Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Apr 22.
Published in final edited form as: J Psychiatr Res. 2026 Mar 25;198:175–181. doi: 10.1016/j.jpsychires.2026.03.040

Who’s Getting Care? Acute Treatment Utilization Following an Emergency Department Visit Among Adults with Major Psychiatric Disorders and Suicidal Thoughts and Behaviors

Ana Rabasco a,b, Jihoon Choi a,b, Ilana Ladis a,b, Sarah A Arias a,b, Lauren M Weinstock a, Ivan Miller a,b, Edwin D Boudreaux c, Carlos A Camargo Jr d, Brandon A Gaudiano a,b
PMCID: PMC13098320  NIHMSID: NIHMS2160795  PMID: 41905117

Abstract

Introduction:

People with major psychiatric disorders (e.g., major depressive disorder [MDD], bipolar disorder [BD], and schizophrenia-spectrum disorders [SSD]) utilize acute treatment at high rates, while struggling to engage in outpatient care. There is limited longitudinal research on differences in treatment utilization among people with different psychiatric diagnoses and concurrent suicidal thoughts and/or behaviors (STBs). The current study aimed to examine patterns of and factors associated with treatment utilization (outpatient treatment, psychiatric inpatient hospitalization, emergency department [ED visit]) among individuals with major psychiatric disorders and STBs presenting to the ED.

Methods:

Adults with MDD (n = 347), BD (n = 298), and SSD (n = 100) with active STBs were recruited from eight EDs across the United States as part of a larger study on treatment for STBs. Chart reviews and participant assessments were conducted during the 52-weeks following participant ED discharge.

Results:

Participants with BD and SSD had higher rates of, and shorter time to, inpatient hospitalization and ED visit over follow-up than people with MDD. There were no differences in rates of outpatient treatment between diagnostic groups. Lack of employment and outpatient mental health treatment utilization were associated with inpatient hospitalization over follow-up. Lack of employment, a substance use disorder, and chronic pain were associated with ED visit over follow-up.

Conclusions:

Results show that treatment utilization differs by major psychiatric diagnosis. These findings suggest that people with BD and SSD may benefit from additional supports, including vocational services, during and following ED visits to minimize future acute treatment utilization.

Keywords: psychiatric disorders, treatment, hospitalization, suicide, emergency department, bipolar disorder, schizophrenia

Introduction

People with major psychiatric disorders – including major depressive disorder (MDD), bipolar disorder (BD), and schizophrenia-spectrum disorders (SSD) – tend to have high rates of emergency department (ED) and inpatient treatment utilization (Irmiter et al., 2007; Ronaldson et al., 2020), while also struggling to engage in outpatient care (McAlpine & Mechanic, 2000). People with major psychiatric disorders have severe and chronic symptoms that impair their functioning, leading to heightened risks for psychiatric crises and the need for acute care (Gonçalves-Pinho et al., 2022; Kessler & Lev-Ran, 2019). These symptoms and related impairments are distinct across diagnostic groups, with MDD characterized by persistent low mood, reduced energy and motivation, poor concentration, and social withdrawal; BD characterized by episodic mood instability, impulsivity, disrupted sleep, and impaired judgment during manic/hypomanic episodes; and SSD characterized by delusions, hallucinations, disorganized speech and behavior, and negative symptoms (American Psychiatric Association, 2022).

Research has shown that a major psychiatric disorder diagnosis and active suicidal thoughts and/or behaviors (STBs) are both risk factors for ED visits (Azar et al., 2020) and hospitalization (Berardelli et al., 2022; Lorine et al., 2015). Approximately 50% of people who frequently visit the ED report having poor mental health (Hunt et al., 2006). Additionally, severity of mental health diagnosis, as well as medical and substance use comorbidity, are associated with increased ED visits (Brennan et al., 2014; Niedzwiecki et al., 2018). Better understanding ED readmissions among people with mental health diagnoses could mitigate the rising rate of ED visits (Hooker et al., 2019) and associated costs (Pickens et al., 2022). Rehospitalization is also a significant concern among individuals with major psychiatric disorders. One study of 35,527 individuals with either SSD or BD discharging from Veterans Affairs inpatient psychiatric settings found that, within seven years, 86% were re-hospitalized (Irmiter et al., 2007). Rates of rehospitalization among individuals with MDD are also substantial, with studies showing rehospitalization rates in MDD of 26.8% over a two-year follow-up period (Cearns et al., 2019) and 22.7% over a median 4.5 year follow up period (Innes et al., 2015). Among people with SSD, suicide attempts and other forms of self-harm have been found to increase the risk for rehospitalization (Sullivan et al., 1997). Other risk factors for psychiatric hospital readmission among people with major psychiatric disorders include previous hospitalizations (Donisi et al., 2016; Hariman et al., 2020; Velelekou et al., 2022; Yan et al., 2019), lower education level (Suzuki et al., 2003), substance misuse (Hariman et al., 2020; Yan et al., 2019), and lack of social support (Velelekou et al., 2022).

