Life-years lost in people experiencing homelessness and other high-risk groups in Denmark: a population-based, register-based, cohort study

BACKGROUND: People experiencing homelessness have a high mortality risk. Estimates of absolute risk, such as life-years lost, provide a more accurate measure of mortality differences in transient populations than conventional life expectancy estimates and enable direct comparisons with the general population and other high-risk populations. We aimed to examine mortality risks and life-years lost in people experiencing homelessness and other high-risk groups by sex, age group, and cause of death.

METHODS: We conducted a nationwide, register-based cohort study of people aged 17-100 years living in Denmark between Jan 1, 2002, and Dec 31, 2020. Data from the Danish Civil Registration System was linked to the Homeless Register and the Cause of Death Register. The outcome was overall and cause-specific mortality during follow-up. We estimated mortality rates per 10 000 person-years at risk and estimated adjusted mortality rate ratios using Poisson regression analysis. We calculated both average and age-specific life expectancy and life-years lost for people experiencing homelessness and compared these figures with those of the general population, as well as with individuals with schizophrenia, alcohol use disorder, and drug use disorder.

FINDINGS: Among 6 286 512 Danish residents, 58 376 (0·9%) people had at least one homeless shelter contact during the study period. Among people with a history of homelessness, the mortality rate was 240·3 (95% CI 236·3-244·4) per 10 000 person-years. In the general population with no history of homelessness, the mortality rate per 10 000 person-years was 117·8 (95% CI 117·6-118·1) in the general population with no history of homelessness, which resulted in an absolute difference of 122·5 (118·5-126·5). The mean excess life-years lost was 15·9 (95% CI 15·8-16·2) for males and 15·3 (14·8-15·7) for females compared with the general population, which was higher than in other high-risk groups, such as people with schizophrenia, alcohol use disorder, and drug use disorder (with and without homelessness history). Most of this excess mortality in people experiencing homelessness was attributed to external causes of death, psychiatric disorders, and diseases of the liver and digestive system. Life-years lost from external causes of death, including suicide and unintentional injuries, was 5·2 (95% CI 5·0-5·5) for males and 3·9 (3·5-4·2) for females, compared with the general population.

INTERPRETATION: The substantial number of life-years lost among people experiencing homelessness is a key public health concern, highlighting pronounced inequalities in health among some of the most vulnerable individuals. Such metrics should be useful to inform policies and preventive efforts to reduce excess mortality in this population.

FUNDING: Independent Research Fund Denmark.