Characteristics of individuals registered in the national registries in social medicine departments in Denmark 2012–2022: a descriptive cohort study Authors: Mia Sarah Fischer Button, Line Lund Kårhus, Poul Frost, Jan Renneberg, Janne Petersen Published: 2026 ABSTRACT AbstractIntroductionUnderstanding the health and social circumstances of individuals experiencing long-term unemployment is essential for effective welfare and rehabilitation efforts. Social medicine departments are key assessment points in Denmark, yet the characteristics of those they assess remain poorly documented. This register-based study aims to describe the demographic, socioeconomic and health characteristics of individuals registered in national registries in social medicine departments in Denmark.MethodsThe study includes all first-time contacts to social medicine departments from 1 January 2012 to 31 December 2022 over 18 years. Individuals were identified through the Danish National Patient Register and the SocMed Capital Region Database. Two matched groups included: Match Group 1, matched on sex, age and residence at a ratio of 1:5, while Match Group 2 was additionally matched on employment status at a ratio of 1:1.ResultsA total of 13 092 individuals were identified in the study population SocMedDep group. The group had a lower educational attainment; notably, 15.4% of the SocMedDep group were employed, compared with nearly 78% in Match Group 1. The SocMedDep group exhibited a higher presence of psychiatric disorders and musculoskeletal disorders than both matched groups. Trends showed a rise in psychiatric diagnoses over time, particularly in the SocMedDep group.ConclusionsThis study highlights the health disparities among individuals referred to social medicine departments in Denmark, showcasing a higher prevalence of mental health issues and prolonged unemployment. The findings indicate a need for targeted intervention reforms in the labour market policies to enhance employment outcomes for individuals with psychiatric and musculoskeletal disorders. WHAT IS ALREADY KNOWN ON THIS TOPIC There is a subgroup of unemployed individuals who remain hard to rehabilitate despite labour market reforms. Social medicine departments in Denmark are key points in assessing these individuals. There is a subgroup of unemployed individuals who remain hard to rehabilitate despite labour market reforms. Social medicine departments in Denmark are key points in assessing these individuals. WHAT THIS STUDY ADDS This study characterises the health and social circumstances of individuals seen at social medicine departments and gives insights on individuals who are experiencing long-term unemployment. This study characterises the health and social circumstances of individuals seen at social medicine departments and gives insights on individuals who are experiencing long-term unemployment. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Understanding the health and social circumstances of individuals seen at social medicine departments is essential for designing effective welfare and rehabilitation efforts. Understanding the health and social circumstances of individuals seen at social medicine departments is essential for designing effective welfare and rehabilitation efforts. Introduction About half of long-term unemployed individuals in Sweden (47%) and Denmark (57%) receive sickness or disability benefit interventions,1 indicating their unemployment is due to illness or disability. Similarly, approximately one-quarter of individuals aged 40–64 years not working, not seeking work or not available for work attribute their status to illness or disability across European Union member states.2 Additionally, across Organisation for Economic Co-operation and Development (OECD) countries, the proportion of those living with disabilities—ranging from long-term physical, mental and intellectual to sensory impairment—has risen over the last 15 years.3 Even in countries with social benefit systems, individuals with disabilities face elevated levels of inequality and poverty.46 This creates a subgroup of working-age individuals excluded from the labour market due to health issues and disabilities, who are likely to be socioeconomically disadvantaged. Likewise, previous research shows that a history of sick leaves increases the risk of transitioning to unemployment or disability pension,7 indicating that previous health issues, even if not resulting in unemployment, influence the risk of future negative employment outcomes. Many European countries, including Denmark, have implemented active labour market policies.1 While these policies have led to an increase in citizens receiving supported employment through flex-jobs—a Danish scheme that provides subsidised employment for individuals with permanently reduced work capacity—a segment remains permanently or long-term excluded from the labour market.8 In Denmark, a reform introduced in 2013 aimed to improve labour market rehabilitation for work-disabled individuals through a more active approach. The reform’s key components were more easily obtainable flex jobs for individuals with limited work capacity, individualised