There is a lack of research on risk factors for both inpatient psychiatric rehospitalization and recurrent ED visits among adults with major psychiatric disorders experiencing concurrent STBs. Furthermore, few studies have compared whether there are differential rates of hospitalization and ED visits between individuals with MDD, BD, and SSD, despite previous research showing differences in hospitalization patterns and healthcare contacts between adults with these diagnoses (Berardelli et al., 2022; Frahm et al., 2019). Clarifying these differences may help healthcare professionals better target the needs of each diagnostic subgroup and improve resource allocation in the ED setting. This is particularly relevant, as EDs are a common point of contact for people at high risk for suicide (Ahmedani et al., 2019; Da Cruz et al., 2011) and for people with major psychiatric disorders (Ceniti et al., 2020; Sirotich et al., 2016) due to a combination of acute symptom exacerbations, suicidal crises, co-occurring substance use, medical comorbidities, and limited access to timely outpatient care. For many, EDs represent the most accessible setting during periods of heightened risk, particularly when outpatient services are unavailable. These shared vulnerabilities contribute to recurrent ED utilization and underscore the ED’s role as a critical safety net for this population (Ahmedani et al., 2019; Sirotich et al., 2016). Additionally, STB rates are elevated in the weeks and months following discharge from both ED and inpatient psychiatric hospitalization settings (Choi et al., 2012; Forte et al., 2019), especially among persons with major psychiatric disorders (Olfson et al., 2016). Although acute care settings are an essential service for people with major psychiatric disorders and can be the appropriate clinical disposition depending on patient presentation, it would be worthwhile to find ways of reducing unnecessary ED visits and hospitalizations. This is due to the potential for iatrogenic patient experiences (Frueh et al., 2005; Ward-Ciesielski & Rizvi, 2021), as well as financial burdens (Stensland et al., 2012; Tibirna et al., 2024), from avoidable acute care treatment. It is imperative to optimize care pathways more broadly among those with major psychiatric disorders.

Outpatient treatment utilization is another factor closely tied to rehospitalization among people with major psychiatric disorders. Research supports that outpatient treatment engagement is associated with reduced rates of hospitalization for this population (Cook et al., 2021; Marcus et al., 2017). Disengagement from outpatient treatment, on the other hand, has been associated with a host of negative outcomes, including more severe psychiatric symptoms and social impairment (Killapsy et al., 2000). One study of people with psychosis found that 45% of participants did not attend a single outpatient treatment session in the six months following inpatient hospitalization (Myers et al., 2017). To prevent suicide and improve outcomes among people with major psychiatric disorders and STBs, it is essential to better understand potential diagnostic differences associated with outpatient treatment (dis)engagement following ED discharge in this exceptionally high-risk subset of patients.

Study Aims

This study addressed a gap in the literature to evaluate patterns of and baseline factors associated with treatment utilization among individuals with MDD, BD, and SSD presenting to the ED who are experiencing concurrent active STBs. By identifying the baseline factors associated with treatment utilization among the patients at highest risk for admission within this already high-risk group, limited ED resources can be best leveraged to mitigate future hospitalizations among this population. This study was an exploratory secondary analysis conducted with data from a larger trial (Boudreaux et al., 2013; Miller et al., 2017).

Aim 1: 1a. Examine and compare rates of index psychiatric inpatient hospitalization, along with 12-month follow-up rates psychiatric inpatient hospitalization, ED visit, and outpatient treatment for participants with MDD, BD, and SSD. 1b. Compare time to psychiatric hospitalization and ED visit over 12-month follow-up among participants with MDD, BD, and SSD.

Aim 2: Explore relevant demographic (sex, age, education, employment, marital status, and whether an individual lived alone or not) and baseline clinical (depression, substance use disorder [SUD], chronic pain, past-week suicide attempt, outpatient mental health treatment utilization) variables associated with psychiatric inpatient hospitalization and ED visit over 12-month follow-up for participants with MDD, BD, and SSD.

