Homelessness, type of homelessness, and risk of cause-specific mortality: a systematic review and meta-analysis of 116 studies comprising 2,563,633 homeless people and 129,292,553 population controls

BackgroundHomelessness might increase the risk of premature mortality, but evidence is scarce, imprecise, and is mostly limited to rough sleepers as opposed to more common types of homelessness. MethodsPublished studies were retrieved through a systematic search of MEDLINE, Embase PsycINFO and Scopus from inception to December 2024. Unpublished data were identified from open-access data archives. We used random-effects meta-analysis to combine effect estimates from published and unpublished data. This review is registered at PROSPERO (CRD42023430984). FindingsWe included 116 studies from Europe, the USA, South America, Africa, Asia, and Australia. The meta-analysis of all-cause mortality comprised 110,892,271 people (1,618,049 exposed to homelessness). The risk of all-cause mortality was significantly increased in people exposed to homelessness (Relative risk [RR] 2{middle dot}12 [95% CI 1{middle dot}91-2{middle dot}57], p<0{middle dot}001, I2=99{middle dot}7%), with risks similar in men (3{middle dot}88, 2{middle dot}69-5{middle dot}06) and women (3{middle dot}46, 2{middle dot}17-4{middle dot}70). This risk was most elevated in people who had slept rough (7{middle dot}63, 3{middle dot}29-11{middle dot}97), followed by those who used low-cost hotels (5{middle dot}18, 1{middle dot}14-9{middle dot}23), then hostels (3{middle dot}44, 2{middle dot}10-4{middle dot}77). In analyses of cause-specific mortality (26,291,900 people, 1,202,205 homeless), summary RR estimates were elevated for 33 of the 36 (92%) causes of death and highest for deaths due to psychoactive substance use disorder (21{middle dot}36, 14{middle dot}44-31{middle dot}67), accidental injuries (13{middle dot}15, 5{middle dot}46-31{middle dot}69), drug-overdose (10{middle dot}80, 6{middle dot}37-18{middle dot}31), and those that are alcohol-related (5{middle dot}93, 1{middle dot}10-22{middle dot}04). No evidence of publication bias was detected. InterpretationHomeless people experience an increased risk of premature mortality across an array of health outcomes. That the most extreme inequities have an interrelated aetiology suggests a cross-sectoral medical, housing, and social care response is required. FundingThe Centre for Homelessness Impact, Health and Care Research Wales, UK Medical Research Council (MR/P023444/1) and the US National Institute on Aging. Research in contextO_ST_ABSEvidence before this studyC_ST_ABSWhile existing systematic reviews have found that homelessness is associated with an increased risk of all-cause mortality, there remains substantial gaps in knowledge. Currently there is no meta-analysis of the risk of death from all-causes and specific disorders, comparison of mortality risk across the spectrum of homelessness experience (e.g., rough sleeping, hostels, single-occupancy low-cost hotels), whether differences exist in the impact on men and women, or in high, or low and middle income countries (LMICs). Accordingly, we searched MEDLINE, Embase, PsycINFO and Scopus from inception to December 2024 for studies reporting mortality outcomes among people with a history of homelessness. Additionally, we searched the grey literature and obtained unpublished individual-participant data for two cohort studies from open-access data archives. Added value of this studyOur systematic review and meta-analysis provides the first comprehensive examination to date of mortality among people who have experienced homelessness. Drawing on 116 studies, there was, in aggregate, a doubling in risk of all-cause mortality in the 1,618,049 people who had experienced homelessness relative to the 109,274,222 population controls (relative risk [RR] 2{middle dot}12, [95% CI 1{middle dot}91-2{middle dot}57], p<0{middle dot}001, I2=99{middle dot}7%). This association was strongest in people who had slept rough (7{middle dot}63, 95% CI 3{middle dot}29-11{middle dot}97) than in a hostel (3{middle dot}44, 95% CI 2{middle dot}10-4{middle dot}77) and was of equal magnitude in men (3{middle dot}88, 95% CI 2{middle dot}69-5{middle dot}06) and women (3{middle dot}43, 95% CI 2{middle dot}10-4{middle dot}70). Summary RRs were extremely high for some causes of death, including RRs of >20 for psychoactive substance use disorder, 10-19 for drug-overdose, accidents, and accidental injuries, and 5-10 for alcohol-related causes - none of which have been previously summarised. No evidence of publication bias was found. Implications of all the available evidenceHomelessness is associated with an increased risk of all-cause mortality. Our analysis suggests these a large proportion of these excess deaths are caused by substance use, drug-overdose, accidents and harmful alcohol consumption. The extreme inequalities in mortality we identified points to the need a cross-sectoral response to improve both health and housing conditions for people who have experienced homelessness.