolated from sterile fluids are examined. For this study, NIDR provided data on culture dates and serotypes of laboratory confirmed IPD episodes. Data on use of healthcare services, including inpatient stays and outpatient visits to specialized medical and primary care, as well as residence in institutional care were sourced from the care registers of the Finnish Institute for Health and Welfare. The Care Register for Healthcare provided data on hospitalizations in all hospitals of Finland and outpatient visits to public specialized care hospitals. The Register of Primary Health Care Visits provided data on various types of primary healthcare encounters (e.g., general practice office, occupational healthcare and emergency room visits, phone calls, home care) in the public sector for the whole observation period. Data on outpatient visits to private providers became gradually available starting from 2020. The data on institutional care at residential care or other round-the-clock housing facilities came from the Care Register for Social Welfare. For each encounter, information on diagnoses, admission and discharge dates, and type of service provider and encounter were retrieved. Data on entitlements to higher-than-regular (special) reimbursement for medication costs came from the Reimbursement Register maintained by the Social Insurance Institution of Finland (Kela). For receiving entitlement to special reimbursement for their condition, a patient needs a medical statement from a clinician. Data on sickness allowances were obtained from the Kela Register for Seasonal Benefits. Sick leaves lasting less than 10 days were not included, as they are not recorded in the Kela Register. Information on sex and dates of birth and death was provided by the Digital and Population Data Services Agency (DVV). Study Population The patient population consisted of residents of continental Finland aged ≥ 18 years with IPD or CAP in 2016–2019. That is, the cohort entry period was prior to coronavirus 2019 (COVID-19). First, all adults with episodes of laboratory-confirmed IPD identified from NIDR during the cohort entry period were included. A patient could enter the cohort multiple times during the entry period but only once during a calendar year. As NIDR combines potential multiple positive culture results within 3 months for the same individual into a single episode, the same patient could not enter the cohort within 90 days. The date of the first positive culture for S. pneumoniae within a calendar year was defined as the index date. Regarding CAP, all adults with an in-person healthcare encounter (inpatient stay or visit) for pneumonia during the cohort entry period, but no laboratory-confirmed IPD during the same calendar year, were eligible (Supplementary Fig. S1). Healthcare encounters with pneumonia were identified with the International Classification for Diseases, 10th version (ICD-10) codes J10–J18 and J86, or International Classification for Primary Care, 2nd version (ICPC-2) code R81 (Supplementary Fig. S1) as primary or secondary diagnosis. Pneumonia-related healthcare encounters within a calendar year were evaluated chronologically using an iterative process of one to four rounds. In each round, the first encounter following a ≥ 90-day pneumonia-free period was designated as the preliminary index date for a new pneumonia case. Then, the following exclusion criteria were employed for preliminary index dates: laboratory-confirmed IPD or a healthcare encounter for IPD (ICD-10 codes: A40.3, B95.3, G00.1, M00.1) within 90 days before or after the index date, hospitalization for any cause within 2 days before the index date (for excluding hospital-acquired pneumonia), a diagnosis code for ventilator-associated pneumonia (ICD-10 code: J95.8) linked with the index hospitalization, and age under 18 years. Three controls were randomly sampled by DVV from the general population matched to each patient with IPD/CAP by age (± 1 year), sex, and municipality of residence at the patient’s index date. The same exclusion criteria were employed for patients and controls. In addition, for describing the serotype coverage of PCV13, PCV20, and PCV21 (Supplementary Table S1), data on all episodes of laboratory-confirmed IPD occurring in the adult population of Finland in 2023–2024 were obtained from NIDR. The analyses of IPD and CAP risk factors, outcomes, and costs were based on cases identified during the pre‑COVID‑19 period (2016–2019) to ensure estimation of clinical and economic burden under stable epidemiological conditions. In contrast, the analysis of pneumococcal serotype coverage was conducted using the most recent available national surveillance data on laboratory‑confirmed IPD episodes from 2023 to 2024. The purpose of analysis was descriptive and epidemiological, characterizing the current distribution of circulating pneumococcal serotypes in adults, and the proportion of these serotypes included in licensed pneumococcal conjugate vaccines, reflecting the potentially vaccine‑preventable disease burden. Exposure The main exposure categories were IPD, inpatient CAP, and outpatient CAP. Patients with CAP were divided into inpatient and outpatient CAP cohorts based on the type of healthcare encounter at the index date. However, those patients initially diagnosed with CAP in the outpatient setting but hospitalized with CAP within 14 days of the index date were classified into the inpatient CAP cohort. Covariates Age, sex, and residence in institutional care were measured at index date. The comorbid conditions considered as risk factors for IPD/CAP included: (1) chronic obstructive pulmonary disease (COPD) and asthma, (2) chronic heart diseases (including coronary heart