- Email: [email protected]
Homeless patients tend to have greater psychiatric needs when presenting to the emergency department,
YAJEM-158542; No of Pages 4 American Journal of Emergency Medicine xxx (xxxx) xxx
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Homeless patients tend to have greater psychiatric needs when presenting to the emergency department,☆,☆☆ Lauren E. Lamparter a,⁎, Megan A. Rech, PharmD b,c, Theresa M. Nguyen, MD FACEP c a b c
Medical Student, MS3, Loyola University Chicago Stritch School of Medicine, 2160 S 1st Ave, Maywood, IL 60513, United States Loyola University Medical Center, Department of Pharmacy, 2160 S 1st Ave, Maywood, IL 60513, United States Loyola University Medical Center, Department of Emergency Medicine, 2160 S 1st Ave, Maywood, IL 60513, United States
a r t i c l e
i n f o
Article history: Received 15 September 2019 Received in revised form 9 October 2019 Accepted 12 October 2019 Available online xxxx Keywords: Homeless Psychiatric Emergency department Chief complaint
a b s t r a c t Introduction: Homeless patients tend to visit Emergency Departments (EDs) more frequently than the nonhomeless population. The goal of this study was to assess differences in chief complaint, medical conditions, and disposition between homeless patients compared to non-homeless patients presenting to an urban ED. Methods: This was a retrospective cohort of homeless patients ages ≥18 years compared to non-homeless controls from January 1, 2017 to December 31, 2017. Exclusion criteria were as follows: direct admission to hospital floor, repeat visits, or leaving without being seen. The primary endpoint of this study was to assess differences in chief complaint of homeless versus non-homeless patients upon presentation to the ED. Our secondary endpoints included differences in ED utilization between the two groups, in terms of length of stay, ambulance use, diagnosis, and disposition. Results: Homeless patients were more likely present to the ED for a psychiatric evaluation (homeless group 34% vs. non-homeless group 4%, p b 0.01) and have a history of a psychiatric diagnosis (56% vs. 10%, p b 0.01) compared to non-homeless controls. Homeless patients also tended to require more ambulance transport (46% vs. 16%, p b 0.01). More homeless patients were transferred to a psychiatric facility (40% vs. 1%, p b 0.01), while the majority of non-homeless patients were discharged home (50% vs. 93%, p b 0.01). Conclusion: This study found that homeless patients had a significantly higher association with psychiatric diagnoses and greater ED utilization than non-homeless. This suggests the importance of increased access to consistent psychiatric care and follow up within the homeless population. © 2019 Elsevier Inc. All rights reserved.
1. Introduction The National Alliance to End Homelessness reported that on any given night in January of 2017, approximately 553,742 people experienced homelessness across the country [1]. A person is classified as homeless if he or she does not have a permanent or temporary address of residency [2]. A survey performed at an urban medical center in 2017 reported that the number of homeless people living exposed without shelter has increased from prior years [3]. Living without shelter places individuals at a greater risk for many acute medical problems, such as exposure injuries or sexual assault, as well as exacerbation of chronic medical problems, including malnutrition, liver disease, dental disease, ☆ This research was presented at the 2019 AAEM/RSA & West JEM Population Health Research Competition in Los Vegas, NV on March 11, 2019. ☆☆ This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The data presented and analyzed in this study is available upon reasonable request. ⁎ Corresponding author. E-mail addresses: [email protected] (L.E. Lamparter), [email protected] (M.A. Rech), [email protected] (T.M. Nguyen).
or psychiatric disorders, which could be otherwise manageable [4]. In addition, a previous study of homeless utilization of the Emergency Department (ED) found that a majority of homeless patient visits to the ED were directly related to excessive alcohol use [2]. As a result of their increased exposures, chronic illnesses, and substance abuse, homeless adults tend to visit EDs more often and more frequently than the nonhomeless general population [2]. It is relevant then, for emergency physicians to understand the specific reasons why homeless patients seek emergency care. The goal of this study was to assess homeless ED utilization by presenting chief complaint, use of ED resources, and final disposition compared to non-homeless controls in order to better understand and potentially prepare to address the needs of this population. 2. Methods This was a retrospective analysis of homeless ED patients compared to non-homeless ED patients from January 1, 2017 – December 31, 2017. Institutional Review Board approval was obtained prior to conducting this study. This study was conducted at an urban, level 1
https://doi.org/10.1016/j.ajem.2019.10.012 0735-6757/© 2019 Elsevier Inc. All rights reserved.