Method

Procedure

This paper is a secondary analysis of the Emergency Department Safety Assessment and Follow-up Evaluation (ED-SAFE) study, which tested the effectiveness of universal suicide screening and a suicide prevention intervention in adults presenting to EDs across the United States (clinical trials registry number: NCT01150994; Boudreaux et al., 2013; Miller et al., 2017). Participants were recruited from eight EDs in three sequential phases. In Phase 1, patients received treatment as usual (TAU). During Phase 2, universal screening was implemented at each recruitment site. Lastly, Phase 3 introduced the Coping Long Term with Active Suicide (CLASP) study intervention in addition to universal screening. For more information on the CLASP intervention, see Miller et al. (2017).

All participants completed telephone follow-up assessments at weeks 6, 12, 24, 36, and 52 with chart reviews conducted at 6 and 12 months. This study received institutional review board approval at each site. For a more comprehensive overview of the study methods, see Boudreaux et al. (2013).

Participants

Participants who did not report a diagnosis of MDD, BD, or SSD were excluded from all study analyses (n = 208). All analyses included only participants with MDD (n = 347), BD (n = 298), and SSD (n = 100) in Phases 1 and 2, as participants in Phase 3 received CLASP, which aimed to improve treatment engagement. Participant demographic and clinical characteristics are presented in Table 1.

Table 1.

Demographic Characteristics by Psychiatric Diagnosis (N = 745)

MDD
(n = 347)		 BD
(n = 298)		 SSD
(n = 100)		 Group Differences
Variable Mean (SD) or N (%) Fχ2 η2/Cramer’s V
Age 37.30 (13.91) 38.96 (11.51) 40.33 (11.38) 2.77 0.01
Sex 7.37* 0.10
 Male 136a (39%) 120a (40%) 54b (54%)
 Female 211a (61%) 178a (60%) 46b (46%)
Ethnicity 1.98 0.05
 Latine 41 (12%) 32 (11%) 16 (16%)
 Non-Latine 306 (88%) 266 (89%) 84 (84%)
Race 3.94 0.07
 White 270 (78%) 237 (80%) 70 (70%)
 POC 77 (22%) 61 (20%) 30 (30%)
Highest Level of Education 38.31*** 0.16
 < High School 60a (17%) 52a (17%) 38b (38%)
 High School 83a (24%) 92a (31%) 34a (34%)
 Vocational/Technical 14a (4%) 10a (3%) 2a (2%)
 Some college 116a (33%) 86a,b (29%) 18b (18%)
 College graduate 53a (15%) 37a,b (12%) 5b (5%)
 Any post-graduate 21a (6%) 21a (7%) 3a (3%)
Sexual Orientation 2.73 0.06
 Heterosexual 303 (89%) 248 (85%) 87 (90%)
 Lesbian/Gay/Bisexual 38 (11%) 44 (15%) 10 (10%)
Marital Status 9.85** 0.12
 Married 84a (24%) 50 a,b (17%) 12b (12%)
 Not-Married 263a (76%) 248a,b (83%) 88b (88%)
Employment 27.68*** 0.20
 Full time 113a (36%) 69b (25%) 9c (10%)
 Non-full time/None 199a (64%) 212b (75%) 84c (90%)
Household 15.82*** 0.15
 Living alone 86a (25%) 82a (28%) 45b (45%)
 Living with others 261a (75%) 216a (73%) 55b (55%)
Substance Use Disorder
 Yes 87a (25%) 111b (37%) 42b (42%) 15.95*** 0.15
 No 260a (75%) 187b (63%) 58b (58%)
Past Week Suicide Attempt
 Yes 98 (28%) 83 (28%) 26 (26%) 0.20 0.02
 No 249 (72%) 215 (72) 74 (74%)
Suicidal Ideation Severity 13.78 (4.08) 14.16 (3.93) 13.95 (3.82) 0.74 0.002
Open in a new tab
*

p < .05

**

p < .01

***

p < .001

Note. Each subscript letter denotes categories whose column proportions do not differ significantly from each other at the .05 level; Follow-up duration was 52-weeks; BD = bipolar disorder; SSD = schizophrenia-spectrum disorder; MDD = major depressive disorder; Substance Use Disorder = self-reported alcohol or drug use disorder

Inclusion criteria required having attempted suicide or experiencing active suicidal ideation (answering yes to question #2 on the Columbia Suicide Severity Rating Scale [C-SSRS]; Posner et al., 2011) in the week prior to the ED visit. Exclusion criteria were: 1) medical/cognitive inability to participate in study procedures, 2) living in a non-community setting, 3) being under state custody or pending legal action, 4) no permanent residence or reliable telephone service, 5) an insurmountable language barrier, or 6) no self-reported diagnosis of MDD, BD, or SSD.