disease, cardiac failure, arrythmias), (3) diabetes (mellitus), (4) inflammatory diseases (including, rheumatic or other autoimmune diseases requiring immunosuppressive therapy), (5) cancer, and (6) renal disease/solid organ or bone marrow transplantation. The presence of comorbid conditions was assessed at index date and defined as having a Kela record indicating a valid entitlement to special reimbursement of medication costs for the abovementioned conditions. Individuals without any of the preceding comorbid conditions were classified as having no risk factors. History of IPD/CAP was defined as any healthcare encounter for CAP or an episode of laboratory-confirmed IPD 91–365 days before the index date. Outcomes The outcomes of interest included recurrent IPD/CAP, admission to institutional care, AMI, death from any cause, and direct disease-specific and all-cause healthcare costs and indirect costs. Recurrent IPD/CAP refers to an episode of laboratory-confirmed IPD or a physical encounter for CAP that was separate from the index episode or encounter; that is, it began over 90 days after the index date. Among controls, this outcome refers to IPD/CAP occurring at any point during the follow-up. For those who did not live in institutional care at the index date, the first admission to institutional care after the index date was identified from the Care Register for Social Welfare; that is, hospitalizations were not included in this outcome. AMI was defined as hospitalization or outpatient visit to specialized care with an ICD-10 code I21 (acute myocardial infarction) or I22 (subsequent myocardial infarction) identified from Hilmo after the index date. For recurrent IPD/CAP, admission to institutional care, and AMI, the follow-up extended up to a maximum 4–7 years (depending on the time of cohort entry) from the index date to the first of migration, death, or the administrative end of the study (31 December 2022). Deaths from any cause as an outcome were identified over 7- and 30-day and 3-year periods from the index date. Cost Assessment Healthcare costs refer to costs associated with healthcare resource use (HCRU). Disease-specific healthcare costs were assessed over a 1-year period (365 days) following the index date and all-cause healthcare costs over 1-year periods before and after the index date. HCRU was defined as disease-specific when at least one of the ICD-10 codes A40.3, B95.3, G00.1, M00.1, J10–J18, or J86 or the ICPC-2 code R81 was among the diagnoses linked with the encounter. As in older adults particularly, it may be difficult to differentiate between the HCRU associated with the treatment of the IPD/CAP and the exacerbation of comorbidities, the increase in all-cause healthcare costs from 1 year before to 1 year after the index date was also assessed and interpreted as attributable to the index IPD or CAP. Total disease-specific costs included sickness allowances as indirect costs. Only sick leaves with IPD/CAP as the main diagnosis were considered. Finally, average annual disease-specific costs accumulated by each cohort were assessed, further dividing HCRU-associated costs into inpatient and outpatient costs. The HCRU-associated costs (in euros) were estimated using respective national unit costs [23] for observed HCRU. The costs of inpatient stays, emergency room, and other outpatient visits in specialized care were calculated using average costs per hospital day or type of visit across all hospitals. Average daily costs of inpatient stays at primary healthcare (regardless of length of stay) were used for valuing hospital days at inpatient units of health centers. Primary care encounters were grouped by type (e.g., outpatient visit, call, or home care) and further by service form and a healthcare professional. The unit costs include all costs from the service provider, such as costs of personnel, materials, facilities, in-hospital medication, and procedures and examinations performed. Unit costs were inflated to 2023 values using the national healthcare price index from Statistics Finland. For sickness allowances, the actual costs reimbursed by Kela were used while the costs for the first 9 days paid by the employer were not available. The share of annual total disease-specific costs caused by pneumococcal pneumonia were calculated on the basis of two Finnish studies that estimated pneumococcal pneumonia to account for approximately 17% out of all pneumonias [20, 21]. One of the estimates was based on data on urinary antigen tested adults (positive for S. pneumoniae) in one hospital district in Finland in 2019 [21] and the other on test results from samples (sputum, urine, nasopharyngeal swab, and blood) of patients aged ≥ 65 years with radiologically confirmed CAP [20]. That is, here, the proportion of 17% was used when estimating the share of total disease-specific costs that were due to pneumococcal pneumonia. Frequencies of Streptococcus pneumoniae Serotypes Covered by PCVs The serotype coverages of PCV13, PCV20, and PCV21 of IPD episodes reported to NIDR during the years 2023–2024 are described. The coverage is further compared between higher-valency PCVs (PCV20 and PCV21), dividing the serotypes to those common to both and those unique to each PCV. Ethical Approval These retrospective secondary registry data were permitted by Findata, the Social and Health Data Permit Authority (permit diary numbers: THL/3226/14.02.00/2022, THL/2333/14.06.00/2024, THL/3005/14.06.00.2024, and THL/2792/14.02.00/2025). Study was conducted following the Declaration of Helsinki, national legislation and regulations of Finland, legislation of European Union (EU), and the International Society of for Pharmacoepidemiology and Guideline for Good Pharmacoepidemiology