This annex provides and overview Luxembourg’s health data infrastructure for the purposes to set up Health System Performance Assessment in Luxembourg. Drawing on the comprehensive Situational analysis of Luxembourg’s health data infrastructure for HSPA purposes (OECD, 2025[1]) it describes the main data custodians and data sources highlighted in Chapter 4, with a particular emphasis on health data infrastructure relevant to the final set of 105 HSPA indicators that populate Luxembourg’s HSPA framework. More information and details on the health data landscape in Luxembourg is available in the Situational Analysis of Luxembourg’s health data infrastructure for HSPA purposes (OECD, 2025[1]), drafted building on 12 interviews with Luxembourg health system custodians in autumn 2024, and desk research.
Health System Performance Assessment Framework for Luxembourg
Health data custodians and governance
In Luxembourg, the health data ecosystem is governed by several health data custodians. Figure A B.1. highlights key stakeholders managing health datasets.
Note: Light blue: core public organisations in health system governance; dark teal: organisations supporting health system governance; grey: highlights health research and partners; light green: healthcare providers; dark blue: other stakeholders.
Source: (OECD, 2025[1]), Situational analysis of Luxembourg’s health data infrastructure for HSPA purposes.
The Ministry of Health and Social Security (M3S) provides strategic oversight on health system governance and health policy development, but plays a smaller role in management of datasets, interacting with several public stakeholders. Nevertheless, the M3S has some datasets in its custody, stemming from its regulatory role, including professional licensing registries and hospital authorisation records, as well as pricing data (e.g. pharmaceuticals and medical devices). Organisations responsible for health data operate with a high degree of independence from the M3S, including those under the authority of the M3S, such as the Health Directorate (DiSa) and the General Inspectorate of Social Security (IGSS).
The Health Directorate (DiSa) is in charge of public health, including its monitoring, development of public health action plans, prevention programmes and health promotion, infectious disease control, and pharmaceutical surveillance while being also the national authority for medicines and health products. It has four analytical departments, which serve as primary custodians for clinical and public health data. DiSa departments collect a wide range of data, including Administrative and Clinical Information on Hospital Stays (DCSH), the Causes of Death registry, hospital infrastructure data, notifiable infectious diseases, environmental risks, and cancer screenings. DiSa also commissions population health surveys from public research organisations, including:
Health Behaviour in School-age Children (HBSC) conducted by the University of Luxembourg which provides important insights on children aged 11‑15.
The European Health Interview Survey (EHIS) conducted by the Luxembourg Institute of Health (LIH), which constitutes an important source of information on health risk factors in the adult population.
The Survey of Health, Ageing and Retirement in Europe, conducted by the Luxembourg Institute of Socio-Economic Research (LISER), capturing health of older people.
The Patient Reported Indicators Survey (PaRIS) survey conducted by LIH, which provides insight to experiences of primary care users with chronic conditions.
Finally, DiSa commissions the Luxembourg Institute of Health to serve as custodian to two clinical registries, including the perinatal health registry (SUSANA), and the trauma and accident registry (RETRACE). The National Cancer Registry, governed by a multi‑organisation management committee, is also managed by LIH.
The National Health Fund (CNS), the national health insurer, collects health data notably through reimbursement claims, capturing among others outpatient care, inpatient physician services, pharmaceuticals purchased outside of hospitals, and laboratory tests. Population coverage varies across datasets (Box B.1). It also maintains a list of health professionals registered for billing purposes of the statutory health insurance, and this builds on the data of the M3S on health workers licensed to practise. Given its role in financing health provision, the CNS produces key data on healthcare spending in Luxembourg.
Luxembourg’s demographic profile is notably influenced by its high number of foreign workers and cross-border commuters for work. Close to half of the residents are not Luxembourgish nationals, and close to half of Luxembourg’s workforce live in neighbouring countries. Healthcare use of the population insured in Luxembourg, including Luxembourg residents and cross-border workers, is captured in claim data. However, information on healthcare consumed abroad, particularly by cross-border workers, is limited to basic claims summaries from foreign insurers. Additionally, in 2021 around 8% of Luxembourg’s residents were not covered by the statutory health insurance, including employees of International organisations and their families (8% of residents), and are excluded from the claims data. Clinical registries capture cases in Luxembourg, regardless of residence, while population surveys focus on residents.
Source: (OECD, 2025[1]), Situational analysis of Luxembourg’s health data infrastructure for HSPA purposes.