Please cite this article as: L.E. Lamparter, M.A. Rech and T.M. Nguyen, Homeless patients tend to have greater psychiatric needs when presenting to the emergency department..., American Journal of Emergency Medicine, https://doi.org/10.1016/j.ajem.2019.10.012
2
L.E. Lamparter et al. / American Journal of Emergency Medicine xxx (xxxx) xxx
trauma and tertiary referral center in Illinois, where the ED serves approximately 50,000 patients per year. Patients were identified via their demographic status as “indigent” or an International Classification of Disease, tenth revision (ICD-10) codes for homelessness and inadequate housing. The ICD-10 codes included the following: Z59.0 homelessness, Z59.1 inadequate housing, Z59.3 problems related to living in residential institution, Z59.5 extreme poverty, and Z59.8 and Z59.9 problem related to housing and economic circumstances, unspecified. Since there is fluidity in the circumstances of the homeless population, (for example, they might lack a permanent address but not live full-time on the streets), a range of ICD-10 codes were needed to capture the variations within this population. In the event that an ICD-10 code for homelessness was not assigned or the patient was not classified as “indigent,” homeless patients were also identified by searching for the study institution’s address in the patient address field. All homeless patients identified by these methods were initially included in the study. These methods of identifying homeless patients were chosen based on a previous study which had used a patient’s lack of address to identify their homeless population [2]. In addition, the standard procedure at the study institution is to list the hospital’s address if a patient is unable to provide an address upon admission. A systematic sample of non-homeless patients were identified by including every tenth ED visit beginning on January 1, 2017. Patients in both groups were excluded if they were under 18 years of age, were directly admitted to the hospital, or left without being seen. If the patient was seen in the ED multiple times during the study period, only the first visit was included. For each patient visit, the following information was recorded: date of visit, mode of transportation to the ED, chief complaint, blood alcohol and urinary toxicology screening results, diagnosis upon discharge, number of prior ED visits, and final disposition. Demographic information, past medical history (including psychiatric or substance abuse history), and insurance status were also collected. Use of hospital resources by homeless patients was assessed by ambulance use, ED length of stay, insurance status, labs ordered, disposition, and repeat visits. Baseline characteristics were described using mean, standard deviation, median, interquartile range (IQR) and percentages. Shapiro Wilks test was used to assess normality of continuous data. Continuous parametric variables were compared using a t-test. Continuous nonparametric data were analyzed utilizing the Mann-Whitney U test. Chi-square test or Fischer's exact test were used to compare categorical variables. A p value of b 0.05 was considered significant.
3. Results In 2017, there were approximately 46,000 ED visits, of which 208 were identified as homeless by ICD-10 diagnostic codes, lack of an address, or demographic status as “indigent.” A systematic sample of 178 non-homeless patients were obtained as a control from all patients who visited in 2017. After exclusion criteria were applied, 68 homeless patients were compared to 70 control patients (Fig. 1). Baseline demographics are displayed in Table 1. Homeless patients were more likely be men (76% homeless group vs. 24% non-homeless group, p b 0.01) and to have an underlying psychiatric diagnosis (55% vs. 10%, p b 0.01) (Table 1). Homeless patients were significantly more likely to present to the ED for a psychiatric evaluation (34% vs. 4%, p b 0.01) (Table 2). Interestingly, only three percent of homeless patients presented with a chief complaint related to an injury while 10% of non-homeless presented after an injury or fall, though this was not a significant difference (p = 0.17) (Table 2). Homeless patients reported greater incidence of illicit drug use (34% vs. 14%, p b 0.01) (Table 1). Though homeless patients were more likely to report illicit drug use, only 9% of homeless presented with acute intoxication, which was not a significant difference from the 4% of non-homeless patients who presented acutely intoxicated (p = 0.28) (Table 2). There
Total Charts Reviewed (n= 386) Homeless and Non-homeless
Exclusion Criteria: • • • •
Final Homeless (n= 68)
Age < 18 (n=48) Direct admit to the floor (n=71) Left without being seen (n= 16) Subsequent visits in 2017 (n=113)
Final Non-homeless (n= 70) Fig. 1. Study flow diagram.
was no difference in the incidence of alcohol abuse between groups (Table 1). Homeless patients were more likely to present to the ED by ambulance transport (46% vs. 16%, p b 0.01) and had longer ED length of stays (median 8 h, range 4–13 vs. median 5 h, range 3–7, p = 0.01) (Table 2). The majority of homeless patients were insured by Medicare or Medicaid (71% vs. 51%, p b 0.01), and while only 19% of the homeless patients were uninsured, they were more likely to be uninsured than the non-homeless patients (19% vs 13%, p b 0.01) (Table 2). In terms of final disposition, homeless patients were significantly less likely to be discharged, compared to non-homeless patients, the majority of whom were discharged home (50% vs. 93% p= b0.01) (Table 2). Of the patients who were not discharged, 40% of the homeless were transferred to psychiatric facilities, compared to 2% of the non-homeless population (p b 0.01) (Table 2). There was no difference between the two groups in regards to total number of ED visits or readmissions. 4. Discussion This study demonstrated a significant increase in homeless patient’s chief complaints related to psychiatric illness compared to the nonhomeless controls, and found that homeless paitents had significantly more psychiatric conditions, illicit drug use, and more frequent use of ED resources such as ambulance transport. Furthermore, homeless patients were significantly more likely to have a history of a psychiatric Table 1 Baseline demographics of homeless compared to non-homeless patients. Demographics
Homeless (n = 68)
Non-Homeless (n = 70)
p value
52 (76) 40 (28–55)
16 (24) 40 (28–55)
b0.01 0.81
33 (49) 27 (40) 8 (12) 7 (10)
32 (46) 27 (39) 11 (16) 15 (21)
0.79
0.07
6 (9) 10 (15) 7 (10) 38 (56)
10 (14) 24 (34) 12 (17) 7 (10)
0.32 0.01 0.24 b0.01
19 (28) 22 (32) 23 (34)
21 (30) 17 (24) 10 (14)
0.79 0.29 b0.01
n (%) Gender, male n (%) Age, median (IQR)* Race, n (%) White African American Other Hispanic ethnicity, n (%) Medical History Diabetes mellitus Hypertension Pulmonary disease Psychiatric illness Social History Alcohol use Tobacco use Illicit drug use *
IQR = Interquartile Range.