Practice. Ethical evaluation was provided by the permit authority, Findata. According to the local legislation of the Finnish Act on the Secondary Use of Health and Social Data (552/2019), secondary use of registry data is considered to be noninterventional study, and no further evaluations were required. In accordance with the Findata data protection policies, individual-level results (N < 3) are not reported. Anonymity of the results was inspected prior to publication by the permit authority. All individual-level data and analytic datasets were handled in secured environment provided by permit author. Statistical Analyses Characteristics of each patient cohort and their population controls, including age, sex (female/male), residence in institutional care, and presence of comorbidities considered as risk factors for IPD/CAP (one or more/none) at index date and history of IPD/CAP within a year preceding the index date were summarized and stratified by age group (18–49/50–64/≥ 65 years). In addition, prevalence of each risk factor is reported as percentages for the three age groups. All outcome analyses were stratified by patient cohort and age groups. Due to variable follow-up time, cumulative incidence of adverse outcomes, including recurrent IPD/CAP, admission to institutional care, and AMI was estimated with the Kaplan–Meier method, and crude rates are reported per 100 person years (PY) with 95% confidence intervals (CIs). Incidence of adverse outcomes over the whole follow-up period was compared between patients and controls with competing-risks regression, and the results are expressed as hazard ratios (HR) with 95% confidence intervals (CI). For AMI, an additional analysis was conducted among patients and controls free of ischemic heart disease at index date by excluding all those who had a valid entitlement for special reimbursement for coronary heart disease and/or a record of a healthcare encounter for ischemic heart disease (ICD-10 codes I20–I25) within 4 years preceding the index date. All-cause mortality is reported as percentages (with 95% CIs) of those who died over 7-day, 30-day, and 3-year periods. Disease-specific and all-cause costs (€) are presented per PY (PPY) to account for loss to follow-up due to mortality (or migration) during the 1-year observation period after the index date. Mean annual disease-specific costs, including direct inpatient and outpatient healthcare costs and indirect costs, were estimated for each patient cohort by dividing the total costs accumulated over a 1-year period by the cohort in question by four. Finally, the annual disease-specific costs due to pneumococcal pneumonia were estimated [20, 21]. Results Description of the Study Population In total, 2894 adults with IPD were identified from NIDR in 2016–2019, with annual numbers ranging from 680 to 764. The total number of adults with inpatient and outpatient CAP identified from the care registers who met the inclusion and exclusion criteria was 129,203 and 126,314, respectively (Supplementary Fig. S1). For inpatient CAP, the annual numbers ranged from 30,527 to 33,712 and from 28,978 to 34,353 for outpatient CAP. The baseline characteristics of each patient cohort and their matched controls are summarized in Table 1. For patients with IPD, the mean age was 65.1 years (standard deviation, 16.73), for patients with inpatient CAP 73.0 years (16.54), and for those with outpatient CAP 58.6 years (20.52). Of the IPD cohort, 18.1% were 18–49 years, 26.3% were 50–64 years, and 55.6% were ≥ 65 years of age. In the inpatient CAP cohort, the respective proportions were 10.2%, 14.5%, and 75.3% and, in the outpatient CAP cohort, 34.2%, 21.9%, and 43.9%. Residence in institutional care was most common among patients aged ≥ 65 years with inpatient or outpatient CAP (~14%) (Table 1). Table 1Baseline characteristics of patients with IPD, inpatient or outpatient CAP, and their population controls stratified by age group and overallCharacteristic18–49 years of age50–64 years of age ≥ 65 years of ageOverallPatientControlPatientControlPatientControlPatientControlIPD, N524157076022811610481028948661 Sex, n (%) Female196 (37.5)588 (37.4)308 (40.5)925 (40.6)828 (51.4)2474 (51.4)1332 (46.0)3987 (46.0) Male328 (62.6)982 (62.5)452 (59.5)1356 (59.4)782 (48.6)2336 (48.6)1562 (54.0)4674 (54.0) Institutional care, n (%)6 (1.1)7 (0.4)18 (2.4)9 (0.4)93 (5.8)208 (4.3)177 (4.0)224 (2.6) History of IPD/CAP, n (%)37 (7.1)7 (0.4)68 (8.9)22 (1.0)276 (17.1)110 (2.3)381a (13.2)139 (1.6) Risk factors, n (%) 1 or more99 (18.9)123 (7.8)334 (43.9)471 (20.6)1008 (62.6)1949 (40.5)1441 (49.8)2543 (29.4) None425 (81.1)1447 (92.2)426 (56.0)1810 (79.4)602 (37.4)2861 (59.5)1453 (50.2)6118 (70.6)Inpatient CAP, N13,18539,41518,71555,85097,303282,532129,203377,797 Sex, n (%) Female6204 (47.1)18,538 (47.0)7918 (42.3)23,628 (42.3)46,172 (47.5)134,374 (47.6)60,294 (46.7)176,540 (46.7) Male6981 (52.9)20,877 (53.0)10,797 (57.7)32,222 (57.7)51,131 (52.5)148,158 (52.4)68,909 (53.3)201,257 (53.3) Institutional care, n (%)644 (4.9)130 (0.3)1400 (7.5)254 (0.5)13,858 (14.2)20,140 (7.1)15,902 (12.3)20,524 (5.4) History of IPD/CAP, n (%)736 (5.6)141 (0.4)1576 (8.4)288 (0.5)10,178 (10.5)5693 (2.0)12,490b (9.7)6122 (1.6) Risk factors, n (%) 1 or more3167 (24.0)3264 (8.3)9317 (49.8)12,280 (22.0)64,269 (66.1)119,731 (42.4)76,753 (59.4)135,247 (35.8) None10,018 (76.0)36,151 (91.8)9398 (50.2)43,570 (78.0)33,034 (33.9)162,801 (57.6)52,450 (40.6)242,550 (64.2)Outpatient CAP, N43,175129,05727,66282,55755,477162,676126,314374,290 Sex, n (%) Female24,123 (55.9)72,098 (55.9)14,793 (53.5)44,141 (53.5)28,056 (50.6)82,444 (50.7)66,972 (53.0)198,683 (53.1) Male19,052 (44.1)56,959 (44.1)12,869 (46.5)38,416 (46.5)27,421 (49.4)80,232 (49.3)59,342 (47.0)175,607 (46.9) Institutional care, n (%)553 (1.3)340 (0.3)768 (2.8)335 (0.4)7539 (13.6)8154 (5.0)8860 (7.0)8829 (2.4) History of IPD/CAP, n (%)2001 (4.6)465 (0.4)1949 (7.0)433 (0.5)5755 (10.4)2738 (1.7)9705c (7.7)3636 (1.0) Risk factors, n (%) 1 or more6373 (14.8)10,001 (7.7)10,424 (37.7)17,540 (21.2)32,893 (59.3)64,947 (39.9)49,690 (39.3)92,488 (24.7) None36,802 (85.2)119,056 (92.3)17,238 (62.3)65,017 (78.6)22,584 (40.7)97,729 (60.2)76,624 (60.7)281,802 (75.3)CAP community-acquired pneumonia, IPD invasive pneumococcal disease, n numberan = 8 patients with laboratory-confirmed IPD within 91–365 days before the index datebn = 99 patients with laboratory confirmed IPD within 91–365 days before the index datec n = 87 patients with laboratory confirmed IPD within 91–365 days before the index date Baseline characteristics of patients with IPD, inpatient or outpatient CAP, and their population controls stratified by age group and overall CAP community-acquired pneumonia, IPD invasive pneumococcal disease, n number an = 8 patients with laboratory-confirmed IPD within 91–365 days before the index date bn = 99 patients with laboratory confirmed IPD within 91–365 days before the index date c n = 87 patients with laboratory confirmed IPD within 91–365 days before the index date Risk Factors Overall, 49.8% of the patients with IPD, 59.4% of those with inpatient CAP, and 39.3% of those with outpatient CAP had at least one risk factor for IPD/CAP (Table 1). The respective proportions among the matched controls were 29.4%, 35.8%, and 24.7%. The proportion of individuals with no risk factors decreased with age. Still, in the age group of 65 years and older, 34–40% of the patients did not have other risk factors at the index date. Among controls, this proportion was around 60%. The prevalence of each risk factor in the patient cohorts with IPD, inpatient CAP, and outpatient CAP stratified by age group is presented in Fig. 1a–c. The most common comorbidity was chronic heart diseases, their prevalence being ~30% in each patient cohort aged ≥ 65 years. About 20% of the patients in the oldest age group suffered from asthma or COPD and 20–24% from diabetes. Prevalence (%) of risk factors in A IPD, B inpatient CAP, and C outpatient CAP cohorts by age group. Sample sizes, IPD: 18–49 years, N = 524; 50–64 years, N = 760; and 65+ years, N = 1610. Inpatient CAP: 18–49 years, N = 13,185; 50–64 years, N = 18,715; and 65+ years, N = 97,303. Outpatient CAP: 18–49 years, N = 43,175; 50–64 years, N = 27,662; and 65+ years, N = 55,477. IPD invasive pneumococcal disease, CAP community-acquired pneumonia, COPD chronic obstructive pulmonary disease Adverse Outcomes Recurrence of IPD and/or CAP Altogether, 827 (28.6%) patients with IPD had a recurrent IPD or CAP during the mean 2.97 years of follow-up. In the inpatient CAP cohort, the number of patients with IPD/CAP recurrence was 41,929 (32.5%) over the mean 2.43 follow-up years, and in the outpatient CAP cohort it was 24,580 (19.5%) over the mean 3.87 follow-up years. Figure 2A demonstrates considerable differences in the cumulative incidence of IPD/CAP during the follow-up across the age groups for each patient cohort and between patients and controls. Overall, the incidence of IPD/CAP during the follow-up was observed to be around 5.5–7.5-fold among patients compared with their matched controls (Supplementary Table S2). While the IPD/CAP incidence was clearly the highest in the oldest age group, the relative rates declined with age. When comparing patients with IPD and their controls, the HR (95% CI) for recurrent IPD/CAP was 17.2 (10.8–27.4) in 18–49-year-olds and 5.81 (5.05–6.69) in ≥ 65-year-olds. For the inpatient CAP cohort versus controls, the respective HRs (95% CI) were 19.8 (18.0–21.8) and 4.87 (4.87–4.91) and for the outpatient CAP cohort versus controls 8.09 (7.65–8.56) and 4.64 (4.52–4.75). A Cumulative incidence of recurrent IPD/CAP in patients with IPD, inpatient CAP, and outpatient CAP and matched controls by age group. B Cumulative incidence of admission to institutional (24–h) care in patients with IPD, inpatient CAP, and outpatient CAP and matched controls by age group. C Cumulative incidence of acute myocardial infarction (AMI) in patients with IPD, inpatient CAP, and outpatient CAP and their matched controls by age group. IPD invasive pneumococcal disease, CAP community-acquired pneumonia Admission to Institutional Care Of the patients with IPD not residing in institutional care at the index date (n = 2777), 323 (11.6%) were admitted to such care during the follow-up. In the inpatient CAP cohort (n = 113,301), the respective number was 28,054 (24.8%), and in the outpatient CAP cohort (n = 117,454) 10,271 (8.7%). Overall, the cumulative incidence of institutional care admission was around 1.5–2-fold among patients in comparison with their controls (Supplementary Table S2). Again, the crude incidence increased steeply with age (Fig. 2b), while relative rates compared with controls declined. The highest relative rates were seen in the inpatient CAP cohorts: in the age groups 18–49 years and 50–64 years, the HRs were 9.17 (95% CI 8.15–10.30) and 9.08 (8.35–9.87) and 2.05 (2.02–2.09) in those aged ≥ 65 years. AMI In the IPD cohort, 101 (3.5%) patients experienced an AMI over the mean 3.69 years of follow-up (crude incidence 0.95/100 PY, 95% CI 0.77–1.15). In the inpatient CAP cohort, the number of patients with AMI was 8262 (6.4%) over the mean of 2.99 years of follow-up (crude incidence 2.14/100 PY, 2.09–2.18). In the outpatient CAP cohort, the respective number was 3874 (3.1%) during the mean 4.29 follow-up years (0.71 per 100 PY, 0.69–0.74). AMI incidence increased with age and remained consistently higher among