Measures

Psychiatric Diagnosis

Participants’ psychiatric diagnoses were assessed using the self-report question: “Have you ever been diagnosed by a doctor or therapist with…?” Response options included: depression, bipolar disorder, alcohol use disorder, any drug use disorder, anxiety disorder, attention deficit disorder, eating disorder, schizophrenia or schizoaffective disorder, or any other psychiatric disorder. SUD was defined as any alcohol use disorder or any drug use disorder self-reported at baseline.

Mental Healthcare Utilization

Medical record reviews were conducted at 6 and 12 months to capture the number and dates of ED and/or psychiatric inpatient hospitalization visits. Of note, all ED visits were recorded, not just those ED visits with a psychiatric reason as the presenting concern. Baseline ED visits resulting in psychiatric admission within two days before (some participants were admitted following informed consent but before baseline assessment) and seven days after the baseline assessment were regarded as patient’s index hospitalization. Participants self-reported outpatient mental health treatment utilization at each assessment timepoint. For the follow-up timepoints, outpatient mental health treatment utilization was combined across the entire follow-up period and then dichotomized (yes/no to having visited a mental health provider).

Suicide Attempts

Suicidal ideation intensity and past-week suicide attempts were assessed at baseline using the C-SSRS, a valid and reliable measure of STBs (Posner et al., 2011).

Statistical Analysis

Statistical comparisons were made across diagnostic groups (MDD, BD, SSD) to examine potential differences in demographic and clinical characteristics using one-way ANOVAs and chi-square tests. Next, index hospitalization and follow-up rates of outpatient therapy, inpatient psychiatric hospitalization, and ED visits were compared using chi-square analyses across the three diagnostic groups.

Because the ED-SAFE dataset does not include duration of follow-up ED visit or hospitalization, a multi-state Markov model could not be used, as we could not calculate the transition intensities based on the duration of state occupancy (sojourn time), which is one of the assumptions of multi-state models. Instead, Kaplan-Meier survival analyses were conducted to examine differences in the amount of time to psychiatric inpatient hospitalization and ED visit (no inpatient admission), respectively, during the follow-up period between diagnostic groups. The presence of a psychiatric inpatient hospitalization or an ED visit over the 52-week study period were entered as the “event” and days from baseline to the first hospitalization or ED visit, respectively, were entered as the “time.” Participants who did not have an outcome were censored at time of withdrawal or their last follow-up interview. Next, Cox regressions were conducted, with employment, sex, and SUD included as covariates (based on significant differences between groups and theoretical relevance to treatment utilization), diagnostic groups as the independent variable (BD, SSD, MDD), and psychiatric inpatient hospitalization or ED visit as the respective outcomes.

Finally, the relationship between relevant demographic (sex, age, education, employment, marital status, living status) and baseline clinical (depression, SUDs, chronic pain, past-week suicide attempt, outpatient mental health treatment) factors and psychiatric inpatient hospitalization and ED visits over follow-up were explored using two binary logistic regressions. Psychiatric diagnosis groups (MDD, BD, SSD) were combined in these analyses to preserve statistical power. Initial tests indicated that multicollinearity was not a concern (all VIF’s < 1.11).

Results

Participant Demographics

Participant demographics by group are reported in Table 1. The SSD group had a significantly greater proportion of participants who were male, had less than a high school level of education, lived alone, and were unmarried than the BD and MDD groups. Both SSD and BD groups had a higher proportion of participants with a self-reported SUD and a lower proportion of participants with full-time employment (see Table 1).

Aim 1a: Treatment Utilization by Diagnostic Group

Chi-square analyses indicated that there were no differences in the rates of attending outpatient therapy over follow-up between participants with MDD (n = 315; 91%), BD (n = 272; 91%), and SSD (n = 93; 93%), χ2 = 0.48, Cramer’s V = .03, p = .79. There were also no differences in index hospitalization between participants with MDD (n = 215; 62%), BD (n = 188; 63%), and SSD (n = 69; 69%), χ2 = 1.67, Cramer’s V = .05, p = .43. However, participants with BD (n = 84; 28%) and SSD (n = 30; 30%) had significantly higher rates of follow-up inpatient hospitalization compared with MDD participants (n = 67; 19%), χ2 = 8.92, Cramer’s V = .11, p = .01. Similarly, participants with BD (n = 133; 45%) and SSD (n = 48; 48%) had significantly higher rates of follow-up ED visits compared with MDD participants (n = 117; 34%) χ2 = 11.04 Cramer’s V = .12, p = .004.