Please cite this article as: L.E. Lamparter, M.A. Rech and T.M. Nguyen, Homeless patients tend to have greater psychiatric needs when presenting to the emergency department..., American Journal of Emergency Medicine, https://doi.org/10.1016/j.ajem.2019.10.012
L.E. Lamparter et al. / American Journal of Emergency Medicine xxx (xxxx) xxx Table 2 Information Obtained from ED Visit. Chief Complaint
Psychiatric evaluation Abdominal/Back pain Suicide Chest pain Intoxication Neurologic (e.g. stroke, seizure) Injury Other Insurance Medicare or Medicaid Private insurance Uninsured Other insurance Arrival to and Time spent in ED Ambulance Use Length of Stay, median hours (IQR) * Disposition Discharged Transferred to psychiatric facility Against Medical Advice Other disposition ED Visits, median (IQR)* Prior ED visits Readmissions in 2017 Total visits to date *
Homeless (n = 68)
Non-Homeless (n = 70)
p value
n (%) 23 (34) 8 (12) 6 (9) 6 (9) 6 (9) 3 (5) 2 (3) 54 (79)
3 (4) 6 (9) 1 (1) 5 (7) 3 (4) 7 (10) 7 (10) 14 (21)
b0.01 0.53 0.06 0.71 0.28 0.33 0.17 b0.01
46 (71) 1 (2) 12 (19) 6 (9)
35 (51) 20 (29) 9 (13) 5 (7)
b0.01
31 (46) 8 (4–13)
11 (16) 5 (3–7)
b0.01 0.01
33 (50) 27 (40) 1 (2) 7 (10)
64 (93) 1 (2) 2 (3) 2 (3)
b0.01
0 (0–1) 0 (0–1) 2 (1–4)
1 (0–1) 0 (0–1) 3 (1–5)
0.01 0.75 0.08
IQR = Interquartile Range.
illness such as bipolar, schizophrenia, anxiety or depression (56% vs. 10% non-homeless, p b 0.01). The high prevalence of psychiatric diagnoses in this homeless population was consistent with the findings of other studies, which have correlated homelessness with psychiatric diagnoses leading to ED visits [5–7]. A recent survey of homeless individuals in Chicago found only 28% of the unsheltered homeless population are receiving mental health care, and another study found that 72% of the chronically homeless have neurocognitive deficits secondary to severe mental illness, post-traumatic stress disorder, seizures, or traumatic brain injury [3,8]. As more than half of the cohort in this study had a history of a psychiatric diagnoses, this implies a need for further study and understanding of the psychiatric care available to this population. Homeless patients also face many barriers to obtaining primary care, including fear of stereotypes and discrimination, difficulty presenting proof of insurance, transportation issues, and scheduling problems [9]. When trying to find food and shelter, making time to prioritize health care appointments can be difficult [9]. Thus, homeless patients are common visitors to EDs across the country, and there is a stereotype that they visit more frequently and more repeatedly than their nonhomeless counterparts [2,5]. This has generally been attributed to differences in medical needs, lack of insurance, or acute intoxication with alcohol and drugs [6]. However, in recent years the former stereotype of an alcoholic homeless patient has been replaced by the mentally ill homeless patient [4]. Our study of ED utilization by homeless patients confirmed that there has been a shift away from old stereotypes. This study shows the homeless population utilizes the ED for different reasons than non-homeless patients, but they are not necessarily using the ED more frequently. Another common stereotype is that homeless patients require greater utilization of resources [10]. In our study, we found that homeless patients were significantly more likely to use ambulance as a means of arrival to the ED and had a median two and a half hour longer length of stay in the ED. A study from Canada compared 3081 homeless patients to 90,345 non-homeless patients and found that homeless patients who were seen and subsequently admitted for psychiatric concerns had hospital visits that were significantly more expensive than patients with other, non-psychiatric concerns [10]. This suggests
3
the importance of providing more specific resources such as increased access to outpatient follow up or medical transportation for the homeless population, which could help reduce overall health care costs and ED length of stay. One of the more interesting findings of this study was that the majority of patients in this study were insured. Homeless patients were more likely to have Medicare or Medicaid (71% vs. 51%, p b 0.01), while non-homless were more likely to have private insurance (2% vs. 29% p b 0.01). There was a significant difference (p b 0.01) between the percentage of uninsured patients among the homeless (19%) and non-homeless (13%). These findings are likely related to recent legislation changes have allowed people to register for Medicare or Medicaid with a social security number instead of an address. The U.S. Department of Health and Human Services has made possible a variety of eligibility pathways for homeless people to qualify for Medicaid [11]. A health care utilization project from 2017 found that homeless adult visitors of teaching hospital EDs were more likely to have Medicare insurance while those presenting to community hospitals were more likely to be uninsured [6]. Despite those who are able to obtain insurance, homeless patients still come to the ED for their acute and chronic medical needs due to the difficulties of obtaining primary care as addressed above [9]. In terms of demographics, the homeless patients in this study were more likely to be men, while the sample of non-homeless patients were more likely to be women (76% homeless vs. 