patients than controls throughout the follow-up (Fig. 2c). The HRs for AMI ranged from 1.06 to 3.45, being the highest for inpatient CAP (Supplementary Table S2). Mortality Of the 2894 patients with IPD, 828 (28.6%) died within the 3-year mortality follow-up. During the first week (7 days) and month (30 days) when the death is likely to be attributable to the index IPD, mortality rates were 6.1% and 9.4% (Fig. 3a–c). Large differences in mortality were seen across age groups in each follow-up period; mortality increased from 1.9% in 18–49-year-olds to 8.3% in ≥ 65-year-olds at 7 days, from 3.1% to 12.8% at 30 days, and from 7.4% to 41.1% at 3 years. Among the matched controls, mortality was clearly lower than among patients in all age groups and overall (Supplementary Table S3). A 7-day, B 30-day, and C 3-year mortality (%) in IPD, inpatient CAP, and outpatient CAP cohorts by age group and overall. Sample sizes, IPD: 18–49 years, N = 524; 50–64 years, N = 760; 65+ years, N = 1610; and overall N = 2894. Inpatient CAP: 18–49 years, N = 13,185; 50–64 years, N = 18,715; 65+ years, N = 97,303; and overall, N = 129,203. Outpatient CAP: 18–49 years, N = 43,175; 50–64 years, N = 27,662; 65+ years, N = 55,477; and overall, N = 126,314. IPD invasive pneumococcal disease, CAP community-acquired pneumonia Altogether 56,636 patients with inpatient CAP (43.8%) and 20,980 with outpatient CAP (16.6%) died during the 3-year mortality follow-up (Fig. 3c). Again, large increases in mortality were observed from the youngest to the oldest age group for each follow-up period (Fig. 3a–c). In the inpatient CAP cohort, mortality increased from 0.5% among the 18–49-year-olds to 4.9% among the ≥ 65 years at 7 days, from 1.3% to 12.4% at 30 days and from 7.1% to 52.6% at 3 years. The respective increases in the outpatient CAP cohort were from 0.1% to 3.8%, from 0.2% to 7.1%, and from 1.2% to 33.2%. Among the patient cohorts, 3-year mortality was the highest in the inpatient CAP cohort overall (43.8%) and in age groups 50–64 years (23.9%) and ≥ 65 years (52.6%). This is likely to reflect the older age distribution of patients with inpatient CAP versus other patients as a similar pattern was seen among the matched controls (Supplementary Table S3). In each age group and overall, mortality remained considerably lower among controls than in the respective outpatient CAP cohorts. Costs Disease-Specific and All-Cause Healthcare Costs Per Patient Year The disease-specific healthcare costs PPY over the 1-year period after the index date were on average ~€12,070 for the whole IPD cohort, ~€9370 for the inpatient CAP cohort, and ~€790 for the outpatient CAP cohort. Figure 4A–C shows the estimated all-cause healthcare costs PPY for the year before and for the year after the index date, including disease-specific PPY costs after the index date and increase in the all-cause PPY costs beyond the disease-specific costs, for each patient cohort stratified by age group. A–C All-cause direct healthcare costs per person year (PPY) 1 year before and 1 year after the index date, and disease-specific healthcare costs PPY in (A) IPD, (B) inpatient CAP, and (C) outpatient CAP cohorts by age group. IPD invasive pneumococcal disease, CAP community-acquired pneumonia In the IPD cohort, the all-cause healthcare costs ranged from €5603 to €12,020 for the year before the index date (Fig. 4a). The disease-specific PPY costs totaled to €8326 among patients 18–49 years of age, €11,843 among those 50–64 years of age, and €13,066 among those ≥ 65 years of age. In these age groups, the all-cause healthcare PPY costs increased €2137, €2191, and €2002 in addition to the disease-specific costs, respectively. In the inpatient CAP cohort (Fig. 4b), the all-cause healthcare cost range was from €7399 to €15,189 for the year before the index date. The disease-specific PPY costs totaled to €6087 among patients aged 18–49 years, at €8367 among those 50–64 years, and to €10,082 among those aged ≥ 65 years. The average increase in the all-cause PPY costs above the disease-specific costs was €1036 among patients aged 18–49 years, €2256 among those aged 50–64 years, and €3506 among those aged ≥ 65 years. In the outpatient CAP cohort (Fig. 4C), the healthcare costs ranged from €2925 to €9660 the year before the index date. The disease-specific PPY costs totaled to €358, €635, and €1254 in the age groups 18–49 years, 50–64 years, and ≥ 65 years, respectively, while the average increase in all-cause PPY costs above these sums were €314, €1107, and €2573. In patients without any predefined risk factors, the increase in all-cause costs from 1 year before to 1 year after the index date, including all-cause costs above the disease-specific costs, was also noteworthy in all patient cohorts (Supplementary Table S4). Of importance, only minor increases were observed in the matched controls (Supplementary Table S5) supporting that the increase observed in the all-cause healthcare costs in patients was attributable to IPD/CAP. Cost Categories and Total Annual Costs Total disease-specific PPY costs, including direct healthcare costs and indirect costs, during 1 year after the index date were highest among those aged ≥ 65 years in the IPD (~€13,080), inpatient CAP (~€10,080), and outpatient CAP (€1260) cohorts (Fig. 5a). The main cost driver was inpatient care, which comprised > 85% of the total costs in the IPD and inpatient CAP cohorts in all age groups. Among patients aged ≥ 65 years, inpatient costs accounted for 95% of the total PPY cost, that is, ~€12,400 for the IPD and ~€9600 for the inpatient CAP cohort. Outpatient costs comprised < 5% of the costs in these patient groups. A Direct disease-specific inpatient and outpatient costs and