The General Inspectorate of Social Security (IGSS) centralises data from across Luxembourg’s social security system, including health insurance claims (including those for long-term care), social security contributions, the employment market, and pensions. The IGSS also accesses the Administrative and Clinical Information on Hospital Stays (DCSH) and health workforce identification numbers via the National Health Fund (CNS). Given the IGSS’ role in governance of social security, the datasets can serve as a source of person-level data on selected socio-demographic characteristics, household composition, employment, work absences, social assistance benefits, and sickness leave. The IGSS also produces and reports financial revenues and expenditures indicators to the OECD and the Eurostat for the System of Health Accounts, including timely estimates from the previous calendar year. The IGSS stores its datasets in a Data Warehouse, which can be used internally for data linkage. Variables across datasets can be accessed by external uses via the Microdata Platform for project-based data linkage, which ensures relevant data protection and data quality assurance measures.
The National Health Observatory (ObSanté) consolidates and uses data from other organisations to produce analytical reports and evidence regarding Luxembourg’s health system. It does not collect data on its own but accesses it through data requests to other organisations depending on needs (IGSS, DiSa, LISER, LIH, University of Luxembourg, STATEC). The Luxembourgish legislation ensures the ObSanté with a right to request, centralise and analyse personal data from public administrations, public institutions, and other Luxembourg’s organisations in pseudonymised form.
The Agence eSanté manages the national electronic health record system (Dossier de Soins Partagé – DSP) and the non-mandatory e‑vaccination registry. The data uploaded to DSP include laboratory and imaging results, hospital discharge report, physician-added information, and patient-added information. Luxembourg has not yet implemented an electronic prescription system. The role of Agence eSanté in the administration of digital solutions makes it well-positioned to collect and disseminate digital health indicators.
The Hospital Federation of Luxembourg (FHL) represents Luxembourg’s entire hospital sector and plays a co‑ordinating role in monitoring of hospital care quality, covering inpatient and day care hospital activities in Luxembourg. The FHL does not work on primary data, but receives information from hospitals and the CNS. The FHL does not publish reports online, although it shares dashboards on quality indicators directly to hospitals.
STATEC, the national statistical office, conducts population surveys including EU-SILC and EHIS, manages census data, and produces demographic statistics from the population registry. The Luxembourg Institute of Socio-Economic Research (LISER) runs the European Survey on Health, Ageing and Retirement (SHARE) in Luxembourg. The Luxembourg National Data Service (LNDS) provides technical expertise to facilitate the sharing and secondary use of data from the public sector and is foreseen to expand its role in health data storage.
The Ministry of Family Affairs, Solidarity, Living Together and Reception of Refugees (Ministry of Family Affairs) oversees long-term care (LTC) facilities and services for older people and people with disabilities, managing relevant data. A regular quality monitoring programme in long-term care is being established.
Luxembourg’s National Cancer Institute (INC), established in alignment with the National Cancer Plan in 2015 as a non-profit collaboration of health system stakeholders, serves as a central co‑ordinator for public and private stakeholders involved in cancer care and management in Luxembourg. The INC publishes annual reports noting aspects of cancer care processes, including indicators, such as number of Multidisciplinary Tumour Board meetings held to review and discuss individual cancer cases.
In addition, data on specific topics are collected by the Medical Chamber (Collège Médical), which represents healthcare professionals, the Chambre des salariés Luxembourg (CSL), which represents private sector workers, and the Ministry of Gender Equality and Diversity (MEGA).
Data flows and linkage
Numerous data transfers and flows between key organisations demonstrate a foundation for collaboration and information exchange. Figure A B.2. illustrates key data flows between custodians and data users. However, data sharing largely operates through bilateral agreements and ad hoc requests, often requiring case‑by-case approvals. As a secondary data user, ObSanté receives aggregated indicators from custodians such as IGSS, DiSa and STATEC, while also engaging upstream in indicator development where needed. For instance, a joint project involving ObSanté, the Luxembourg Institute of Health (LIH) and the Hospital Federation led to the establishment of standardised reporting on medical imaging waiting times.
Note: The figure does not illustrate all health data available in Luxembourg, and represents only a selection (e.g. the Surveillance system of mortality is described in the baseline report). CNS – National Health Insurance; IGSS – General Inspectorate of Social Security; DiSa – Health Directorate; M3S – Ministry of Health and Social Security; FHL – Hospital Federation of Luxembourg; LIH – Luxembourg Institute of Health; ObSanté – National Health Observatory; DSP – Electronic Health Record system.
