Importance: Homelessness is a complex challenge with an estimated yearly economic burden of $6 billion in the United States. Mitigating homelessness requires an understanding of determinants of homelessness, their interaction with health factors, and quantification of impact. Objective: To investigate the health, social and policy factors influencing homelessness in a longitudinal integrative machine learning analysis. Data Sources and Study design: This retrospective longitudinal study integrated Global Burden of Disease (GBD), Health Inequality, and Housing and Urban Development (HUD) datasets for 3131 counties in the United States. We used the disease burden data of 2014, health inequalities data of 2001-2014, and homelessness count of 2015. Primary Outcome and Measurement Results: Homelessness, the burden of disease, health inequalities, economic policies, ethnic, social, and racial factors. Methods: Spearman rank correlation test was performed to check pairwise associations. A unified probabilistic model with temporal causality was fitted using a data-driven structure learning algorithm. The resulting associations adjusted for other variables in the network were quantified using network inference algorithms. Finally, counterfactual analysis was performed to quantify the potential impact of the learned interventions. Results: The total burden of homelessness was significantly (p<0.001) and positively associated with rates of HIV and hepatitis mortality. Inference from the unified probabilistic model indicated that a state with a high hepatitis mortality rate had a 9% higher homelessness. Further, the rate of rheumatic heart disease mortality had a 29% decrease with the provision of shelter in young adults experiencing homelessness (p<0.001). Finally, states with moderate tax progressivity had a mitigating effect on homelessness as compared to both high and low tax progressivity ( 2% and 5% respectively). We evaluated the counterfactual impact of policy interventions to provide more support to cancer patients to prevent homeless and provision of shelter to prevent rheumatic heart disease mortality in young adults experiencing homelessness. Conclusion and Relevance: Control of infectious diseases and the implementation of tax policies are critical interventions for the reduction of homelessness in the United States.

The authors have declared no competing interest.

None

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

HUD.gov / U.S. Department of Housing and Urban ...https://www.hud.gov

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.

Yes