24% non-homeless, p b 0.01). This is consistent with local data from the 2017 point in time survey of Chicago homeless which found that 83% of all Chicago’s homeless were male [3]. This difference in gender has the potential to influence the reasons a particular patient might come to the ED and thus be a significant limiting factor. It would be interesting to compare a non-homeless cohort with more men to the predominantly male homeless population. This study is limited due to its retrospective design and potential for biases in identification and selection of patients. Documentation may have limited the identification of homeless individuals. Although there was not a unified method to assess the housing status of patients, we attempted to cast a wide net by using multiple methods to search for homeless individuals, including ICD-10 codes, status as “indigent”, or lack of an address and replacement of the patient’s address with the hospital address. This suggests the need for a more unified operational definition of homelessness and a screening tool to identify this patient population. Also, it is common for homeless patients to list the address of a shelter, a friend or even a fictitious address when asked to provide their own address upon hospital admission, so the lack of an address might not be a consistent means of identifying this population [12]. This difficulty in identification of homeless patients could have also contributed to the small number of data. A wider study over more than one year could have increased our sample size of 68 and allowed us to see an increase in the significance of our data; however, for this particular study, we were interested in studying patients from one year at one particular hospital. 5. Conclusion Homeless patients were significantly more likely to present to the ED with a chief complaint related to a psychiatric condition than nonhomeless patients. Homeless patients were also significantly more likely to utilize ED resources, such as more frequent ambulance transport and longer ED stays, in a given visit than the non-homeless patients. Contrary to common misconceptions, homeless patients did not have more frequent ED visits than the non-homeless patients. However, they did have greater incidence of illicit drug use. The results of this study highlights the importance of consistent mental health care in this population. Both training the ED staff to care for patients with significant mental health conditions and setting up better outpatient mental health follow up are important considerations to address to improve
Please cite this article as: L.E. Lamparter, M.A. Rech and T.M. Nguyen, Homeless patients tend to have greater psychiatric needs when presenting to the emergency department..., American Journal of Emergency Medicine, https://doi.org/10.1016/j.ajem.2019.10.012
4
L.E. Lamparter et al. / American Journal of Emergency Medicine xxx (xxxx) xxx
the quality of care for the homeless patients who utilize the ED frequently for psychiatric concerns. Declaration of Competing Interest There are no conflicts of interest in this study. References [1] The State of Homeless in America. Retrieved from https://endhomelessness.org/ homelessness-in-america/homelessness-statistics/state-of-homelessness-reportlegacy/. (2017). Last accessed on 9/9/19. [2] Pearson DA, Bruggman AR, Haukoos JS. Out-of-hospital and emergency department utilization by adult homeless patients. Ann Emerg Med 2007;50(6):646–52. https:// doi.org/10.1016/j.annemergmed.2007.07.015. [3] City of Chicago 2017 Homeless Point-in-Time Count and Survey Report (pp. 1-17, Rep.). (2017). Chicago, IL: Voorhees Center for Neighborhood and Community Improvement, University of Illinois at Chicago. [4] Institute of Medicine (US) Committee on Health Care for Homeless People. Homelessness, Health, and Human Needs. Washington (DC): National Academies Press (US); 1988. 3, Health Problems of Homeless People.
[5] Sun R, Karaca Z, Wong HS. Characteristics of homeless individuals using emergency department services in 2014. Agency Health Res Qual 2017: 1–13. [6] Lin W, Bharel M, Zhang J, et al. Frequent emergency. department visits and hospitalizations among homeless people with medicaid: implications for medicaid expansion. Am Pub Health Assoc 2015;105(S5):S716–22. [7] Kushel MB, Perry S, Bangsberg D, et al. emergency department use among the homeless and marginally housed: results from a community-based study. Am J Public Health 2002;92(5):778–84. [8] Stergiopoulos V, Cusi A, Bekele T, et al. Neurocognitive impairment in a large sample of homeless adults with mental illness. Acta Psychiatr Scand 2015 Apr;131(4): 256–68. [9] Health quality ontario interventions to improve access to primary care for people who are homeless: A Systematic Review. Ont Health Technol Assess Ser 2016;16: 1–50. [10] Hwang SW, Weaver J, Aubry T, et al. Hospital costs and length of stay among homeless patients admitted to medical, surgical, and psychiatric services. Med Care 2011 Apr;49(4):350–4. [11] Condensed Version of a Primer on How to Use Medicaid to Assist Persons Who Are Homeless to Access Medical, Behavioral Health, and Support Services. https://www. hhs.gov/programs/social-services/homelessness/research/how-to-use-medicaid-toassist-homeless-persons/index.html# (1-6). [12] Feldman BJ, Calogero CG, Elsayed KS, et al. Prevalence of Homelessness in the Emergency Department Setting. West JEM 2017;18(3):366–72.