Source: (OECD, 2025[1]), Situational analysis of Luxembourg’s health data infrastructure for HSPA purposes.
Luxembourg uses unique patient identifiers (matricula) to enable data linkage across datasets. Linkages have previously been conducted for selected use cases, including calculation of indicators such as hospital readmissions using the Administrative and Clinical Information on Hospital Stays (DCSH) dataset, 30‑day post-hospitalisation mortality through linkage between DCSH and the mortality registry, and cancer survival analyses linking the cancer registry with the mortality registry data.
The Microdata Platform hosted by the General Inspectorate of Social Security, established in 2018, is a key enabling infrastructure for secondary data use. It provides secure access to linked administrative microdata for research and analytical purposes through controlled virtual environments, where researchers can work with requested variables from pseudonymised datasets covering claims, hospital stays, employment, social benefits and demographic information. This infrastructure enabled a project on development of methodology for monitoring health workforce in Luxembourg, which quantified physician and nurse activity by linking licences to practise, claims, and employment records.
Data gaps and limitations
Luxembourg’s health data infrastructure, while comprehensive for hospital care, mortality surveillance, and cancer monitoring, presents several notable gaps. Firstly, diagnoses and clinical detail are not available for primary care and outpatient specialist visit, as the only source for these is claims data containing billing codes without information on conditions treated or services provided. This limitation prevents the identification of patient cohorts with specific chronic conditions and constrains the ability to track disease management in ambulatory settings or monitor care pathways across primary, specialist, and hospital care. As a result, continuity and co‑ordination of care across settings cannot be comprehensively assessed using administrative data alone. Cohorts with major chronic conditions such as cardiovascular diseases and diabetes cannot be identified through disease registries either, as registries have not been established. The e‑vaccination card system, despite increasing uptake concentrated in recently vaccinated cohorts, is not integrated in the electronic health record and its voluntary nature means it cannot provide reliable population-level data on childhood vaccination coverage.
Additionally, possibilities for data use for health system monitoring are constrained by other factors. Health data governance in Luxembourg is characterised by multiple custodians with distinct mandates, resulting in fragmented data ownership and reliance on bilateral agreements for data sharing. Although a unique personal identifier (matricula) enables data linkage, most linkages are currently performed on a project-by-project basis rather than embedded in routine monitoring, and some datasets cannot be linked due to missing or restrictive legal provisions governing secondary use. Timeliness of updates available to stakeholders varies across datasets, in part due to a lack of automation in data transfers. Finally, population coverage presents particular challenges given Luxembourg’s unique demographic composition: while residents insured by the CNS are captured across nearly all data sources, information on healthcare consumed abroad (e.g. by CNS-insured cross-border workers) is limited to basic claims summaries from foreign insurers, and international organisation employees and their families are not captured in CNS administrative data.
Surveys complement administrative data sources – in Luxembourg, surveys are at present the primary source for measuring the prevalence of chronic conditions such as cardiovascular diseases, diabetes, and mental health disorders, as well as the quality of care in primary care settings from a patient perspective. However, the periodic nature and smaller sample sizes, as well as reliance on maintaining high response rates, limit the information use for continuous monitoring and local-level analysis.
This section highlights selected sources of data for Luxembourg’s HSPA framework, drawing from the national health data infrastructure and international sources. Nearly three‑quarters (74%) of indicators in Luxembourg’s HSPA (Section 3.2) are linked to international sources, such as surveys, international organisation data collections, publications and databases, while the rest are sourced from country-specific sources and methodologies.
Many of the selected indicators draw on well‑established international data collections that provide comparable, high‑quality information on health system performance. These sources support cross‑country analysis by offering consistent definitions, validated methodologies, and regular updates.
Health data Questionnaire collects core health datasets, such as insurance coverage, long-term care, pharmaceuticals consumption, risk factors, screening rates and remuneration of health workers. It is managed by the OECD, and conducted annually.
JHAQ (Joint Health Accounts Questionnaire) collects internationally comparable data on countries’ health expenditures, following the System of Health Accounts framework (standards for classifying and monitoring health spending) and jointly administered annually by OECD, Eurostat, and WHO. It enables detailed tracking of spending by function, provider, and financing scheme.
