287 Homeless Patients in France: A National Case-Cohort Prospective Study

Abstract

Homeless people represent a vulnerable population. Homelessness is associated with reduced life expectancy, increased risk of comorbidities, and increased incidence of psychiatric illness. Public hospitals and their emergency departments (EDs) are known to be used frequently by these patients. They can be seen as difficult to treat, and have an increased incidence of substance abuse and risk of violence in the ED. We aimed to analyse a large sample of homeless patients to determine the quality of care delivered to homeless people in French EDs. We tested the hypothesis that homeless patients experience suboptimal care by the provision of fewer health care resources. We conducted a prospective multicenter case-control study in 31 EDs in France. Our institutional review board authorized the study without the need for signed informed consent. We defined a homeless patient as a patient who currently lives on the street or in a shelter. During 72 hours from March 3, 2015, all homeless patients who visited the participating EDs were included in the study. One control patient was prospectively recruited after each case was included: the next patient who visited the ED with similar severity triage level (on a one to four scale), similar age (±ten years) and same sex. We retrieved demographic and social characteristics of included patients, along with their vital parameters, and characteristics of ED utilization. The primary outcome measures were length of stay, number of investigations per patient and treatment in the ED. A total of 212 homeless patients and 212 control patients were included in the study. Mean age was 44 (SD 13) years in both groups, and 87% were male. Homeless patients were more likely to have visited the ED in the past 28 days than other patients (47% vs 10%, P < .001). They presented with similar rates and types of comorbidities than control patients, except for a more frequent history of substance abuse. Heart rate, blood pressure, temperature, capillary blood glucose and Glasgow Coma Scale score were similar in both groups. Chief complaint was “housing demand” for 30 (14%) homeless patients. After excluding them, we found no difference in the type of chief complaint except for alcohol abuse, more frequent in homeless patients (20% vs 4%, P < .001). We found a similar median waiting time to physician assessment in the two groups (58 minutes for both), although mean length of stay was longer for homeless patients than for control patients (6.2 vs 3.9 hours, P < .001). We found no significant difference in the rate of radiological or biological investigations between the two groups. Similarly, we found no significant difference for the rate of oral or parenteral treatment administration, and admission rate was similar in the two groups (9% vs 7%, P = .6). Amongst the 182 analyzed homeless patients that visit the ED beside a housing demand, 53 (29%) were uninsured. We did not find a difference in the level of medical care delivered in French ED to homeless patients when compared to matched control. Resource consumption was similar for both groups, as was the admission rate. Nevertheless, homeless patients visit ED more often for an alcohol-related complaint, are often uninsured and have higher rates of return visit.