This is a registry-based cohort study with a matched reference group, investigating employment rates in dialysis and/or transplantation patients in the period 2005–2019. Index time was defined as the first initiation of either dialysis or kidney transplantation. As most transplantation patients take sickness leave during the peri- and postoperative period, employment status at index time was based on their status four weeks before transplantation. This study adheres to the STROBE and RECORD checklists.

The study adheres to the tenets of the Declaration of Helsinki. The study only includes pseudonymized registry data and therefore does not require review of an ethics committee or consent from participants according to Danish Law (Scientific Ethical Committees Act § 14, part 2). The study was reviewed and approved of Statistics Denmark (project number 708072/MMK).

The study populations were defined as patients initiating chronic dialysis treatment (dialysis cohort) or having kidney transplantation (transplantation cohort) based on registrations in the Danish Nephrological Register (Dansk Nefrologisk Selskabs Landsregister, DNSL) between 2005 and 2019. Inclusion criterion was age 18–65 years. Exclusion criteria were patients not included in the Civil Registration System (CPR) three years before and after index time or no entry in the DREAM database (Danish Register for Evaluation of Marginalization). A patient can enter both the dialysis cohort and the transplantation cohort.

Almost all patients were matched with three references from the background population (> 97% of the study population), and all patients were matched with at least one reference. The reference group was matched by year of birth, gender, and latest known municipality.

Employment was classified as being without social benefits registered in DREAM. Unemployment was defined as receiving social benefits or being on sick leave. Permanently out of the workforce was defined as early retirement or pension (specific DREAM codes are listed in Supplementary Table 1).

The primary outcome was the employment rates at five different time points for the dialysis cohort or transplantation cohort: 3 years before, 1 year before, at the initiation of dialysis or transplantation, 1 year and 3 years after. The secondary outcome was differences in employment rate over 3 time periods; 2005–2009, 2010–2014, and 2015–2019, for respectively dialysis or transplantation patients compared to the reference group.

Data was linked between 4 databases: DNSL, DREAM, Statistics Denmark, and the Danish National Patient Register.

DNSL, as a part of the Danish Clinical Quality Program– National Clinical Registries (RKKP), includes all patients with ESKD since 1990 and contains information on dialysis modality, changes between dialysis modalities and time of kidney transplantation. As no private nephrology clinics exist in Denmark, DNSL covers the entire population.

The DREAM register includes all citizens who have received transfer income since 1991 and reports weekly on which benefits have been paid. Because of changes in the DREAM register, we excluded patients or controls if they had no entry in DREAM. People who never received any state transfer payment before their death and died before 2008 were not registered in DREAM. After 2008 all Danish people with any monthly income from either state transfers or regular wage had been registered in the DREAM database.

Statistics Denmark provided information concerning socio-economic conditions, including education and income. In addition, this database includes information such as ethnic background and municipality of residence.

The Danish National Patient Register contains information about contacts at Danish hospitals, including dates of contacts and diagnosis codes according to the International Classification of Diseases, Tenth Revision (ICD-10).

In the registries, ethnicity is reported as Danish if the person, regardless of birthplace, has at least one parent who is born in Denmark and is a Danish citizen, or if the person was born in Denmark regardless of the parent’s status. A person is reported as immigrated if born outside of Denmark, none of the parents are born in Denmark and registered with Danish citizenship.

The following ICD-10 diagnoses were used: DE10-14 for diabetes, DI109 and DI151 for hypertension, DI20-25 for ischemic heart disease, and DF31-DF33 for affective disorders. The validity and sensitivity of identifying ischemic heart disease and diabetes is high using ICD codes in Denmark, is low regarding hypertension and unknown when identifying affective disorders [19].

Based on a random sample of the Danish population matched on index year and age at index year, income quartiles were estimated, and each patient’s income was evaluated in relation to these quartiles.

Data were analysed using SAS Enterprise Guide version 8.3 Update 7 (8.3.7.202) (64-bit). The distribution of characteristics at index time was presented by numbers (percentages) and differences between patients and the reference group were assessed by chi-square tests.

Employment rates are presented in Sankey diagrams demonstrating proportions for each group (patients and controls) and how proportions change from one time point (only) to the adjacent time point (i.e., it is not possible to follow one patient from the first bar to the last).

Employment rates at index time were presented for three groups having year of index in 2005–2009, 2010–2014 or 2015–2019. Changes in employment were assessed with logistic regression analyses and differences between patients and controls were assessed by an interaction.