Please cite this article as: L.E. Lamparter, M.A. Rech and T.M. Nguyen, Homeless patients tend to have greater psychiatric needs when presenting to the emergency department..., American Journal of Emergency Medicine, https://doi.org/10.1016/j.ajem.2019.10.012
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Homeless patients tend to have greater psychiatric needs when presenting to the emergency department,☆,☆☆ Lauren E. Lamparter a,⁎, Megan A. Rech, PharmD b,c, Theresa M. Nguyen, MD FACEP c a b c
Medical Student, MS3, Loyola University Chicago Stritch School of Medicine, 2160 S 1st Ave, Maywood, IL 60513, United States Loyola University Medical Center, Department of Pharmacy, 2160 S 1st Ave, Maywood, IL 60513, United States Loyola University Medical Center, Department of Emergency Medicine, 2160 S 1st Ave, Maywood, IL 60513, United States
a r t i c l e
i n f o
Article history: Received 15 September 2019 Received in revised form 9 October 2019 Accepted 12 October 2019 Available online xxxx Keywords: Homeless Psychiatric Emergency department Chief complaint
a b s t r a c t Introduction: Homeless patients tend to visit Emergency Departments (EDs) more frequently than the nonhomeless population. The goal of this study was to assess differences in chief complaint, medical conditions, and disposition between homeless patients compared to non-homeless patients presenting to an urban ED. Methods: This was a retrospective cohort of homeless patients ages ≥18 years compared to non-homeless controls from January 1, 2017 to December 31, 2017. Exclusion criteria were as follows: direct admission to hospital floor, repeat visits, or leaving without being seen. The primary endpoint of this study was to assess differences in chief complaint of homeless versus non-homeless patients upon presentation to the ED. Our secondary endpoints included differences in ED utilization between the two groups, in terms of length of stay, ambulance use, diagnosis, and disposition. Results: Homeless patients were more likely present to the ED for a psychiatric evaluation (homeless group 34% vs. non-homeless group 4%, p b 0.01) and have a history of a psychiatric diagnosis (56% vs. 10%, p b 0.01) compared to non-homeless controls. Homeless patients also tended to require more ambulance transport (46% vs. 16%, p b 0.01). More homeless patients were transferred to a psychiatric facility (40% vs. 1%, p b 0.01), while the majority of non-homeless patients were discharged home (50% vs. 93%, p b 0.01). Conclusion: This study found that homeless patients had a significantly higher association with psychiatric diagnoses and greater ED utilization than non-homeless. This suggests the importance of increased access to consistent psychiatric care and follow up within the homeless population. © 2019 Elsevier Inc. All rights reserved.
1. Introduction The National Alliance to End Homelessness reported that on any given night in January of 2017, approximately 553,742 people experienced homelessness across the country [1]. A person is classified as homeless if he or she does not have a permanent or temporary address of residency [2]. A survey performed at an urban medical center in 2017 reported that the number of homeless people living exposed without shelter has increased from prior years [3]. Living without shelter places individuals at a greater risk for many acute medical problems, such as exposure injuries or sexual assault, as well as exacerbation of chronic medical problems, including malnutrition, liver disease, dental disease, ☆ This research was presented at the 2019 AAEM/RSA & West JEM Population Health Research Competition in Los Vegas, NV on March 11, 2019. ☆☆ This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The data presented and analyzed in this study is available upon reasonable request. ⁎ Corresponding author. E-mail addresses: [email protected] (L.E. Lamparter), [email protected] (M.A. Rech), [email protected] (T.M. Nguyen).
or psychiatric disorders, which could be otherwise manageable [4]. In addition, a previous study of homeless utilization of the Emergency Department (ED) found that a majority of homeless patient visits to the ED were directly related to excessive alcohol use [2]. As a result of their increased exposures, chronic illnesses, and substance abuse, homeless adults tend to visit EDs more often and more frequently than the nonhomeless general population [2]. It is relevant then, for emergency physicians to understand the specific reasons why homeless patients seek emergency care. The goal of this study was to assess homeless ED utilization by presenting chief complaint, use of ED resources, and final disposition compared to non-homeless controls in order to better understand and potentially prepare to address the needs of this population. 2. Methods This was a retrospective analysis of homeless ED patients compared to non-homeless ED patients from January 1, 2017 – December 31, 2017. Institutional Review Board approval was obtained prior to conducting this study. This study was conducted at an urban, level 1
https://doi.org/10.1016/j.ajem.2019.10.012 0735-6757/© 2019 Elsevier Inc. All rights reserved.