sickness allowances per person year (PY) during the 1-year follow-up in IPD, inpatient CAP, and outpatient CAP cohorts by age group. Mortality-adjusted person time, IPD: 18–49 years, 502 PY; 50–64 years, 688 PY; and 65+ years, 1295 PY. Inpatient CAP: 18–49 years, 12,819 PY; 50–64 years, 16,756 PY; and 65+ years, 74,956 PY. Outpatient CAP: 18–49 years, 42,994 PY; 50–64 years, 26,934 PY; and 65+ years, 48,047 PY. B Total annual disease-specific healthcare cost andsickness allowances in IPD, inpatient CAP, and outpatient CAP cohorts by age group. C Total annual disease-specific costs and sickness allowances in inpatient andoutpatient CAP cohorts by age group and estimated proportion of pneumococcal pneumonia (17% out of all pneumonia cases). IPD invasive pneumococcal disease, CAP community-acquired pneumonia In the outpatient CAP cohort, outpatient costs were the main cost driver in the youngest age groups 18–49 and 50–64 years old accounting for ~66% and ~47% of the total disease-specific costs, respectively. The proportion of inpatient costs increased with age comprising ~66% of the total costs for those aged ≥ 65 years. The reimbursements paid for disease-specific sick leaves were the largest in the IPD cohort at €310 PPY, followed by inpatient and outpatient CAP cohorts at €52 PPY and €32 PPY, respectively. Sickness allowances comprised ~6% of all disease-specific costs in the IPD cohort, ~3% for inpatient CAP, and ~10% for outpatient CAP cohort among patients aged 18–49 and 50–64 years (Fig. 5a). In total, the mean annual disease-specific costs for the CAP cohorts combined summed up to ~€270 million and to ~€7.5 million for the IPD cohort. Figure 5b shows the total annual disease-specific costs for each CAP cohort by age group. For inpatient CAP, these costs were estimated at ~€189 million among patients aged ≥ 65 years. Of the yearly total disease-specific costs, pneumococcal pneumonia was estimated to account for ~€46 million among CAP cohorts of which inpatient and outpatient costs for patients aged ≥ 65 years comprised ~76% (Fig. 5c). Serotype Coverage In 2023–2024, a total of 1179 laboratory-confirmed IPD episodes were identified in the adult population (≥ 18 years), of which the serotype compositions of PCV13, PCV20, and PCV21 covered approximately 56.5%, 71.9%, and 81.3%, respectively. In the age group of ≥ 65 years, serotype coverages of the different vaccines were 43.0%, 61.1%, and 77.7%, respectively. Serotype coverages of PCV13, PCV20, and PCV21 by age group are presented in Fig. 6. The serotypes common for both PCV20 and PCV21 (Supplementary Table S1) caused 57.8% of all IPD cases (Supplementary Fig. S2). The serotypes unique to PCV20 and PCV21 caused 3.3% and 19.9% of IPD episodes in adults aged ≥ 65 years (Supplementary Fig. S2). Serotype coverage of PCV13, PCV20 and PCV21 of laboratory confirmed invasive pneumococcal disease episodes in adults (≥ 18, N = 1179) in Finland 2023–2024 by age group. Total number of episodes, 18–49 years, N = 264; 50–64 years, N = 278; and 65+ years, N = 637. PCV pneumococcal conjugate vaccine Discussion This large registry-based cohort study, including almost 3000 patients with IPD and over 250,000 patients with CAP provides comprehensive evidence on the clinical and economic burden of pneumonia in Finland. Approximately half of those with IPD or CAP were aged 65 years or older. A significantly higher morbidity and mortality among patients compared with population controls across age groups was observed. The estimated total annual costs related to CAP were almost €270 million and those related to IPD about €7.5 million, of which ~70% and 56%, respectively, were incurred by patients 65 years or older. The costs of pneumococcal disease alone, including IPD (€4.2 million) and pneumococcal CAP (€35.0 million), were estimated at €39.2 million in adults aged 65 years and older. Up to 40% of these oldest patients had no measured risk factors, underscoring that advanced age itself confers significant vulnerability to these infections. This finding aligns with international evidence that age is an important risk factor for pneumococcal disease and highlights a gap in current prevention strategies that focus primarily on individuals with chronic conditions [9]. Adverse Outcomes Overall, around 20–30% of the patients had recurrent pneumonia, and 10–25% were admitted to institutional care, with the highest rates being observed in the age group ≥ 65 years of the inpatient CAP cohort. Compared with their matched controls, the incidence of AMI was 1.3-fold in patients with IPD and 1.4–1.8-fold in patients with CAP. These results are in line with the well-established link between severe respiratory infections and cardiovascular complications [11, 12], and emphasize the long-term health consequences of IPD and CAP. Moreover, our results suggest that the increased risk of AMI may extend to patients with less severe CAP treated in the outpatient setting. This is a novel observation and therefore needs to be investigated in future studies. All-cause mortality was substantially higher for patients than for controls throughout follow-up, particularly among those aged ≥ 65 years, for whom 30-day case fatality exceeded 12% in the IPD and inpatient CAP cohorts and was ~7% in the outpatient CAP cohort. Economic Burden and Work Absenteeism This study shows that to the society, direct, and indirect costs associated with IPD and CAP were considerable (€7.5 and €269 million per year, respectively). Inpatient care was the main cost driver, comprising up to 95% of the costs among those aged ≥ 65 years. Importantly, indirect costs related to work absenteeism were also substantial, especially given the ongoing increase in retirement age in Finland. We