Aim 1b: Rate and Time To Inpatient Hospitalization and ED Visit Over Follow-Up by Diagnostic Group

Psychiatric Inpatient Hospitalization

A Kaplan-Meier survival analysis showed that the MDD group had a significantly longer mean time to psychiatric inpatient hospitalization (312.11 days; 95% CI, 300.15 to 324.08) than the BD group (296.57 days; 95% CI, 282.79 to 310.56) and the SSD group (274.06; 95% CI, 246.03 to 302.10). Survival distributions for the three groups were statistically significantly different, χ2(2) = 9.26, p = .01. Pairwise log rank comparisons showed statistically significant differences in survival distributions for the BD versus MDD groups, χ2(1) =6.64, p =.01, and for the SSD versus MDD groups, χ2(1) =6.38, p =.01, with both BD and SSD groups having significantly shorter time to psychiatric inpatient hospitalization compared to the MDD group. The survival distributions for the BD and SSD groups were not statistically significantly different, χ2(1) = 0.42, p =.52 (see Fig. 1).

Figure 1.

Figure 1.

Open in a new tab

Rate of survival (i.e., lack of psychiatric inpatient hospitalization) over the course of the study period by psychiatric diagnosis

Note. BD = bipolar disorder; SSD = schizophrenia-spectrum disorder; MDD = major depressive disorder

A Cox regression showed a similar pattern of results. Participants with BD had significantly higher risk of psychiatric inpatient hospitalization over the study period than participants in the MDD group (hazard ratio, 1.42, 95% CI, 1.07 – 1.89, p =.02). However, there was no significant difference in psychiatric inpatient hospitalization between the SSD and MDD groups (hazard ratio, 1.42, 95% CI, 0.97 – 2.09, p =.08) or the BD and SSD groups (hazard ratio, 1.00, 95% CI, 0.69–1.45, p =.99).

ED Visit

A Kaplan-Meier survival analysis showed that MDD group had a significantly longer mean time to ED visit (276.70 days; 95% CI, 262.70 to 290.70), than the BD group (243.12 days; 95% CI, 225.99 to 260.25) and the SSD group (239.68 days; 95% CI, 210.45 to 268.92). The survival distributions for the three groups were statistically significantly different, χ2(2) = 12.02, p = .002. Pairwise log rank comparisons showed statistically significant differences in survival distributions for the BD versus MDD groups, χ2(1) = 9.03, p = .003, and for the SSD versus MDD groups, χ2(1) = 7.73, p = .005, with both BD and SSD groups having significantly shorter time to ED visit compared to the MDD group. The survival distributions for the BD and SSD groups were not statistically significantly different, χ2(1) = 0.30, p = .58 (see Fig. 2).

Figure 2.

Figure 2.

Open in a new tab

Rate of survival (i.e., lack of ED visit) over the course of the study period by psychiatric diagnosis

Note. BD = bipolar disorder; SSD = schizophrenia-spectrum disorder; MDD = major depressive disorder

A Cox regression showed the same pattern of results. Participants with BD had significantly higher risk of ED visit over the study period than participants in the MDD group (hazard ratio, 1.24, 95% CI, 1.01–1.54, p = .04), as did participants in the SSD group (hazard ratio, 1.36, 95% CI, 1.02–1.82, p =.04). There was no significant difference in ED visit between the BD and SSD groups (hazard ratio, 0.91, 95% CI, 0.69–1.21, p =.52).

Aim 2: Demographic and Clinical Factors Associated with Inpatient Hospitalization and ED Visit over Follow-Up Among Participants with MDD, BD, and SSD

A binary logistic regression examining the relationship between sex, age, education, employment, marital status, living status, depression, SUD, past-week suicide attempt, chronic pain, and outpatient mental health treatment and follow-up psychiatric inpatient hospitalization showed that not being fully employed and having utilized outpatient mental health treatment were significantly associated with follow-up psychiatric inpatient hospitalization (see Table 2).

Table 2.