Please cite this article as: L.E. Lamparter, M.A. Rech and T.M. Nguyen, Homeless patients tend to have greater psychiatric needs when presenting to the emergency department..., American Journal of Emergency Medicine, https://doi.org/10.1016/j.ajem.2019.10.012
2
L.E. Lamparter et al. / American Journal of Emergency Medicine xxx (xxxx) xxx
trauma and tertiary referral center in Illinois, where the ED serves approximately 50,000 patients per year. Patients were identified via their demographic status as “indigent” or an International Classification of Disease, tenth revision (ICD-10) codes for homelessness and inadequate housing. The ICD-10 codes included the following: Z59.0 homelessness, Z59.1 inadequate housing, Z59.3 problems related to living in residential institution, Z59.5 extreme poverty, and Z59.8 and Z59.9 problem related to housing and economic circumstances, unspecified. Since there is fluidity in the circumstances of the homeless population, (for example, they might lack a permanent address but not live full-time on the streets), a range of ICD-10 codes were needed to capture the variations within this population. In the event that an ICD-10 code for homelessness was not assigned or the patient was not classified as “indigent,” homeless patients were also identified by searching for the study institution’s address in the patient address field. All homeless patients identified by these methods were initially included in the study. These methods of identifying homeless patients were chosen based on a previous study which had used a patient’s lack of address to identify their homeless population [2]. In addition, the standard procedure at the study institution is to list the hospital’s address if a patient is unable to provide an address upon admission. A systematic sample of non-homeless patients were identified by including every tenth ED visit beginning on January 1, 2017. Patients in both groups were excluded if they were under 18 years of age, were directly admitted to the hospital, or left without being seen. If the patient was seen in the ED multiple times during the study period, only the first visit was included. For each patient visit, the following information was recorded: date of visit, mode of transportation to the ED, chief complaint, blood alcohol and urinary toxicology screening results, diagnosis upon discharge, number of prior ED visits, and final disposition. Demographic information, past medical history (including psychiatric or substance abuse history), and insurance status were also collected. Use of hospital resources by homeless patients was assessed by ambulance use, ED length of stay, insurance status, labs ordered, disposition, and repeat visits. Baseline characteristics were described using mean, standard deviation, median, interquartile range (IQR) and percentages. Shapiro Wilks test was used to assess normality of continuous data. Continuous parametric variables were compared using a t-test. Continuous nonparametric data were analyzed utilizing the Mann-Whitney U test. Chi-square test or Fischer's exact test were used to compare categorical variables. A p value of b 0.05 was considered significant.
3. Results In 2017, there were approximately 46,000 ED visits, of which 208 were identified as homeless by ICD-10 diagnostic codes, lack of an address, or demographic status as “indigent.” A systematic sample of 178 non-homeless patients were obtained as a control from all patients who visited in 2017. After exclusion criteria were applied, 68 homeless patients were compared to 70 control patients (Fig. 1). Baseline demographics are displayed in Table 1. Homeless patients were more likely be men (76% homeless group vs. 24% non-homeless group, p b 0.01) and to have an underlying psychiatric diagnosis (55% vs. 10%, p b 0.01) (Table 1). Homeless patients were significantly more likely to present to the ED for a psychiatric evaluation (34% vs. 4%, p b 0.01) (Table 2). Interestingly, only three percent of homeless patients presented with a chief complaint related to an injury while 10% of non-homeless presented after an injury or fall, though this was not a significant difference (p = 0.17) (Table 2). Homeless patients reported greater incidence of illicit drug use (34% vs. 14%, p b 0.01) (Table 1). Though homeless patients were more likely to report illicit drug use, only 9% of homeless presented with acute intoxication, which was not a significant difference from the 4% of non-homeless patients who presented acutely intoxicated (p = 0.28) (Table 2). There
Total Charts Reviewed (n= 386) Homeless and Non-homeless
Exclusion Criteria: • • • •
Final Homeless (n= 68)
Age < 18 (n=48) Direct admit to the floor (n=71) Left without being seen (n= 16) Subsequent visits in 2017 (n=113)
Final Non-homeless (n= 70) Fig. 1. Study flow diagram.
was no difference in the incidence of alcohol abuse between groups (Table 1). Homeless patients were more likely to present to the ED by ambulance transport (46% vs. 16%, p b 0.01) and had longer ED length of stays (median 8 h, range 4–13 vs. median 5 h, range 3–7, p = 0.01) (Table 2). The majority of homeless patients were insured by Medicare or Medicaid (71% vs. 51%, p b 0.01), and while only 19% of the homeless patients were uninsured, they were more likely to be uninsured than the non-homeless patients (19% vs 13%, p b 0.01) (Table 2). In terms of final disposition, homeless patients were significantly less likely to be discharged, compared to non-homeless patients, the majority of whom were discharged home (50% vs. 93% p= b0.01) (Table 2). Of the patients who were not discharged, 40% of the homeless were transferred to psychiatric facilities, compared to 2% of the non-homeless population (p b 0.01) (Table 2). There was no difference between the two groups in regards to total number of ED visits or readmissions. 4. Discussion This study demonstrated a significant increase in homeless patient’s chief complaints related to psychiatric illness compared to the nonhomeless controls, and found that homeless paitents had significantly more psychiatric conditions, illicit drug use, and more frequent use of ED resources such as ambulance transport. Furthermore, homeless patients were significantly more likely to have a history of a psychiatric Table 1 Baseline demographics of homeless compared to non-homeless patients. Demographics
Homeless (n = 68)
Non-Homeless (n = 70)
p value
52 (76) 40 (28–55)
16 (24) 40 (28–55)
b0.01 0.81
33 (49) 27 (40) 8 (12) 7 (10)
32 (46) 27 (39) 11 (16) 15 (21)
0.79
0.07
6 (9) 10 (15) 7 (10) 38 (56)
10 (14) 24 (34) 12 (17) 7 (10)
0.32 0.01 0.24 b0.01
19 (28) 22 (32) 23 (34)
21 (30) 17 (24) 10 (14)
0.79 0.29 b0.01
n (%) Gender, male n (%) Age, median (IQR)* Race, n (%) White African American Other Hispanic ethnicity, n (%) Medical History Diabetes mellitus Hypertension Pulmonary disease Psychiatric illness Social History Alcohol use Tobacco use Illicit drug use *
IQR = Interquartile Range.
