European data sources for computing burden of (potential) vaccine-preventable diseases in ageing adults

The Netherlands, Denmark and Norway use the proposed case definition in their surveillance (Table 1). In Italy, only a subset of laboratories within the European AMR surveillance network (EARS-net) use this case definition. Hungary uses a non-agent specific definition for Hospital-Associated Bloodstream Infections (HA-BSI) [19], Belgium and Spain do not collect ExPEC-specific data. None of the countries reported summary BoD measures (Table 2). Stratification by age-group and risk factors was reported as possible in the six countries for which morbidity data are available (Hungary, Denmark, Norway, Italy, and The Netherlands). ExPEC-mortality and sequelae data were rarely available. The latter could only be retrieved in The Netherlands. All countries collected AMR data.

A total of 940 publications were identified in the literature review, of which 41 were included for data extraction covering eleven countries (Additional files 3 and 4). Most publications came from the United Kingdom (UK) (n = 9), Denmark (n = 6) and Spain (n = 4) (Table 3). Additionally, seven publications included multiple countries for data extraction. Respectively, twelve and three publications focused on E.coli bacteraemia and urosepsis. Sixteen studies provided general Bloodstream Infection (BSI) morbidity data. Only eight studies presented E.coli bacteraemia incidence. Seven publications concerned BSI caused by AMR E.coli strains. Several countries presented data on E.coli BSI mortality (Denmark, Finland, Norway, Sweden and the UK) or AMR E.coli-related BSI mortality (Germany and Spain). A large European study reporting ICU-specific mortality data was also identified. Many publications focusing on AMR E.coli BSI were found. Two large AMR-focused surveillance systems were identified (SENTRY and EARS-Net) [20, 21]. Only one study reported DALYs for AMR E.coli, and no study reported disability/utility weights, disease duration, or the need for hospitalisation due to ExPEC (hospitalisation needs).

Belgium, Italy and Spain use both the norovirus and syndromic (acute gastroenteritis) case definitions in their surveillance (Table 1). Denmark uses only the norovirus case definition. The Netherlands collects aggregated laboratory surveillance data on norovirus. Hungary and Norway collect data on norovirus outbreaks. The survey results indicated that the majority of countries collected norovirus morbidity data (6 out of 7), mortality data (4 out of 7), and hospitalisation data (4 out of 7) (Table 2). In addition, Belgium, Denmark and the Netherlands estimated DALYs specifically for norovirus.. However, only a few countries assessed the level of under-reporting or under-ascertainment (i.e. completeness) of morbidity data, and morbidity and mortality data were rarely stratified by age or risk factors. No country collected sequelae data.

The literature review yielded 371 publications, of which 24 were selected for data extraction (Additional files 3 and 4). Data from eight countries were retrieved, mostly from the Netherlands (n = 6), the UK (n = 6) and Germany (n = 5) (Table 2). Four studies (from the UK and the Netherlands) estimated DALYs based on incidence data obtained via the population-based studies Sensor [22] and IID2 (Infectious Intestinal Disease in the community study) [23, 24]. Morbidity data were retrieved in 24 publications, half of them describing pathogen-specific data, in most cases stratified by age. Sequelae data were identified in three syndromic data studies. Norovirus and syndromic disease mortality data were identified in eight and three studies, respectively.

Spain and Italy use the non-bacteraemia PnPn case definition in their surveillance, whereas the other countries use only the all-cause pneumonia case definition (Table 1). Overall, PnPn-specific BoD data is lacking, as no country collected summary BoD data, and only Spain and Italy reported PnPn-morbidity and mortality data (Table 2). On the other hand, more data was available for all-cause pneumonia; Belgium and the Netherlands reported that DALY and/or QALYs have already been calculated or could be calculated for all-cause pneumonia. All-cause pneumonia incidence and hospitalisation data are available in most countries, and mortality data are available in the Netherlands, Belgium and Hungary.

A total of 784 publications were identified, of which 40 were included for data extraction (Additional files 3 and 4). Most studies were conducted in Spain (n = 9), the Netherlands (n = 8), the UK (n = 7) and Italy (n = 5) (Table 2). Only two studies reported QALYs; both were based on the CAPITA study from the Netherlands on all-cause pneumonia [25, 26]. Two studies from France and Switzerland determined the QoL for all-cause pneumonia [27, 28]. For most studies, the incidence was based on hospital discharge data, either for all-cause pneumonia or PnPn. Incidence data on outpatients with all-cause pneumonia were only reported for France, the Netherlands and Belgium. Serotype-specific PnPn incidence data were available in three studies. Several studies reported the attributable fraction for pneumococci using hospital discharge data. Most studies reported age stratification data and half of them reported data for risk groups. Several studies reported mortality after PnPn or after all-cause pneumonia.