Variables Associated with Acute Treatment Utilization over Follow-up

Inpatient Hospitalization
Variable B SE Odds Ratio 95% Confidence Interval
Sex (male vs. female) −0.05 0.19 0.96 0.66 to 1.39
Age −0.01 0.01 0.99 0.98 to 1.01
Education
 Graduated High School −0.05 0.31 0.96 0.52 to 1.75
 Graduated College 0.08 0.25 1.08 0.66 to 1.78
Employment (full time vs. not full time or none) 0.57 0.23 1.76* 1.12 to 2.77
Marital Status (married vs. not married) 0.10 0.25 1.10 0.67 to 1.81
Living Alone (yes vs. no) 0.34 0.21 1.40 0.93 to 2.10
Substance use disorder (yes vs. no) 0.37 0.20 1.45 0.99 to 2.13
Depression −0.02 0.02 0.98 0.95 to 1.02
Past-week suicide attempt (yes vs. no) 0.23 0.20 1.26 0.86 to 1.87
Chronic pain at baseline (yes vs. no) 0.14 0.21 1.15 0.77 to 1.73
Outpatient mental health treatment utilization 0.77 0.21 2.15*** 1.42 to 3.28
Emergency Department Visit
Variable B SE Odds Ratio 95% Confidence Interval
Sex (male vs. female) −0.05 0.17 0.96 0.69 to 1.33
Age −0.01 0.01 0.99 0.98 to 1.00
Education
 Graduated High School 0.10 0.27 1.11 0.65 to 1.89
 Graduated College 0.05 0.23 1.05 0.68 to 1.64
Employment (full time vs. not full time or none) 0.59 0.20 1.81** 1.23 to 2.67
Marital Status (married vs. not married) 0.34 0.22 1.41 0.91 to 2.19
Living Alone (yes vs. no) 0.27 0.19 1.30 0.91 to 1.88
Substance use disorder (yes vs. no) 0.52 0.18 1.68** 1.19 to 2.38
Depression −0.002 0.02 0.99 0.97 to 1.03
Past-week suicide attempt (yes vs. no) 0.11 0.18 1.12 0.79 to 1.59
Chronic pain at baseline (yes vs. no) 0.45 0.19 1.57* 1.09 to 2.26
Outpatient mental health treatment utilization −0.28 0.17 0.76 0.54 to 1.06
Open in a new tab
*

p < .05

**

p < .01

***

p < .001

Note. The reference group for Education was < high school diploma education level

The binary logistic regression was repeated with the same variables above but with ED visit (without psychiatric admission) as the outcome. Not being fully employed, presence of a SUD (alcohol or drug), and chronic pain were significantly associated with follow-up ED visit (see Table 2).

Discussion

People with co-occurring major psychiatric diagnoses and STBs presenting to ED settings are an exceptionally high-risk group of patients (Rabasco et al., 2024), who tend to repeatedly present for acute treatment (Azar et al., 2020) and struggle with engaging in outpatient therapy (McAlpine et al., 2000). However, there is a lack of research on the potential differences in treatment utilization among people with different major psychiatric disorder diagnoses and co-occurring STBs presenting to EDs. This paper examined differences in acute and outpatient treatment utilization between people with the diagnoses of MDD, BD, and SSD over a one-year follow-up, along with demographic and clinical factors associated with acute treatment utilization over follow-up.

We found that people with BD and SSD had higher rates of and shorter time to inpatient hospitalization and ED visit than people with MDD. When controlling for relevant covariates of employment, sex, age, and SUDs, participants with BD continued to have higher rates of inpatient and ED treatment utilization than participants with MDD over follow-up. Participants with SSD only had had higher rates of ED treatment utilization than participants with MDD, not inpatient psychiatric hospitalization. However, the effect size of the inpatient hospitalization difference between SSD and MDD participants was the same as for BD and MDD participants, suggesting that the sample of SSD participants may have been too small to capture significant differences between groups.

In general, these findings align with and extend previous research conducted with older adults (aged 60+), which found that patients with BD were four times more likely to have a psychiatric hospitalization compared with patients with unipolar depression (Bartels et al., 2000). It also contributes to the literature showing that people with BD and SSD have worse outcomes following ED visits compared with people who have other psychiatric disorders (Rabasco et al., 2024). This suggests that with BD and SSD may require additional support in ED settings to help prevent subsequent acute treatment utilization. This support likely needs to extend beyond simply increased outpatient discharge planning, due to the fact that the current study found no differences in outpatient treatment utilization by diagnostic group over follow-up. This may have been due to the overall high rates of outpatient treatment among all participants over the follow-up period (> 90%). Furthermore, there may have been nuances to outpatient treatment, such as frequency of visits or type of provider, that differed between diagnostic groups that our available data did not capture. Although it would be beneficial for future research to more comprehensively explore facets of outpatient treatment that may prevent acute treatment utilization among people with BD and SSD, it is also important to explore other avenues of support for people with these diagnoses in ED settings. For example, some research suggests that people with BD and SSD face greater mental health stigma compared with individuals who have other psychiatric diagnoses (Perich et al., 2022), and mental health stigma is associated with a host of negative outcomes (e.g., STBs, decreased medication compliance) that may require escalation to acute care (Oexle et al., 2017; Sickel et al., 2014). Therefore, it may be worthwhile for ED settings to provide brief psychoeducation interventions that address mental health stigma among this population.