Please cite this article as: L.E. Lamparter, M.A. Rech and T.M. Nguyen, Homeless patients tend to have greater psychiatric needs when presenting to the emergency department..., American Journal of Emergency Medicine, https://doi.org/10.1016/j.ajem.2019.10.012
L.E. Lamparter et al. / American Journal of Emergency Medicine xxx (xxxx) xxx Table 2 Information Obtained from ED Visit. Chief Complaint
Psychiatric evaluation Abdominal/Back pain Suicide Chest pain Intoxication Neurologic (e.g. stroke, seizure) Injury Other Insurance Medicare or Medicaid Private insurance Uninsured Other insurance Arrival to and Time spent in ED Ambulance Use Length of Stay, median hours (IQR) * Disposition Discharged Transferred to psychiatric facility Against Medical Advice Other disposition ED Visits, median (IQR)* Prior ED visits Readmissions in 2017 Total visits to date *
Homeless (n = 68)
Non-Homeless (n = 70)
p value
n (%) 23 (34) 8 (12) 6 (9) 6 (9) 6 (9) 3 (5) 2 (3) 54 (79)
3 (4) 6 (9) 1 (1) 5 (7) 3 (4) 7 (10) 7 (10) 14 (21)
b0.01 0.53 0.06 0.71 0.28 0.33 0.17 b0.01
46 (71) 1 (2) 12 (19) 6 (9)
35 (51) 20 (29) 9 (13) 5 (7)
b0.01
31 (46) 8 (4–13)
11 (16) 5 (3–7)
b0.01 0.01
33 (50) 27 (40) 1 (2) 7 (10)
64 (93) 1 (2) 2 (3) 2 (3)
b0.01
0 (0–1) 0 (0–1) 2 (1–4)
1 (0–1) 0 (0–1) 3 (1–5)
0.01 0.75 0.08
IQR = Interquartile Range.
illness such as bipolar, schizophrenia, anxiety or depression (56% vs. 10% non-homeless, p b 0.01). The high prevalence of psychiatric diagnoses in this homeless population was consistent with the findings of other studies, which have correlated homelessness with psychiatric diagnoses leading to ED visits [5–7]. A recent survey of homeless individuals in Chicago found only 28% of the unsheltered homeless population are receiving mental health care, and another study found that 72% of the chronically homeless have neurocognitive deficits secondary to severe mental illness, post-traumatic stress disorder, seizures, or traumatic brain injury [3,8]. As more than half of the cohort in this study had a history of a psychiatric diagnoses, this implies a need for further study and understanding of the psychiatric care available to this population. Homeless patients also face many barriers to obtaining primary care, including fear of stereotypes and discrimination, difficulty presenting proof of insurance, transportation issues, and scheduling problems [9]. When trying to find food and shelter, making time to prioritize health care appointments can be difficult [9]. Thus, homeless patients are common visitors to EDs across the country, and there is a stereotype that they visit more frequently and more repeatedly than their nonhomeless counterparts [2,5]. This has generally been attributed to differences in medical needs, lack of insurance, or acute intoxication with alcohol and drugs [6]. However, in recent years the former stereotype of an alcoholic homeless patient has been replaced by the mentally ill homeless patient [4]. Our study of ED utilization by homeless patients confirmed that there has been a shift away from old stereotypes. This study shows the homeless population utilizes the ED for different reasons than non-homeless patients, but they are not necessarily using the ED more frequently. Another common stereotype is that homeless patients require greater utilization of resources [10]. In our study, we found that homeless patients were significantly more likely to use ambulance as a means of arrival to the ED and had a median two and a half hour longer length of stay in the ED. A study from Canada compared 3081 homeless patients to 90,345 non-homeless patients and found that homeless patients who were seen and subsequently admitted for psychiatric concerns had hospital visits that were significantly more expensive than patients with other, non-psychiatric concerns [10]. This suggests
3
the importance of providing more specific resources such as increased access to outpatient follow up or medical transportation for the homeless population, which could help reduce overall health care costs and ED length of stay. One of the more interesting findings of this study was that the majority of patients in this study were insured. Homeless patients were more likely to have Medicare or Medicaid (71% vs. 51%, p b 0.01), while non-homless were more likely to have private insurance (2% vs. 29% p b 0.01). There was a significant difference (p b 0.01) between the percentage of uninsured patients among the homeless (19%) and non-homeless (13%). These findings are likely related to recent legislation changes have allowed people to register for Medicare or Medicaid with a social security number instead of an address. The U.S. Department of Health and Human Services has made possible a variety of eligibility pathways for homeless people to qualify for Medicaid [11]. A health care utilization project from 2017 found that homeless adult visitors of teaching hospital EDs were more likely to have Medicare insurance while those presenting to community hospitals were more likely to be uninsured [6]. Despite those who are able to obtain insurance, homeless patients still come to the ED for their acute and chronic medical needs due to the difficulties of obtaining primary care as addressed above [9]. In terms of demographics, the homeless patients in this study were more likely to be men, while the sample of non-homeless patients were more likely to be women (76% homeless vs. 24% non-homeless, p b 0.01). This is consistent with local data from the 2017 point in time survey of Chicago homeless which found that 