Homeless people in England died at a median age of 51.6, linked-data study found

Homelessness is one of the most severe health exposures that epidemiology measures. In a study linking hospital admissions to death records in England, homeless people who had recently been in hospital died at a median age of 51.6 years, against 71.5 years for a comparison group drawn from the most deprived neighbourhoods [s1] — and nearly a third of those deaths were from causes that timely health care could have treated [s1].

A twenty-year gap, and a treatable share

The 2019 analysis linked records for 3,882 homeless hospital admissions across 17 specialist integrated homeless health schemes between late 2013 and 2016 to 600 subsequent deaths [s1]. The median age at death of 51.6 years (interquartile range 42.7–60.2) sat roughly two decades below the 71.5 years (60.7–79.0) in the high-deprivation comparison sample [s1]. That the comparison group was itself poor matters: the gap is not homelessness versus affluence but homelessness versus already-deprived housed people, which isolates the added toll of losing stable housing.

The causes were not dominated by any single condition. The leading underlying causes were external causes such as injury and poisoning (21.7% of deaths), cancer (19.0%), and digestive disease (19.0%) [s1]. After weighting for age and sex, 30.2% of the homeless deaths were from causes considered amenable to timely and effective health care, compared with 23.0% in the comparison group [s1]. In other words, a substantial part of the excess is not inevitable; it reflects care that did not reach people in time.

What the exposure does across the board

A 2014 Lancet review set the scale of the problem: on any given night more than 400,000 people are homeless across the European Union and more than 600,000 in the United States [s2]. The review found homeless people carry higher premature mortality than the general population — especially from suicide and unintentional injuries — alongside raised rates of infectious disease, mental disorders, substance misuse, and non-communicable diseases, with evidence of accelerated ageing [s2]. It also flagged a paradox familiar to clinicians: homeless people engage less with routine care and adhere less to treatment, yet attend emergency departments far more often than housed people [s2], which is exactly the pattern that produces treatable deaths.

The most quantified mortality estimate comes from a 2018 Lancet meta-analysis of 337 studies covering four overlapping socially excluded groups — homeless people, prisoners, sex workers, and people with substance use disorders [s3]. Pooled all-cause standardised mortality ratios were 7.88 for men (95% CI 7.03–8.74) and 11.86 for women (10.42–13.30), meaning death rates roughly eight to twelve times those of the general population [s3]. The highest category-specific ratios were for injury, poisoning, and other external causes, at 7.89 in men and 18.72 in women [s3]. Because that meta-analysis combines homeless populations with three other excluded groups, its numbers describe the broader "inclusion health" population rather than homelessness alone — but homeless people are a large part of it, and the direction is unambiguous.

Housing as the structural determinant

The 2014 review is explicit that the availability of low-cost housing is thought to be the most important structural driver of homelessness, interacting with individual factors such as poverty, family breakdown, mental illness, and substance misuse [s2]. That framing puts homelessness squarely in the cost-of-living story: it is where housing affordability, income, and health collide most visibly. The English study noted that homelessness there had risen 165% since 2010 [s1], a reminder that the exposure tracks housing-market conditions rather than being fixed.

The limits of the evidence

These are observational data, and the people who become homeless differ from the housed in ways — chronic illness, addiction, adverse childhoods — that independently shorten life [s2][s3]. Some of the mortality gap reflects those prior conditions rather than homelessness itself, and the linked-data study captured people who had been admitted to hospital, who are sicker than the wider homeless population [s1]. The 2018 meta-analysis also reported very high statistical heterogeneity between studies, so its pooled ratios are best read as an order of magnitude rather than a precise figure [s3]. None of these caveats reverses the central finding; they bound how tightly it can be attributed to housing status alone.

What to watch

The most policy-relevant number in this literature is the treatable fraction. If roughly 30% of homeless deaths are amenable to health care [s1], the question is whether integrated homeless health services — the schemes the English study was built around — measurably shift it. Evaluations of "Housing First" and specialist inclusion-health teams are the place that question is being tested, and the metric that would signal progress is a narrowing of the amenable-mortality gap, not just more contact with services.

Sources

• [s1] Causes of death among homeless people: a population-based cross-sectional study of linked hospitalisation and mortality data in England — Wellcome Open Research, published 11 March 2019. https://doi.org/10.12688/wellcomeopenres.15151.1
• [s2] The health of homeless people in high-income countries — The Lancet, published 25 October 2014. https://doi.org/10.1016/S0140-6736(14)61132-6
• [s3] Morbidity and mortality in homeless individuals, prisoners, sex workers, and individuals with substance use disorders in high-income countries: a systematic review and meta-analysis — The Lancet, published 20 January 2018. https://doi.org/10.1016/S0140-6736(17)31869-X