Finally, the current research found that among participants with MDD, BD, and SSD, a lack of employment and having utilized outpatient mental health treatment were associated with increased risk for psychiatric inpatient hospitalization over follow-up, while lack of employment, having a SUD, and experiencing chronic pain were associated with increased risk for an ED visit over follow-up. Our findings align with and build upon past research showing that substance use (Irmiter et al., 2007) and unemployment (Lin et al., 2022) are associated with increased risk for acute treatment utilization among adults with a major psychiatric diagnosis. In addition, having a pain diagnosis has been found to be predictive of high rates of inpatient utilization among adults with co-occurring SUD and a major psychiatric diagnosis (Painter et al., 2018). Interestingly, having utilized outpatient mental health treatment in the six months prior to the baseline assessment was associated with increased risk for inpatient hospitalization over follow-up. Previous research has found that although utilization of any psychotherapy is associated with lower 30-day psychiatric inpatient readmission rates, those patients who have contact with more providers and services have a higher likelihood of psychiatric inpatient readmission (Huff, 2000). Therefore, it would be worthwhile for future research to provide a more fine-grained examination of how outpatient mental health treatment utilization relates to future inpatient hospitalizations, as patterns of results may change depending on the quantity and frequency of services.

These findings reinforce the importance of providers in ED settings providing relevant resources, such as vocational supports, substance use treatment referrals, or pain management treatment referrals, as part of the discharge plans for patients with major psychiatric diagnoses, which may be particularly helpful in reducing future acute treatment utilization.

Limitations

Although this study includes a number of strengths, such as a longitudinal design and multi-site recruitment, the results of this work should be interpreted in light of its limitations. First, acute treatment utilization over follow-up was assessed using chart review data, which may have resulted in missing the ED visits and inpatient hospitalizations occurring outside of the identified health systems. Second, all ED visits over the study follow-up were included in the analyses, not just those with suicidality or psychiatric factors as the presenting concern. This allowed for a more comprehensive understanding of treatment utilization for this population, as people with major psychiatric disorders also tend to have significant medical concerns (Bahorik et al., 2017). However, it would be beneficial for future research in this area to differentiate between medical and psychiatric ED visits, in order to develop more individualized interventions psychiatric versus medical ED recidivism within this population.

Third, participants self-reported their mental health diagnoses, which were not confirmed with diagnostic interviews. Although this may have led to some participants misreporting their mental health diagnoses, the sample characteristics by diagnostic group align with those found in previous research (e.g., people with SSD had higher rates of being unmarried and being unemployed; Thornicroft et al., 2004). They also have lower education levels compared with people who have BD (Vreeker et al., 2016), thus supporting the validity of the self-reported diagnoses. Fourth, because the ED-SAFE dataset does not include duration of follow-up ED visit or hospitalization, we were unable to conduct multi-state models, which would have allowed for examining multiple acute treatment episodes across follow-up. Future research could use multi-state models to explore how psychiatric diagnosis may play a role in time to repeated acute treatment episodes, rather than just the first ED visit or hospitalization. Finally, the sample size of participants in the SSD group was small (n = 100), reducing the power to detect differences in treatment utilization outcomes between groups, especially when including multiple covariates in the models.

Conclusion

In sum, ED visits and psychiatric inpatient hospitalization were higher over the one-year study follow-up period for those participants with BD and SSD than for participants with MDD. People with BD and SSD may require additional supports following ED visits to minimize future acute treatment utilization. In addition, lack of employment emerged as significantly associated with both ED visit and psychiatric inpatient hospitalization over follow-up, reinforcing that employment is an important treatment target for this population. This study aligns with and extends previous research by providing a more nuanced examination of diagnostic differences in treatment utilization among people with major psychiatric diagnoses and STBs presenting to the ED setting.

