Luxembourg Health Map 2025: the hospital system in figures

The Carte sanitaire is the report the Observatoire national de la santé publishes every two years to plan hospital capacity. This reuse draws out eight key figures, all read or recomputed from the open data, without drawing a verdict.

In 1986 the country had 36 hospital establishments; by 2025 only 13 remain, consolidated through a long series of mergers. Four hospital centres hold most of the beds; six specialised establishments (cardiology, radiotherapy, neuropsychiatry, rehabilitation) and one establishment for people at the end of life (Haus OMEGA) round out the offering; two establishments, the Mondorf thermal centre and the Laboratoire national de santé, have no beds. In all, 2,686 installed beds on 30 June 2025.

Sources: Carte sanitaire 2025, synthesis document (Observatoire national de la santé), cartographie ; Tableau 8 (installed beds by establishment, 2025) ; classification des 13 établissements : tab311 (loi modifiée du 8 mars 2018) · Dark green: hospital centres; light green: specialised establishments; ochre: Haus OMEGA. Without beds: Centre thermal et de santé de Mondorf-les-Bains, Laboratoire national de santé. · Report an error

With 3.92 hospital beds per 1,000 inhabitants, Luxembourg has fewer than France (5.3), Belgium (5.4), Germany (7.6) and the EU average (5.1) in 2024 according to Eurostat; ten EU countries have fewer. According to the Observatoire national de la santé, the fall in the ratio since 2012 (5.15) is "mainly due to strong population growth, while the absolute number of beds has remained fairly stable" (our translation). With 15.2% aged 65 and over, the lowest share in the Union, hospital capacity "may be slightly lower than that of its neighbours", the report writes.

The Carte sanitaire makes two statements. In its conclusion, the capacities set by law "offer the possibility of covering the population's needs"; in its state of play, the authorisations provide from 2026 for more beds "in order to compensate for a possible shortfall in hospital beds and to meet growing demand". The report adds that judging the adequacy of beds "also requires analysing bed occupancy rates, reasons for hospitalisation and the average length of hospital stays" (our translations). This reuse does not settle the question; it gathers those indicators below, and budget-lu documents ageing on the public-finances side.

Sources: Eurostat, hlth_rs_bds1 (lits d'hôpital disponibles, toutes catégories, pour 1 000 habitants, 2024 ; 2012 en repère) ; ratio 2025 recalculé : 2 686 lits / 685 075 habitants (Tableau 10) · Eurostat, demo_pjanind (part des 65 ans et plus au 1er janvier 2025) · Tableau 1 (STATEC, population au 1er juillet, traitement ObSanté) ; independent recomputation des classes 65+ et 75+ · tab516-2 (part des journées d'hospitalisation par classe d'âge, 2014-2024p ; données IGSS, traitement ObSanté) · tab515-1 (occupation des lits aigus, centres hospitaliers, lits de soins palliatifs inclus, 2019-2024p ; données IGSS) · Tableau 40 (durée moyenne de séjour, activité opposable, 2014-2024p ; données IGSS) · tab215 (Eurostat proj_23ndbi, extraction du 17 janvier 2026 ; personnes de 65 ans et plus pour 100 personnes de 15 à 64 ans ; indicateur sans lien avec la dépendance physique) · Report an error

The hospital law sets a ceiling of "planned" beds; the ministry "authorises" a share of them; establishments "install" a share again. For 2024-2025, 88.2% of the planned ceiling (3,153) was authorised and 96.6% of authorised beds were installed (2,686 beds on 1 July 2025). The law amended on 19 December 2025 raises the ceiling to 3,545 beds; 2,813 beds have been authorised since 1 January 2026, to which the law adds a health-crisis reserve of at most 200 acute beds (50 per hospital centre). The report calculates that if all 3,545 beds were authorised and installed, for 690,959 residents in December 2025 (STATEC), the ratio "would rise to 5.1 hospital beds per 1,000 inhabitants".

Outside this count: 631 day-hospital beds installed on 1 July 2025 (743 authorised, 759 from 2026). The 174 intensive-care beds declared in July 2025 (93 adult, 54 specialised adult, 27 neonatal and paediatric) are included in the acute beds. The 2,813 beds authorised on 1 January 2026 are authorised beds, not installed beds.

Sources: Tableau 14 (lits planifiés 2024-2025, autorisés 2024-2025, installés 01.07.2025) et tab319-1 (planifiés et autorisés depuis le 01.01.2026) ; annexe 2 de la loi hospitalière modifiée du 8 mars 2018, modifiée le 19 décembre 2025 · tab319-2 (lits d'hospitalisation de jour autorisés et installés au 1er juillet, 2019-2026) · tab316 (lits de soins intensifs déclarés installés, juillet 2025) · tab318 (lits aigus autorisés 2024-2025 et 2026-2030, installés 2023 et 2025, par établissement) · synthesis document, p. 10 (scénario 5,1 pour 1 000) · Report an error

The number of hospital doctors rose from 1,047 in 2014 to 1,505 in 2025 (+43.7%), including 165 doctors in specialist training (57 in 2014). But the striking figure is elsewhere: of the 10,413 employees in the hospital sector, 49.4% live outside Luxembourg. The salaried workforce is today half cross-border, driven mainly by France (staff living in France +77% since 2014).