rehabilitation programmes and stricter criteria for granting disability pension. As part of the reform, social medicine departments in hospitals in each of the five regions of Denmark were to play a central role in clarifying the impact of social and health-related factors on functional limitations among the work-disabled.9 This includes assessment of the expected prognosis taking into account untried medical or therapeutic treatments and activities of rehabilitation under consideration of the International Classification of Functioning, Disability and Health model.10 The assessments are provided by a team of healthcare professionals including psychiatrists, rheumatologists, psychologists, physical therapists and social workers with the aim of either finding a timely resolution for the case or providing guidance for ways to move forward in the work rehabilitation process. The individuals who are assessed at the social medicine departments are referred mostly by the municipal job centres when the effort to rehabilitate to the labour market or the resolution of their case is not straightforward, and a more detailed examination and assessment is warranted. In most cases, these individuals are expected to have complex problems combining social and medical issues. To date, a more detailed description of this group on a national level has not been carried out. This register-based study aimed to describe the number of individuals registered in the national registers in contact with a social medicine department in Denmark. Further, this study aimed to describe the distribution of demographic, socioeconomic and health characteristics in this group and compare these with the distributions in a similar population regarding age, sex and municipality as well as employment status. Methods Setting The study was conducted within Denmark’s tax-funded welfare system, which supports individuals unable to work. Municipalities manage sick leave payments after 14 days (until 2018) or 30 days (after 2018). If sick leave exceeds 8 weeks, individuals are referred to municipal job centres. The job centre conducts interdisciplinary rehabilitation meetings with citizens requiring a rehabilitation plan to reintegrate into the labour market. Additionally, meetings are held for citizens being evaluated for flex jobs, disability pensions or more extensive long-term rehabilitation plans. Individuals can be referred to social medicine departments by the municipality if deemed relevant. The SocMed Capital Region Database contains data from hospital charts of individuals seen at the department of social medicine, Frederiksberg Hospital since 2012 and includes patients referred from the Capital Region only. The database contains over 150 variables from patient charts, registered in a REDCap database by secretaries. At an individual level, it contains data on healthcare use, social history, employment history, physical and mental illnesses, functioning and more (www.socmeddatabasen.dk). Study population and design To gather a Danish national cohort of individuals seen at social medicine departments (SocMedDep group), we included all first-time contacts from 1 January 2012 to 31 December 2022. To be included, the individual should be over 18 years old, have a valid Danish personal identification number (CPR) and be registered with either (1) a contact at one of the social medicine departments (online supplemental table S1 for department codes) or (2) a contact with the primary diagnosis of International Classification of Diseases (ICD)−10-code DZ048S (referencing a social medical examination) in the Danish National Patient Registry (DNPR)11 or (3) included in the SocMed Capital Region Database. We included patients both on diagnosis code and registered department and from the SocMed Database to ensure capturing all contacts possible through the registers at our disposal, as this is the first national registry study on these individuals. The specific department codes were obtained through interviews with current and former professors or chief doctors and/or secretaries at the departments in the five regions of Denmark, as well as searching for any department names with the phrasing ‘social’ in its name. The index date was defined as the day an individual had a registered visit to a social medicine department. In cases with registrations in both the SocMed Capital Region Database and in DNPR <2 months apart, the index date was set to the date registered in DNPR. The SocMedDep group was gathered at Statistics Denmark12 and matched with two populations. Match Group 1 was individually matched 1:5 on sex, birth year and municipality of residence at the index date. The ratio was chosen to increase statistical precision while avoiding substantial loss of comparators. Sex and birth year (age) were chosen as they are known confounders to both unemployment and baseline risk of morbidity and mortality. Municipality was chosen to account for regional variation in socioeconomic conditions and differences in municipal handling of recipients of benefits. Match Group 2 was matched 1:1 on the previous criteria and, additionally, on employment status. Employment