Cardiovascular disease in the homeless: a systematic review of observational and interventional studies

BackgroundCardiovascular disease causes excess mortality and morbidity in socially excluded (inclusion health) populations. However, no specific guidelines exist for the treatment of cardiovascular disease in these individuals, including homeless populations. We aimed to better understand the disease burden and to identify evidence-based interventions in this setting by reviewing studies examining targeted interventions and prevalence of cardiovascular disease in homeless individuals.MethodsIn this systematic review, we searched Embase for observational and interventional studies of cardiovascular disease in homeless populations published on or before Dec 19, 2018, using the keywords “heart diseases”, “adrenergic beta-antagonists”, “cardiotonic agents”, “cardiovascular diseases”, “cardio-renal syndrome”, “hypertension” and “homeless [population]”. Two independent reviewers selected eligible articles and extracted relevant data, and used the Newcastle-Ottawa Scale to grade the quality of studies. The main outcome, assessed by a random-effects meta-analysis, was prevalence of cardiovascular disease (International Classification of Diseases-10 definition).FindingsOf 273 full-text articles that were retrieved, 17 (six case-control, 11 cohort) met the inclusion criteria, none of which considered interventions for cardiovascular disease. 13 (76%) of 17 were high quality, and all were based in Europe or North America, including 765 459 individuals, of whom 32 721 were homeless. 12 (71%) of 17 studies were published before 2011. Homeless individuals were more likely to have cardiovascular disease than were non-homeless individuals (pooled odds ratio [OR] 2·96 [95% CI 2·80–3·13]; p<0·0001; p_{heterogeneity}<0·0001). North American homeless individuals were more likely to have cardiovascular disease (pooled OR 3·07 [2·90–3·25]; p<0·0001; p_{heterogeneity}=0·0070) than their European counterparts (pooled OR 2·15 [1·81–2·54]; p<0·0001; p_{heterogeneity}=0·125). Compared with non-homeless individuals, hypertension was more likely in homeless people (pooled OR 1·38–1·09, 1·75; p=0·0070; heterogeneity p=0·94.InterpretationHomeless populations are at increased risk of cardiovascular disease. However, studies to-date are limited in number, geography and contemporary relevance with no interventional studies and considerable heterogeneity between existing studies. The increased burden of cardiovascular disease in homeless populations requires better characterisation and interventions need to be designed and tested urgently. Cardiovascular disease needs to be studied further in this setting if integrated care in homeless individuals is to become a reality.FundingNational Institute of Health Research Programme Development Grant (RP-DG-0117–10003).