did not estimate productivity losses and included sickness allowances only after the first 9 days. Still, sickness allowances amounted to €0.2 million for the IPD cohort and €2.3 million for the CAP cohort, highlighting the societal impact of these diseases beyond direct medical expenses. Costs of IPD and CAP seemed to increase with age, being highest in patients aged ≥ 65 years, which is in line with previous studies [3]. In a previous Finnish publication, the direct disease-specific healthcare costs were estimated at ~€4700 per episode of inpatient CAP and ~€6500 per episode of IPD (bacteremia) in the age group ≥ 65 years [24]. In this study, the estimated 1-year disease-specific healthcare costs were considerably higher, ~€13,000 PPY and ~€10,000, respectively. The reasons for larger estimated costs in our study include longer follow-up period, identification of HCRU linked with pneumonia in any diagnosis position (versus primary diagnosis), adjustment for mortality during the follow-up, and inflation (2023 versus 2010 prices). More specifically, our estimates include costs associated with IPD/CAP beyond the initial episode up to 1 year and are also adjusted for mortality during this time period. Significant increases in all-cause healthcare costs (i.e. attributable costs) that were not directly linked to treatment of IPD or CAP, assumably resulting from increased frailty and exacerbation of comorbidities, were also observed in this study. This suggests that IPD/CAP may adversely impact the patient’s quality of life beyond the acute episode. The notable increases in all-cause healthcare costs even in patients without any measured comorbid conditions suggests that there may be other risk factors that were not included as risk factors in our study but also that IPD and CAP are associated significant attributable costs beyond direct treatment costs of the infection. For example, smoking and high alcohol consumption have been linked to an increased risk of pneumonia [25]. Estimates of the proportion of S. pneumoniae of all pneumonia cases vary widely from > 10% to 50% [26–28]. According to a recent estimate, pneumococcal pneumonia accounts for approximately 17% of all CAP cases diagnosed in the adult population in Finland [21]. This estimate is also in line with a previous Finnish study, where in radiologically confirmed CAP, pneumococcus was the etiological factor in 17% of the cases [20]. Based on these estimates, S. pneumoniae accounts for ~€46 million of the yearly total disease-specific costs of noninvasive pneumonia in all ages and ~€35 million in adults aged 65 years and older in Finland. Serotype Coverage and Implications for Vaccination This study shows that a substantial proportion of the pneumococcal disease burden was attributable to serotypes included in expanded conjugate vaccines, with coverage reaching ~80% of identified adult cases for PCV21. Interestingly, in adults aged 65 years and older the coverage of PCV13 and PCV20 has decreased in recent years from 47.5% and 66.5%, respectively, in 2016–2022 [3] to 43.0% and 61.1% in 2023–2024, possibly reflecting serotype replacement as a consequence of the pediatric national immunization program (NIP) [29, 30]. In Finland, PCV10 was introduced into the pediatric NIP in 2010 [30]. In many European countries, introducing PCV13 to pediatric vaccination programs has reduced the incidence of vaccine serotypes in children but also in unvaccinated adults as a result of indirect protection [30–32]. At the same time, due to serotype replacement, the incidence of nonvaccine serotypes has increased more than 60% in older adults aged ≥ 65 years. Finland’s adult NIP is limited to certain risk groups, whereas in many European countries, vaccination is recommended and funded for all adults aged ≥ 65 years [6, 7]. Our findings suggest that broader vaccination strategies, including the use of PCVs with expanded serotype coverage, provides an opportunity to significantly reduce the burden of disease, particularly in adults over 65 years of age. Strengths and Limitations The strengths of this study include the use of comprehensive national registry data, enabling production of robust population-level estimates, and linkage of individual-level exposures and outcomes. In addition, reporting indirect costs due to work absenteeism can be considered a strength, as there are few previous publications [16]. However, several limitations should be noted. First, tobacco smoking and alcohol use—both major risk factors for IPD and CAP—could not be assessed due to the lack of such data in national registries. Overall, reliance on the register data when defining risk factors may lead to overestimation of the proportion of population with no risk factors. Moreover, our study did not include data on such potential risk factors as migration background, urban/rural residence, employment, or other socioeconomic status. Second, potential misclassification of diagnoses and incomplete capture of outpatient visits to private providers prior to 2020 may have hampered the identification of patients exposed to CAP and also affected incidence estimates of recurrent CAP. Third, studies following patients during their hospital admission for pneumonia have observed the AMI risk to be peak within the first few days of exposure [13, 15]. In our study, an AMI occurring during hospitalization for CAP was not counted as an event, as the time sequence between the infection and AMI could not be determined. Consequently, our estimates of AMI incidence are likely to be conservative. Fourth, indirect costs were most likely underestimated, as only costs associated with sick leaves lasting ≥ 10 days could be considered. Fifth, we estimated attributable