status was added because those referred to social medicine departments tend to be individuals outside of the labour market. Without this second match group it would be difficult to determine whether observed differences reflected characteristics in our study population or broader patterns associated with unemployment in general. It was not possible to match directly on municipality; therefore, a three-class variable on the social class of the municipality was used instead. The ratio of 1:1 was chosen because of lack of possible controls to choose from. This variable was obtained from a socioeconomic index used to compare municipalities based on factors such as average wealth and level of education, determined by the Ministry of the Interior and Health of Denmark.13 Data sources The Danish Civil Registration System (CRS) contains a unique CPR for individuals with residence in Denmark. The CRS contains data on sex, date of birth, place of residence, citizenship and more, and allows cross-referencing with all other registers through the use of the CPR. The DNPR holds information on diagnoses and dates of all hospital contacts. The DNPR contains a range of data, including date and department of contact and diagnoses. Hospitals are required by law to submit information on all visits. The data quality in DNPR is considered to be good and especially suited for cohort studies.11 The Employment Classification Module (AKM) registry provides information on occupation and employment status of the Danish population liable for tax. The education registry contains information about finished education for individuals with a valid CPR number. Information is gathered from student authorisation registries as well as surveys for individuals with missing data. The family income registry contains data on personal and household income. Households include children living at home under the age of 25 years. The Danish substance abuse treatment registry (“Register over Stofmisbrugere i Behandling” SIB database) contains data on individuals who have undergone treatment for alcohol or drug abuse. Variables of interest Sex was stratified into men and women. Transgender citizens will appear as the sex they legally have at index date. Age was stratified into three categories (18–39 years, 40–59 years and 60+ years). The stratification at 40 years was chosen due to rules regarding disability pension changing slightly at that age. Immigration was defined as either being born in Denmark or not, and missing information on migration (<0.05%) was interpreted as no migration. Education was defined as the highest completed education at index date and grouped according to: not finished primary school (less than 9 years of education); primary school (9–10 years); high school equivalent or trade school (12–15 years); bachelor or higher (15+ years) and no information. This categorisation was chosen due to our own experience from the social medicine department that the study population has little or no education and uphold after analysing data to have groups of a meaningful size. Household income was classified into quartiles by comparing each individual’s household income in the index year with the national Danish income distribution for individuals of the same age, sex and calendar year. Employment status was gathered at AKM and categorised into employed; other incl. retired; not working and on cash benefit. AKM date is gathered at the end of the index year (online supplemental table S2). Years since last employment was grouped into 0 years; >0–5 years; >5–10 years; and >10 years and was retrieved from the AKM database. Addiction clinic was defined as a registration in the SIB database at any time before the index date. Comorbidities were derived from DNPR and grouped as in the M3 Comorbidity Index.14 The 61 M3 disease groups used are found in online supplemental table S4a. To identify comorbidities, a look-back of 10 years prior to the index date was used.14 15 The M3 Comorbidity Index was developed for comparing health outcomes at the population level rather than among clinical patients and includes groupings largely affecting mortality and morbidity.14 It was chosen in this study since the SocMedDep group is diverse with regard to diagnoses. The M3 score was categorised as 0 (low); >0–1 (moderate) or >1 (high). Multimorbidity was defined as the presence of five or more M3 disease groups. Psychiatric diagnosis was defined as the presence of a diagnosis from ICD-10 Chapter V: Mental and Behavioural Disorders (ICD-10 codes F00-F99) 10 years prior to the index date. Musculoskeletal disorder (MSD) diagnosis was defined according to the definition used in the Global Burden of Disease Study 2017,16 and the included diagnoses can be found in online supplemental table S3. To describe healthcare utilisation, the number of registered prescriptions was defined using the Anatomical Therapeutic Chemical classification17 second level and refers to the number of prescriptions within an organ group in the year prior to the index date. The number of laboratory tests performed was defined as the absolute number of tests performed in the year prior to the index