Homeless Persons in the Emergency Room: A Single-Center Study on the Performance and Completeness of Recommended Diagnostic Tests

Homeless persons are subject to higher mortality, more chronic diseases, and multimorbidities than the general population (1, 2, 3, 4, 5). International studies have shown that homeless persons receive poorer medical treatment (3, 4). The interaction of social, physical, and psychological health problems causes additional needs (4).

For the first time in Germany we collected clinical and demographic data as well as quality indicators for the treatment of homeless patients in an emergency department. Subsequently we set out a discharge checklist, which is intended to strengthen the awareness of special needs and yield better-quality treatment.

Methods

We compared in a retrospective data analysis the treatment of homeless patients (homeless group, “HLG”) with that of a control group (non-homeless group, “NHLG”) in the time period 2010 to 2022 in a large emergency admissions unit in Stuttgart. 1139 patients were identified as HLG and were assigned to a precarious living/residential situation. This entails accommodation in emergency shelters as well as life without accommodation or in a bivouac/temporary shelter. 1479 cases of treatment were assigned to these patients. In the NHLG, 22,529 randomly selected persons from the remaining cases of the study period were studied.

We analyzed demographic data, reasons for presenting to the emergency room, rates of pathologic vital parameters, documentation quality, and inpatient admission rates.

The analysis is descriptive. To compare rates, we reported odds ratios with 95% confidence intervals (CI).

Results

Both groups included more men than women (men: women HLG 58.8%: 41.2%; NHLG 53.4%: 46.5%). The age distribution did not follow a normal pattern. The HLG had an age peak between 35 and 55 years, compared with the relatively homogenous age distribution in the NHLG. The median age in the HLG was 48 years and in the NHLG, 44 years.

In the HLG we identified eight reasons for attending, which in a minimum of 5% of cases in the HLG led to attendance in the emergency room (Figure 1).

Figure 1

Most common reasons for homeless persons attending emergency admission (“HLG”, n = 1479)

The proportion/rate of vital parameters we collected was for all categories lower in the HLG than in the NHLG. This was particularly pronounced for the respiratory rate (31.8% HLG versus 57.1% NHLG, OR 0.19; 95% confidence interval: [0.17; 0.21]), temperature (23.9% HLG versus 67.4% NHLG, OR 0.15 [0.13; 0.17]), Glasgow coma scale (GCS) (8.9% HLG versus 41.6% NHLG, OR 0.14 [0.11; 0.16]), and pupillary status (0.6% HLG versus 39.8% NHLG, OR 0.01 [0.005; 0.02]).

The rate of pathologic readings among the documented vital parameters differed substantially in the categories GCS (50.3% HLG versus 30.9% NHLG, OR 2.23 [1.59; 3.15]) and respiratory rate (69.4% HLG versus 58.2% NHLG, OR 1.65 [1.35; 2.10]).

The inpatient admission rate was lower for the HLG in the subgroups alcohol intoxication and generalized tonic-clonic seizure than in the NHLG (alcohol intoxication 17.9% HLG versus 52.1% NHLG, OR 0.2 [0.16; 0.25]; generalized tonic-clonic seizure 12.0% HLG versus 58.5% NHLG, OR 0.1 [0.06; 0.15]).

Rates of discharge against medical advice were low in the HLG (1.76%) and the NHLG (1.31%).

Discussion

The most common reasons for presenting to emergency admissions in the HLG were intoxication, falls, and generalized tonic-clonic seizures. Established quality indicators, such as in-hospital mortality and diagnostic agreement were inappropriate because of the low admission rate. For this reason we analyzed the quality of the documentation of the vital parameters.

Fewer vital parameters were documented in the HLG than in the NHLG. The lack of documentation of important parameters should be considered as an indicator for a lower quality of treatment than in the NHLG. We also confirmed this observation in the most common diagnostic subgroups (publication is in preparation). We interpret our finding of a lower rate of documented Glasgow coma scores in the context of a higher rate of pathologic values in the HLG to mean a higher risk for the lacking detection of a pathological Glasgow coma score. Lacking temperature readings in the HLG are also critical because this group of patients is more vulnerable to environmental influences.

Since only 12 patients in the HLG (1.1%) met the criterion for frequent re-attendances (≥4/year) and since all attendances in our emergency room require triage, medical assessment, and discharge we did not evaluate these cases separately. Since a total of 155 doctors from different specialties were involved in the treatment we consider our observations as an expression of a systemic problem. We think that implicit bias is a plausible approach to an explanation.

The additional needs of homeless persons are not reflected in the treatment process of an emergency room. This can trigger feelings of hopelessness in the medical personnel when confronted with multiple needs that cannot be tackled in this setting (4).

In emergency medicine, using checklists is an established method for managing risk. As a consequence of our findings we therefore developed a checklist as a rapidly implementable measure, in order to reduce putting patients at risk as a result of incomplete monitoring of vital parameters. From our literature search and evaluation of activity reports of organizations providing services for homeless people, as well as from information provided by people working in such organizations we identified continued outpatient treatment and basic care of basic needs as further factors that could be tackled in an emergency setting by means of a checklist (Figure 2). Overall, this should help to improve the documentation and treatment quality as well as tackle additional needs.

Figure 2

Checklist used at Klinikum Stuttgart hospital in the setting of outpatient care for homeless people

Johannes Heymer, Anna Sophie Hein, Matthias Ott, Martina Müller-Schilling, Tobias Schilling, Alexander Krohn, Florian Dengler

Conflict of interest statement
The authors declare that no conflict of interest exists.

Manuscript received on 11 April 2024, revised version accepted on 2 September 2024.

Translated from the original German by Birte Twisselmann, PhD.

Cite this as:
Heymer J, Hein AS, Ott M, Müller-Schilling M, Schilling T, Krohn A, Dengler F: Homeless persons in the emergency room—a single-center study on the performance and completeness of recommended diagnostic tests. Dtsch Arztebl Int 2024; 121: 812–3. DOI: 10.3238/arztebl.m2024.0183

References

1. Aldridge RW, Menezes D, Lewer D, et al.: Causes of death among homeless people: a population-based cross-sectional study of linked hospitalisation and mortality data in England. Wellcome Open Res 2019; 4: 49.

2. Lutchmun W, Gach J, Borup C, Froeschl G: Chronic diseases and multi-morbidity in persons experiencing homelessness: results from a cross-sectional study conducted at three humanitarian clinics in Germany in 2020. BMC Public Health 2022; 22: 1597.

3. Wadhera RK, Khatana SAM, Choi E, et al.: Disparities in care and mortality among homeless adults hospitalized for cardiovascular conditions. JAMA Intern Med 2020; 180: 357–66.

4. Albert R, Baillie D, Neal H: Unmet needs in street homeless people: a commentary on multiple interconnected needs in a vulnerable group. Future Healthc J 2023; 10: 103–6.

5. Bertram F, Hajek A, Dost K, et al.: The mental and physical health of the homeless—evidence from the National Survey on Psychiatric and Somatic Health of Homeless Individuals (the NAPSHI study). Dtsch Arztebl Int 2022; 119: 861–8.