JQNMHS (Joint Questionnaire on Non‑Monetary Health Statistics) gathers non‑monetary health statistics such as workforce, infrastructure, hospital capacity, and service utilisation and hospitalisation diagnoses, jointly administered annually by OECD, Eurostat, and WHO to facilitate cross‑country monitoring of system resources.
HCQO (Healthcare Quality and Outcomes) is the OECD’s collection of comparable indicators on healthcare quality, conducted once every two years. It covers dimensions such as healthcare effectiveness, safety, and patient‑centredness to benchmark performance across health systems.
In addition to the regular data collections listed above, the OECD undertakes work to develop and expand indicators to facilitate adoption of validated common calculation methodologies for indicators that capture specific topic areas, such as mental health, climate, emergency care or cancer. Luxembourg’s HSPA includes indicators for instance from the OECD Cancer Pilot data collection, and the OECD Long-term care pilot data collection.
Internationally conducted health surveys complement administrative data based indicators and provide information on health behaviours and risk factors (Eurostat administers EHIS every five to six years and EU-SILC every two years). Additionally, surveys offer insights into health status and care needs among specific populations, such as young people (HSBC is a WHO collaborative cross-national study done every four years) and older people (SHARE is co‑ordinated by a dedicated international research infrastructure established under EU law run every two years). Survey microdata can often be disaggregated by income, education, sex, age, and degree of urbanisation, which enables equity-related analyses across socio-economic and demographic groups.
Health status and mortality data
Luxembourg maintains a comprehensive Causes of Death registry with coverage for all deaths occurring in Luxembourg since 1963, including ICD‑10 coded diagnoses. The mortality surveillance system also provides near real-time reporting to EuroMOMO, aiming to detect and measure excess deaths that may be related to seasonal infections, extreme weather and other public health threats.
The cancer registry, operational with complete coverage since 2013, captures incidence, staging, treatment pathways, and survival follow-up for all cancer cases diagnosed or treated in Luxembourg. Other registries include SUSANA, capturing perinatal maternal and neonatal outcomes, and RETRACE, capturing anonymous incidence data from emergency departments on injuries and accidents, including suicide attempts. However, disease registries do not exist for other conditions such as cardiovascular diseases or diabetes. These mortality and disease‑specific registries support immediate calculation of many indicators, notably in the Health Status domain, such as life expectancy, causes of death, and cancer incidence.
Additionally, population surveys including EHIS, EU-SILC, and SHARE collect self-reported health status, chronic conditions, and activity limitations. Prevalence of chronic diseases and mental health status is also captured through surveys.
Hospital care quality and safety data
Information on quality of hospital care can be calculated using the DCSH dataset, which provides comprehensive administrative and clinical information for all inpatient stays, including day cases. Administrative data encompasses admission and discharge dates, residence, and insurance status, while clinical data includes diagnoses, procedures, and Diagnosis-Related Group (DRG) classification with severity levels. As the dataset can be linked, calculation of hospital-based indicators, including the Value‑based Care domain, with readmission rates, length of stay, and post-hospitalisation outcomes through linkage with the Causes of Death registry. Patient Safety indicators can similarly be calculated from the DSCH dataset. In addition, Luxembourg has available information through mandatory reporting of nosocomial infections and a pharmacovigilance register maintained by DiSa.
Emergency care is not systematically integrated into DCSH unless patients are subsequently admitted to hospital, although an emergency department monitoring system is under development. Data on outpatient hospital services, including specialist consultations and ambulatory procedures not requiring admission, remains limited (see the ambulatory and primary care data section) below.
The Hospital Federation (FHL) co‑ordinates quality monitoring with indicators calculated by hospitals to produce benchmarking dashboards for hospitals covering selected quality indicators. Although not yet conducted in Luxembourg, condition-specific collection of patient-reported care outcomes could inform on care for specific conditions.
Ambulatory and primary care data
Health insurance claims data for SHI-reimbursed services is a key source of data available on outpatient and primary care data, including consultations with general practitioners and specialists, prescriptions dispensed through pharmacies, diagnostic tests and medical imaging. However, diagnoses and detailed activities in primary care and outpatient specialist claims are absent, which differs from the detailed clinical coding available for hospital data. The lack of clinical detail in ambulatory claims also limits care pathway tracking between primary, specialist, and hospital care, as transitions between settings cannot be linked to specific conditions or treatment plans.