date. The number of psychiatric contacts and the number of somatic contacts were defined as visits to corresponding hospital departments more than 12 hours apart and measured a year prior to the index date. Statistical analysis Variables have been analysed descriptively with the software package SAS Enterprise Guide V.8.3 64-bit (SAS Institute, North Carolina, USA). For categorical variables, percentages have been used. We performed two subgroup analyses of disease distribution in the M3 Index: one among patients with a psychiatric diagnosis and another among those with a musculoskeletal disease in the past 10 years. The project was approved and registered in the Capital Region’s inventory (P-2022-821). The use of data gathered from patient charts in the SocMed Capital Region Database was approved by the Danish Patient Safety Authority (case number 31–1521-145). No ethical approvals are required for registry studies in Denmark. Patient and public involvement Patients or the general public have not been involved in the design, reporting or plans for this study. Results After excluding 259 individuals due to the lack of suitable matches, a total of 13 092 individuals were included in the SocMedDep group, with Match Group 1 comprising 65 448 individuals and Match Group 2 including 13 092. Online supplemental table S4 shows that excluded individuals were similar to those included although differences were a higher percentage under 40 years, male and having had zero somatic hospital contacts within the last year prior to the index date in the excluded group. In the SocMedDep group, 11 377 (87%) were found in DNPR, with 6061 registered in a relevant department as well as with an ICD-10 code, and 5090 (39%) were in DNPR but not in the SocMed database (limited to the Capital Region of Denmark), while 1715 (13%) were only in the SocMed database and 6287 (48%) were found in both DNPR and the SocMed database. Figure 1 A and B, show the distribution of individuals included from the different pathways of inclusion over time and between the capital region and the other regions. Of note is the difference between inclusion before and after 2017. Before 2017, almost half of the inclusion is from the SocMed Database, whereas most of the included individuals after 2017 are found in DNPR by either department code, diagnosis or both. The distribution of individuals in the different inclusion pathways, stratified by geography (Panel A) and by year (Panel B). Less than 10 individuals were identified in the Central and Southern Denmark regions, preventing their inclusion. Registered visits increased in the Capital Region of Denmark and Region Zealand (figure 2). The Capital Region of Denmark had over 900 annual visits after 2017, while Region Zealand saw over 600 annually after 2017. Individuals at Danish social medicine departments (SocMedDep group) registered in each administrative region per year. Only years and regions with more than five individuals have been depicted. Table 1 outlines sociodemographics; the SocMedDep group had a higher proportion of women, mostly aged 40–59 years. In the SocMedDep group, 40.4% had completed a high-school level of education, similar to both matched groups; however, a much smaller percentage had finished a longer education. Furthermore, fewer individuals in the SocMedDep group had completed a bachelor’s degree or higher (17.4%) compared with Match Group 1 (36.1%) Table 1Demographic, education, economy, employment, comorbidity and healthcare use according to patient population at the social medicine departments (SocMedDep) and the matched samples†VariableValueSocMed dep groupMatch group 1Match group 2 N % (CI) N % (CI) N % (CI)13 09265 44813 092SexMan524640.1 (39.2 to 40.9)26 22740.1 (39.7 to 40.4)524640.1 (39.2 to 40.9)Woman784659.9 (59.1 to 60.8)39 22159.9 (59.6 to 60.3)784659.9 (59.1 to 60.8)Age, years<40321124.5 (23.8 to 25.3)16 04524.5 (24.2 to 24.8)321124.5 (23.8 to 25.3)41–59863566.0 (65.1 to 66.8)43 17666.0 (65.6 to 66.3)863566.0 (65.1 to 66.8)≥6012469.5 (9.0 to 10.0)62279.5 (9.3 to 9.7)12469.5 (9.0 to 10.0)Education5–10303323.2 (22.4 to 23.9)27944.3 (4.1 to 4.4)308123.5 (22.8 to 24.7)>10426432.6 (31.8 to 33.4)60789.3 (9.1 to 9.5)414431.7 (30.9 to 32.5)Contact with addiction clinicYes6224.8 (4.4 to 5.1)8541.3 (1.2 to 1.4)7535.8 (5.4 to 6.2)Number of somatic contacts (1 year prior to index date)0335525.6 (24.9 to 26.4)48 92974.8 (74.4 to 75.1)916870.0 (69.2 to 70.8)1–10392430.0 (29.2 to 30.8)11 56717.7 (17.4 to 18.0)218816.7 (16.1 to 17.4)11–20317024.2 (23.5 to 25.0)31454.8 (4.6 to 5.0)9397.2 (6.7 to 7.6)21–30143511.0 (10.4 to 11.5)9931.5 (1.4 to 1.6)3933.0 (2.7 to 3.3)>3012089.2 (8.7 to 9.7)8141.2 (1.2 to 1.3)4043.1 (2.8 to 3.4)Number of psychiatric contacts (1 year prior to index date)011 93291.1 (90.6 to 91.6)64 23898.2 (98.0 to 98.3)12 07392.2 (91.7 to 92.7)1–104213.2 (3.0 to 3.5)6180.9 (0.9 to 1.0)4923.8 (3.4 to 4.1)11–202902.2 (2.0 to 2.5)2430.4 (0.3 to 0.4)2041.6 (1.4 to 1.8)21–301921.5 (1.3 to 1.7)1210.2 (0.2 to 0.2)1120.9 (0.7 to 1.0)>302572.0 (1.7 to 2.2)2280.4 (0.3 to 0.4)2111.6 (1.4 to 1.8)Number of laboratory tests performed (1 year prior to index date)<10406531.1 (30.3 to 31.9)37 