This reliance on the cross-border labour pool overlaps the labour market and the economic fabric already mapped in the series. The employee statistic excludes self-employed doctors, temporary agency staff and subcontracting; it describes residence, not nationality.

Sources: Tableau 28 (IGSS, Rapport général sur la sécurité sociale ; médecins agréés ou salariés auprès des établissements hospitaliers au 1er janvier, chaque médecin compté une fois ; MEVS = médecins en voie de spécialisation) · Tableau 29 (Centre commun de la sécurité sociale, calcul IGSS ; salariés du secteur hospitalier au 31 mars, par pays de résidence, hors médecins libéraux, intérim et sous-traitance) ; part frontalière = France + Belgique + Allemagne, independent recomputation · Report an error

In 2024, the four emergency departments and the national paediatric emergency department recorded 268,683 unplanned visits, i.e. 734 a day. Adults account for 71.4% of visits (including 27,165 patients aged 75 and over, 10.1%) and children for 28.6%. According to the report, the share of non-residents rose from 10.0% in 2019 to 13.2% in 2024. The mean time from admission to first medical contact is 72.8 minutes; the synthesis notes that the mean times at levels 3 and 4 of the Canadian triage scale exceed the target times. A visit lasts 179.9 minutes on average (adults 192.3, children 147.2).

Three different bases coexist in the tables: 268,683 visits in total, 267,237 in the mean-duration calculation and 201,976 in the time-to-first-contact calculation. The latter is a weighted mean of CHEM, CHL (adults), HRS and the national paediatric emergency department: CHdN did not record this indicator in 2024. The rates by age are overestimated, the table notes, because visits by non-residents are divided by the resident population only. In inpatient care, the share of stays by non-residents rose from 6.2% in 2014 to 11.0% in 2024 (provisional data).

Sources: Direction de la Santé, données 2024 transmises par les quatre services d'urgence (CHdN, CHEM, CHL, HRS) et le service national d'urgence pédiatrique (CHL) ; tab1 (passages par catégorie), tab3 (recours par âge, population STATEC au 1er juillet 2024), tab4 (délai moyen admission → premier contact médical, cibles CTAS 2016), tab5 (durée moyenne aux urgences) · Moyenne pondérée CHEM, CHL adultes, HRS et service national d'urgence pédiatrique ; le CHdN n'a pas recensé cet indicateur en 2024 · fig518-1 (séjours hospitaliers des résidents et non-résidents, 2014-2024 ; données IGSS, traitement ObSanté) · part des non-résidents : synthesis document, p. 29, et remarque du tab3 · Report an error

Fascicule 3 counts 14 MRI units, 14 CT scanners and 9 mammographs installed at the end of 2025 (12, 15 and 7 at the end of 2023). Relative to population, the fleet is close to the OECD average (45 CT, MRI and PET units per million inhabitants vs 51, 2023). Use stands out more: 283 CT exams per 1,000 inhabitants in 2023, vs 231 in Belgium and 228 in France, the highest level of the five countries the report compares; and 121,576 MRI exams performed in 2025 (+27.2% since 2023).

Two counts coexist: Fascicule 3 counts equipment installed at year end; the table of equipment by establishment (Direction de la santé, radiation protection) counts 13 CT scanners, 13 MRI units, 8 gamma cameras, 2 PET scanners and 8 mammographs on 1 July 2025. Luxembourg's CT-exam figure is provisional and Germany's is an estimate (note in the report).

Sources: Fascicule 3, Tableaux 1/6/8 (équipements installés, niveau national, fin d'année 2023-2025) · tab332 (Direction de la santé, division de la radioprotection ; équipements au 1er juillet 2025) · Figure 85 (OCDE, Panorama de la santé 2025 ; scanners TDM, IRM et TEP par million d'habitants, 2023 ; moyenne OCDE non pondérée) · Fascicule 3, Figures 21, 30 et 37 (OCDE ; examens IRM et CT pour 1 000 habitants, 2023 ; LU provisoire, DE estimé) et Tableau 7 (examens IRM réalisés) · Report an error

For a planned MRI exam performed as an outpatient in a hospital centre, the national median time between booking and the exam was 31 days in 2025 (33 in 2023, 27 in 2024), on 101,088 exams; 78.1% of exams were performed within 60 days and 45.1% within 30 days. For mammograms of the Programme Mammographie, the organised breast-cancer screening programme, the median fell from 41 days in 2023 to 21 days in 2025, on 25,946 mammograms. Depending on the centre, the 2025 median ranges from 29 to 43 days for MRI and from 12 to 60 days for screening.

Scope: planned exams performed as outpatients in hospital centres; wait measured between the booking date and the exam date; medians computed by ObSanté (Fascicule 3). The MRI medians (Figure 20) and the distribution by band (Figure 19) rest on different exam bases, hence distinct totals. At the end of 2025, the national monthly MRI median was 30 days in October, 34 in November and 36 in December.

Sources: Fascicule 3, Figure 20 (délais médians IRM, par centre hospitalier et par mois ; données FHL, traitement ObSanté) et Figure 19 (distribution par tranche de délais, base d'examens différente) · Fascicule 3, Figure 4 (délais médians des mammographies du Programme Mammographie), Figure 2 (distribution) et Tableau 3 (mammographies PM réalisées) · Report an error