Population surveys play a key role with information on access barriers and unmet healthcare needs through EU-SILC and EHIS, and patient satisfaction and experience among people with chronic diseases treated in primary care through the PaRIS survey. From these sources, indicators in the Integrated Care, Access to care, and Equity domains can be immediately calculated including indicators on unmet healthcare needs, barriers to access mental healthcare, and PREMs.
Healthcare infrastructure and workforce
The professional licensing registry maintained by the Ministry of Health and Social Security (M3S) contains all health professionals authorised to practice in Luxembourg, updated when authorisation status changes. The National Health Insurance Fund (CNS) maintains a billing registration database for healthcare professionals with reimbursement authorisation, including information on physicians’ and other health professionals’ specialisations. A new methodology developed in 2024 links licensing data, claims data, and employment records via the IGSS Microdata Platform to measure the number of practicing professionals and their activity levels. However, some health professionals remain incompletely captured in available data. Salaried hospital staff, including nurses and allied health professionals, do not appear individually in claims data as their services are reimbursed through hospital budgets rather than individual billing. Additionally, up to 20% of medical claims identify physicians’ associations rather than individuals, complicating analysis of individual activity patterns.
Healthcare infrastructure data is available to the Ministry of Health and Social Security (M3S) every five years for the authorisation process and shared with DiSa for its role in hospital planning, and includes beds categorised by type and department, equipment, and departmental organisation. In addition, ObSanté collects hospital infrastructure data directly from hospitals for the Carte sanitaire publication every two years.
Such data support calculation of many Workforce, Infrastructure and Sustainability domains indicators, including for example indicators on practicing doctors and nurses and hospital bed capacity.
Financial status and spending
Data on health expenditures is available through CNS claims data, and reported by service type, provider category and function to international reporting by IGSS. On the national level, spending can be available by geographic area to capture spending on services provided abroad. IGSS similarly has an overview of the operating balances of the health and maternity insurance, as well as reserves available in long-term care and health insurance.
Prevention, screening, and determinants of health
Behavioural health determinants are captured via population surveys, such as EHIS, EU-SILC and HBSC provide regular information on smoking, alcohol consumption, diet, physical activity, and body mass index (BMI) among adults and adolescents. Among non-behavioural determinants, environmental health monitoring captures aspects such as air quality, water quality, and radiation exposure.
Regarding primary prevention, childhood vaccination data is sourced from vaccination coverage surveys conducted among parents of young children every five to six years provide the only population-level estimates of childhood immunisation rates. A voluntary e‑vaccination system exists, however cannot be yet reliably used to monitor coverage, although COVID‑19 vaccination is completely captured through mandatory e‑card recording. Vaccination against influenza among older people can be monitored through claims data, as the service is reimbursed.
Cancer screening data is available for organised screening programmes for breast and colorectal cancer through data directly from the structured programmes. Cervical cancer screening is conducted opportunistically, with some capture possible through laboratory claims though coverage estimation is less complete than for organised programmes. These data enable calculation of many Integrated Care (Prevention) and Determinants of health domain indicators, including indicators on breast and colorectal cancer screening coverage, smoking rate, and alcohol consumption prevalence, physical activity levels, obesity rates, and environmental health exposures.
Data to capture health inequalities
In Luxembourg’s HSPA framework, the domain of Equity includes indicators from surveys, that allow for relevant breakdowns, as well as indicators specific to some population groups. Throughout the framework, many indicators are disaggregated by population group, including breakdowns that provide further information on inequalities outside of the Equity domain.
The IGSS maintains data on employment status, occupation, social benefits, and household composition for all individuals contributing to social security. STATEC collects information on access to care, healthcare, education and income through surveys such as EU-SILC and EHIS, as well as housing conditions and other socio-economic factors. While the decennial census provides detailed information on educational attainment, current legislation does not permit linking census education data with the population registry and other health data, limiting the ability to analyse health outcomes by education level.
Additionally, national data on specific population group is available to the Ministry of Gender Equality and Diversity through collection from organisations dealing with domestic violence, or to the IGSS from linked administrative data to identify people who receive social benefits, such as Allocation de vie chère.
References
[1] OECD (2025), Situational analysis of Luxembourg’s health data infrastructure for HSPA purposes, OECD, Paris, https://www.oecd.org/content/dam/oecd/en/about/projects/the-oecd-technical-support-to-national-hspa-development/Luxembourg-HSPA-Baseline-report.pdf.