06056.6 (56.2 to 57.0)579644.3 (43.4 to 45.1)10–49414131.6 (30.8 to 32.4)17 68527.0 (26.7 to 27.4)380629.1 (28.3 to 29.9)≥50488637.3 (36.5 to 38.2)10 70316.4 (16.1 to 16.6)349026.7 (26.0 to 27.4)M3 score*0 (Low)324424.8 (24.0 to 25.5)40 93462.5 (62.2 to 62.9)528640.4 (39.5 to 41.2)>0–1 (Moderate)782859.8 (58.9 to 60.6)21 47332.8 (32.4 to 33.2)636848.6 (47.8 to 49.5)>1 (High)202015.4 (14.8 to 16.1)30414.7 (4.5 to 4.8)143811.0 (10.5 to 11.5)Multimorbidity (presence of fiveor more M3 disease groups)Yes8356.4 (6.0 to 6.8)8261.3 (1.2 to 1.4)4453.4 (3.1 to 3.7)Psychiatric diagnosis (10 years prior to index date)Yes714154.5 (53.7 to 55.4)766211.7 (11.5 to 12.0)436533.3 (32.5 to 34.2)Musculoskeletal disorder diagnosis (10 years prior to index date)Yes716254.7 (53.8 to 55.6)16 01224.5 (24.1 to 24.8)457935.0 (34.2 to 35.8)Number of registered prescriptions (1 year prior to index date)<5613846.9 (46.0 to 47.7)51 22978.3 (78.0 to 78.6)832763.6 (62.8 to 64.4)5–10466635.6 (34.8 to 36.5)11 41317.4 (17.1 to 17.7)352526.9 (26.2 to 27.7)>10228817.5 (16.8 to 18.1)28064.3 (4.1 to 4.4)12409.5 (9.0 to 10.0)*The M3 Comorbidity Index using 61 disease groups were used in this study.†Match Group 1 (matched on sex, age and municipality of residence) represents the general population and Match Group 2 (matched further on employment status) represents the background population with the same employment status. Demographic, education, economy, employment, comorbidity and healthcare use according to patient population at the social medicine departments (SocMedDep) and the matched samples† The M3 Comorbidity Index using 61 disease groups were used in this study. Match Group 1 (matched on sex, age and municipality of residence) represents the general population and Match Group 2 (matched further on employment status) represents the background population with the same employment status. Employment differed significantly, with only 15.5% in the SocMedDep group in some form of employment compared with almost 78% in Match Group 1. The SocMedDep group had fewer in the other category (including those on disability pensions), more individuals receiving cash benefits and longer unemployment durations aligning more with Match Group 2. Regarding household income, 73.3% of the SocMedDep group fell into the lowest quartile (compared with 32.4% in Match Group 1 and 70.2% in Match Group 2). Immigrants made up 24.3% of the SocMedDep group, higher than Match Group 1 (16.4%) but similar to Match Group 2 (25.6%). The SocMedDep group and Match Group 2 had higher addiction clinic registrations than Match Group 1. The SocMedDep group exhibited greater healthcare utilisation, with over 20% having more than 20 somatic hospital contacts in the prior year versus <3% in Match Group 1 and 6% in Match Group 2. They also had more psychiatric contacts, as did Match Group 2. The SocMedDep group had more laboratory tests done than Match group 1. In the SocMedDep group, only 24.8% had an M3 score of zero and 6.4% had multimorbidity. Prescription medication use was higher in the SocMedDep Group, particularly when compared with Match Group 1 and, over 17% of the SocMedDep group had more than 10 registered prescriptions versus 4.3% in Match Group 1 and 9.5% in Match Group 2. When looking over time, the percentage of individuals with psychiatric disorders increased in the SocMedDep group. This increase was also seen in Match Group 2 but cannot be found in Match Group 1. For MSD, there was no marked change over time (figure 3). Percentage of psychiatric or musculoskeletal disorders per year for the SocMedDep group, Match Group 1 (matched on sex, age and municipality of residence) and Match Group 2 (matched further on employment status). When looking at the individual M3 disease groups (online supplemental table S5a), the SocMedDep group had higher rates of alcohol and drug abuse, obesity, diabetes, chronic lung disease and heart disease than Match Group 1; these were similar in Match Group 2. The SocMedDep group had a higher rate of sleep disorders, metabolic disorders and endocrine disorders than both matched groups. The M3 subgroup analysis of individuals with a psychiatric diagnosis (online supplemental table S5b) showed that over 30% in all three groups had anxiety and behavioural disorders and over 35% had major psychiatric disorders. Further, there was an over-representation of major psychiatric disorders, alcohol abuse and drug abuse in Match Group 2 compared with both the SocMedDep group and Match Group 1. Subgroup analysis of individuals with a MSD showed that over 18% in the SocMedDep group had anxiety and behavioural disorders versus just over 5% in both other groups. There was a higher rate of obesity in both the SocMedDep group and Match Group 2. Discussion This register-based study highlights national disparities in reporting contacts with social medicine departments, resulting in a cohort predominantly from Denmark’s eastern regions. Individuals referred to these departments are primarily from lower-income groups, with limited education, higher immigration rates and prolonged labour market exclusion. When matched on employment, sociodemographic differences