Case definitions and methods to ascertain RSV cases differ widely by country (Table 1). In Belgium and Spain, laboratory testing is usually conducted only in a selection of hospitalized cases. In the Netherlands, while a subset of ARI cases undergoes laboratory testing for RSV and real-time SARI surveillance is being piloted, no official laboratory confirmation is needed for applying ICD-10 codes in hospitalized cases. Many hospitals in Norway conduct influenza like illness (ILI) surveillance with RSV and influenza panel testing, which results in RSV-positivity among ILI cases. Under-ascertainment and underreporting may bias the official frequency estimates as the completeness of reported frequency data are not assessed in any of the countries (Table 2). The only exception might be the Netherlands, where efforts have been made to assess data completeness for RSV cases identified in primary care. RSV sequelae data are available from Denmark, and are also currently being studied in the Netherlands, in the context of the RESCEU (Respiratory Syncytial virus Consortium in Europe) project [27]. Spain has reported that hospital records could be used as a possible source of sequelae data. As several countries use SARI case definitions for RSV (Belgium, Spain, the Netherlands, Denmark), there are available data on hospitalisations and RSV-specific mortality. In Belgium, sentinel hospitals report case-fatality and RSV-SARI mortality data, which can be stratified by age and risk groups. There are also case-fatality data in Spain, but only for younger age-groups and for a subset of cases which undergo laboratory testing.

The literature search yielded a total of 711 publications, of which 46 were included in the data extraction (Additional files 3 and 4). The majority of studies originated from the UK (n = 12), followed by Sweden (n = 4), France (n = 3), Poland (n = 3), Spain (n = 3), Germany (n = 2), Greece (n = 2), the Netherlands (n = 2), Switzerland (n = 2), and Belgium (n = 1) (Table 2). In the literature review, the most widely used case definitions for sampling are based on ILI or ARI cases, or both. Although one article reported age-specific DALYs due to hospitalised lower or upper respiratory tract infections (LRTIs and URTIs) caused by RSV, no study described country-specific summary BoD estimates. In a study by Shi et al. [28], simple BoD data (RSV incidence, hospital admission rates, RSV positivity among hospitalised ARI cases and hospital case fatality ratios for RSV) were estimated for industrialised vs developing countries, but no European-level data were described. Age stratification and risk factor data (co-morbidities, such as cardiopulmonary and cardiovascular diseases), immunocompromised health status, and seasonality were widely reported in publications. Publications analysing RSV surveillance data in the Netherlands and the UK contained robust data on RSV-attributable mortality. Case fatality rates were estimated in literature reviews performed in French and Swedish hospitals [28, 29].

Belgium, Denmark, Italy, the Netherlands, Norway and Spain reported using the abovementioned S.aureus case definition in their surveillance (Table 1). Hungary uses the official non-specific ECDC case definition of HAI, including the one for BSI [19]. Only Belgium reported that DALYs could be calculated, although only for general nosocomial infections but not for those which are pathogen-specific (Table 2). None of the surveyed countries collect disability/utility weights. All countries, except for Norway, have S.aureus incidence data available or data which is easily retrievable and can be stratified by risk factors. Several countries stated that they collect sequelae and hospitalisation data. Very few countries collect mortality data, while all countries collect AMR data. Additionally, some countries collect non-BSI data, such as S.aureus pneumonia and surgical site infection cases.

The literature search yielded 185 publications, of which 101 publications remained for data extraction after full-text review (Additional files 3 and 4). Most publications came from Denmark (n = 14), followed by Spain (n = 13), Germany (n = 9), England (n = 8), France (n = 7), and Finland (n = 6). Two large European surveillance systems were identified: the EARS-Net [20], and the Central Asian and Eastern European Surveillance of Antimicrobial Resistance (CAESAR) [29]. Both collect data on blood cultures and AMR, which can be used to calculate the incidence of S.aureus bacteraemia. EARS-Net also allows age and gender stratification (Table 2). None of the publications reported summary BoD measures. Incidence data could be extracted from 51 publications. In most cases, data were stratified by age. Sequelae and hospitalisation data were often retrievable. Disease duration was only reported in three studies. In total, 75 studies reported on S.aureus-specific or all-cause mortality. Many publications included risk factor analyses related to mortality data. Almost all publications reported AMR in S. aureus infections.