83% of all Chicago’s homeless were male [3]. This difference in gender has the potential to influence the reasons a particular patient might come to the ED and thus be a significant limiting factor. It would be interesting to compare a non-homeless cohort with more men to the predominantly male homeless population. This study is limited due to its retrospective design and potential for biases in identification and selection of patients. Documentation may have limited the identification of homeless individuals. Although there was not a unified method to assess the housing status of patients, we attempted to cast a wide net by using multiple methods to search for homeless individuals, including ICD-10 codes, status as “indigent”, or lack of an address and replacement of the patient’s address with the hospital address. This suggests the need for a more unified operational definition of homelessness and a screening tool to identify this patient population. Also, it is common for homeless patients to list the address of a shelter, a friend or even a fictitious address when asked to provide their own address upon hospital admission, so the lack of an address might not be a consistent means of identifying this population [12]. This difficulty in identification of homeless patients could have also contributed to the small number of data. A wider study over more than one year could have increased our sample size of 68 and allowed us to see an increase in the significance of our data; however, for this particular study, we were interested in studying patients from one year at one particular hospital. 5. Conclusion Homeless patients were significantly more likely to present to the ED with a chief complaint related to a psychiatric condition than nonhomeless patients. Homeless patients were also significantly more likely to utilize ED resources, such as more frequent ambulance transport and longer ED stays, in a given visit than the non-homeless patients. Contrary to common misconceptions, homeless patients did not have more frequent ED visits than the non-homeless patients. However, they did have greater incidence of illicit drug use. The results of this study highlights the importance of consistent mental health care in this population. Both training the ED staff to care for patients with significant mental health conditions and setting up better outpatient mental health follow up are important considerations to address to improve
Please cite this article as: L.E. Lamparter, M.A. Rech and T.M. Nguyen, Homeless patients tend to have greater psychiatric needs when presenting to the emergency department..., American Journal of Emergency Medicine, https://doi.org/10.1016/j.ajem.2019.10.012
4
L.E. Lamparter et al. / American Journal of Emergency Medicine xxx (xxxx) xxx
the quality of care for the homeless patients who utilize the ED frequently for psychiatric concerns. Declaration of Competing Interest There are no conflicts of interest in this study. References [1] The State of Homeless in America. Retrieved from https://endhomelessness.org/ homelessness-in-america/homelessness-statistics/state-of-homelessness-reportlegacy/. (2017). Last accessed on 9/9/19. [2] Pearson DA, Bruggman AR, Haukoos JS. Out-of-hospital and emergency department utilization by adult homeless patients. Ann Emerg Med 2007;50(6):646–52. https:// doi.org/10.1016/j.annemergmed.2007.07.015. [3] City of Chicago 2017 Homeless Point-in-Time Count and Survey Report (pp. 1-17, Rep.). (2017). Chicago, IL: Voorhees Center for Neighborhood and Community Improvement, University of Illinois at Chicago. [4] Institute of Medicine (US) Committee on Health Care for Homeless People. Homelessness, Health, and Human Needs. Washington (DC): National Academies Press (US); 1988. 3, Health Problems of Homeless People.
[5] Sun R, Karaca Z, Wong HS. Characteristics of homeless individuals using emergency department services in 2014. Agency Health Res Qual 2017: 1–13. [6] Lin W, Bharel M, Zhang J, et al. Frequent emergency. department visits and hospitalizations among homeless people with medicaid: implications for medicaid expansion. Am Pub Health Assoc 2015;105(S5):S716–22. [7] Kushel MB, Perry S, Bangsberg D, et al. emergency department use among the homeless and marginally housed: results from a community-based study. Am J Public Health 2002;92(5):778–84. [8] Stergiopoulos V, Cusi A, Bekele T, et al. Neurocognitive impairment in a large sample of homeless adults with mental illness. Acta Psychiatr Scand 2015 Apr;131(4): 256–68. [9] Health quality ontario interventions to improve access to primary care for people who are homeless: A Systematic Review. Ont Health Technol Assess Ser 2016;16: 1–50. [10] Hwang SW, Weaver J, Aubry T, et al. Hospital costs and length of stay among homeless patients admitted to medical, surgical, and psychiatric services. Med Care 2011 Apr;49(4):350–4. [11] Condensed Version of a Primer on How to Use Medicaid to Assist Persons Who Are Homeless to Access Medical, Behavioral Health, and Support Services. https://www. hhs.gov/programs/social-services/homelessness/research/how-to-use-medicaid-toassist-homeless-persons/index.html# (1-6). [12] Feldman BJ, Calogero CG, Elsayed KS, et al. Prevalence of Homelessness in the Emergency Department Setting. West JEM 2017;18(3):366–72.
Please cite this article as: L.E. Lamparter, M.A. Rech and T.M. Nguyen, Homeless patients tend to have greater psychiatric needs when presenting to the emergency department..., American Journal of Emergency Medicine, https://doi.org/10.1016/j.ajem.2019.10.012