diminished; however, the SocMedDep group had higher psychiatric and MSD rates and greater healthcare utilisation, while the comparison group exhibited higher addiction rates. A rigorous methodology was employed to compile a comprehensive list of individuals seen at social medicine departments across Denmark, encompassing a wide range of departments and diagnostic codes. Social medicine departments counsel municipalities as well as patients. Whether the activity at the departments falls under treatment or not is subject to different legal interpretations in different parts of Denmark. It is obligatory for hospitals and clinics to report treatment activity to the health authorities according to The Danish Health Act (Danish title: ‘Sundhedsloven’)18 and The Danish Healthcare Professionals Authorisation Act (Danish Title: ‘Lov om Autorisation af Sundhedspersoner og om Sundhedsfaglig Virksomhed’).19 Since not all regions have interpreted their activity as treatment, the data are incomplete. While the clinical database in the Capital Region provided enhanced insight into the Capital Region, individuals from Funen and Jutland were under-represented. Despite this, Denmark’s demographic and legislative uniformity supports the nationwide relevance of our findings.20 Our results support the existing literature, indicating that individuals referred to social medicine departments, and their employment-matched counterparts, display greater health-related issues, lower education and lower income than those not matched on employment status. These findings align with studies suggesting that long-term unemployment is associated with heightened social and health-related challenges.2123 The data suggest that health-related factors are pivotal in differentiating social medicine patients from other unemployed individuals. The higher prevalence of psychiatric and somatic diseases in this group suggests that these patients are hard to rehabilitate and handle in the municipal system alone and are therefore referred to social medicine departments. Higher substance abuse rates in the employment-matched group indicate a potentially underserved subgroup. It points to a selection in the referral to the social medicine departments and potentially lacking services to two distinct groups of individuals—those with psychiatric disease and those with musculoskeletal disease. The rise in psychiatric diagnoses among social medicine patients, plateauing after 2018, mirrors national mental health trends,24 suggesting that a larger proportion of individuals referred are primarily affected by mental health issues. The strain on Denmark’s psychiatric healthcare system, due to resource shortages and recruitment challenges,25 may contribute to this trend, facilitating municipalities to refer individuals to social medicine departments. There are several strengths and limitations of this study. First of all, data in this study build on Danish national registries that are linked at the individual level and offer a high degree of coverage. Second, we designed inclusion to capture all contacts possible within those registries. It is however a limitation that a true national cohort could not be captured, although we also considered this an important result in itself. It was a limitation that diagnoses, especially psychiatric ones, were only hospital diagnoses as access to general practitioner diagnoses could have enriched the detail regarding minor psychiatric illnesses and therefore there is a risk that our findings underestimate psychiatric illness. This cohort, characterised by long-term unemployment and multifaceted challenges, warrants further investigation into outcomes following visits to social medicine departments, particularly in relation to their reintegration into the labour market. Recent reports indicate a decline since 2015 in individuals receiving cash benefits in Denmark,26 paralleling the absolute number of patients seen in social medicine departments. This correlation raises questions about the role of social medicine departments in influencing this trend. Given that cash benefits in Denmark are temporary and many individuals in our cohort have been unemployed for extended periods, further research into the impacts of social medicine department visits on employment outcomes is essential. In conclusion, this register-based study reveals an incomplete national registration of social medicine patients, which calls for initiatives to standardise registration at the national level. It further shows that individuals referred to social medicine departments are predominantly suffering from psychiatric or MSD, often face long-term unemployment and have limited pathways to rehabilitation or disability pensions. The results indicate the need for more targeted labour market reforms and social benefit programmes to improve employment outcomes for individuals with disabilities, also previously highlighted in international reports, for example, from the OECD.27 Further research is needed to understand the mechanisms driving these patterns and to explore potential interventions that could improve outcomes for these vulnerable populations. Supplementary material