costs by comparing all-cause healthcare costs before and after the index date within the patient cohorts and interpreted the increase in all-cause costs as attributable to the index IPD/CAP. As we did not compare the changes in all-cause costs between patients and controls matched for comorbidities and other factors affecting these costs, we may have attributed costs that are due to progress in patient’s underlying conditions to the index IPD/CAP. Of note, lack of adjustment for confounding factors beyond age, sex, and region may influence all our comparisons between patients and controls. Finally, we did not stratify our analyses by sex, as our main objective was to provide data to support public health measures for preventing IPD/CAP, which are not sex-specific. Nevertheless, the consistency of findings across multiple outcomes and cohorts supports the validity of our conclusions. Conclusions In this large nationwide study, IPD and CAP were associated with substantial morbidity, mortality, and economic burden, particularly among adults aged ≥ 65 years. Adverse outcomes—including recurrence, institutional care, and AMI—were frequent. Substantial proportion of older patients had no predefined medical risk factors, emphasizing age as a major determinant of vulnerability. Most costs were incurred by older adults, with pneumococcal disease alone accounting for ~€39 million annually in this age group. These findings support age‑based pneumococcal vaccination and broader use of higher‑valency conjugate vaccines to reduce disease burden and societal costs. Supplementary Information Below is the link to the electronic supplementary material. Supplementary file1 (PDF 350 KB) Acknowledgements The authors would like to sincerely thank Elina Kataja for her invaluable support on project administration and Riku Korhonen for reviewing the manuscript. Authors are grateful to Minna Vehkala and Essi Peltonen for their help with finalizing the manuscript. Author Contributions Conceptualization: Jenni Kononoff, Veli-Jukka Anttila, Peter Klemets, Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen. Funding acquisition: Jenni Kononoff. Data curation: Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen. Formal analysis: Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen. Investigation: Jenni Kononoff, Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen. Methodology: Jenni Kononoff, Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen. Project administration: Jenni Kononoff, Mari Pölkki, and Maarit J. Korhonen. Resources: Jenni Kononoff, Mari Pölkki, and Maarit J. Korhonen. Software: Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen. Supervision: Jenni Kononoff, Mari Pölkki, and Maarit J. Korhonen. Validation: Veli-Jukka Anttila, Peter Klemets, Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen. Visualization: Mari Pölkki and Maarit J. Korhonen. Writing—original draft preparation: Jenni Kononoff, Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen. Writing—review and editing: Jenni Kononoff, Veli-Jukka Anttila, Peter Klemets, Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen. All authors approved the final draft of the manuscript. All authors meet the criteria for authorship by the guidelines published by the International Committee of Medicinal Journal Editors (2004). Funding The study and publication activities, including the rapid service fee, were funded by MSD Finland Oy, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA. Data Availability The datasets generated during and/or analyzed during the current study are not publicly available due to the authors not having permission to share the data. Declarations Conflict of Interest Jenni Kononoff is an employee of Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA, and may hold stock or stock options in Merck & Co., Inc., Rahway, NJ, USA. Veli-Jukka Anttila has received lecture fees from MSD, Pfizer, Roche, GSK, BMS, Biogen, Sanofi, Novartis, and Tillotts and personal fees for consulting or advisory boards from MSD, Pfizer and Tillotts. Peter Klemets has received grants/contracts from MSD and Pfizer, support for attending meeting/travel from MSD. Marita Kalaoja, Mari Pölkki, and Maarit J. Korhonen are employees of Oriola Finland Oy. Ethical Approval This retrospective secondary registry data was permitted by Findata, the Social and Health Data Permit Authority (permit diary numbers: THL/3226/14.02.00/2022, THL/2333/14.06.00/2024, THL/3005/14.06.00.2024, and THL/2792/14.02.00/2025). Study was conducted following the Declaration of Helsinki, national legislation and regulations of Finland, legislation of European Union (EU), and the International Society of for Pharmacoepidemiology and Guideline for Good Pharmacoepidemiology Practice. Ethical evaluation was provided by the permit authority, Findata. According to the local legislation of the Finnish Act on the Secondary Use of Health and Social Data (552/2019), secondary use of registry data is considered as noninterventional study, and no further evaluations were required. In accordance with the Findata data protection policies, individual level results (N < 3) are not reported. Anonymity of the results was inspected prior to publication by the permit authority. All individual-level data and analytic datasets were handled in secured environment provided by the permit author. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Mari Pölkki and Maarit J. Korhonen contributed equally. References Associated Data Supplementary Materials Supplementary file1 (PDF 350 KB) Data Availability Statement The datasets generated during and/or analyzed during the current study are not publicly available due to the authors not having permission to share the data.