References

  1. Ahmedani BK, Westphal J, Autio K, Elssis F, Peterson EL, Beck A et al. , 2019. Variation in patterns of health care before suicide: A population case-control study. Prev Med. 127, 105796. 10.1016/j.ypmed.2019.105796 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Azar KMJ, Petersen JP, Shen Z, Nasrallah C, Pesa J, LaMori J, Pressman A, 2019. Serious mental illness and health-related factors associated with regional emergency department utilization. Popul. Health Manag. 23(6), 430–437. 10.1089/pop.2019.0161 [DOI] [PubMed] [Google Scholar]
  3. Bahorik AL, Satre DD, Kline-Simon AH, Weisner CM, Campbell CI, 2017. Serious mental illness and medical comorbidities: Findings from an integrated health care system. J Psychosom Res. 100, 35–45. 10.1016/j.jpsychores.2017.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bartels SJ, Forester B, Miles KM, Joyce T, 2000. Mental health service use by elderly patients with bipolar disorder and unipolar major depression. Am J Geriatr Psychiatry. 8(2), 160–166. 10.1097/00019442-200005000-00011 [DOI] [PubMed] [Google Scholar]
  5. Berardelli I, Sarubbi S, Rogante E, Erbuto D, Cifrodelli M, Guiliani C et al. , 2022. Exploring risk factors for re-hospitalization in a psychiatric inpatient setting: a retrospective naturalistic study. BMC Psychiatry. 22(1), 821. 10.1186/s12888-022-04472-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Boudreaux ED, Miller I, Goldstein AB, Sullivan AF, Allen MH, Manton AP et al. , 2013. The Emergency Department Safety Assessment and Follow-up Evaluation (ED-SAFE): method and design considerations. Contemp Clin Trials. 36(1), 14–24. 10.1016/j.cct.2013.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brennan JJ, Chan TC, Hsia RY, Wilson MP, Castillo EM, 2014. Emergency department utilization among frequent users with psychiatric visits. Acad Emerg Med. 21(9), 1015–1022. 10.1111/acem.12453 [DOI] [PubMed] [Google Scholar]
  8. Cearns M, Opel N, Clark S, Kaehler C, Thalamuthu A, Heindel W, et al. , 2019. Predicting rehospitalization within 2 years of initial patient admission for a major depressive episode: a multimodal machine learning approach. Transl Psychiatry. 9(1), 285. 10.1038/s41398-019-0615-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ceniti AK, Heinecke N, McInerney SJ, 2020. Examining suicide-related presentations to the emergency department. Gen Hosp Psychiatry. 63, 152–157. 10.1016/j.genhosppsych.2018.09.006 [DOI] [PubMed] [Google Scholar]
  10. Choi JW, Park S, Yi KK, Hong JP, 2012. Suicide mortality of suicide attempt patients discharged from emergency room, nonsuicidal psychiatric patients discharged from emergency room, admitted suicide attempt patients, and admitted nonsuicidal psychiatric patients. Suicide Life Threat Behav. 42(3), 235–43. 10.1111/j.1943-278X.2012.00085.x [DOI] [PubMed] [Google Scholar]
  11. Cook JA, Burke-Miller JK, Razzano LA, Steigman PJ, Jonikas JA, Santos A, 2021. Serious mental illness, other mental health disorders, and outpatient health care as predictors of 30-day readmissions following medical hospitalization. Gen Hosp Psychiatry. 70, 10–17. 10.1016/j.genhosppsych.2021.02.004 [DOI] [PubMed] [Google Scholar]
  12. Da Cruz D, Pearson A, Saini P, Miles C, While D, Swinson N et al. , 2011. Emergency department contact prior to suicide in mental health patients. Emerg Med. 28(6), 467–471. 10.1136/emj.2009.081869 [DOI] [Google Scholar]
  13. Donisi V, Tedeschi F, Wahlbeck K, Haaramo P, Amaddeo F, 2016. Pre-discharge factors predicting readmissions of psychiatric patients: a systematic review of the literature. BMC Psychiatry. 16(1), 449. 10.1186/s12888-016-1114-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. American Psychiatric Association. 2022) Diagnostic and statistical manual of mental disorders (5th ed., text rev.). [Google Scholar]
  15. Forte A, Buscajoni A, Fiorillo A, Pompili M, Baldessarini R, 2019. Suicidal risk following hospital discharge: A review. Harv Rev Psychiatry. 27(4), 209–216. 10.1097/hrp.0000000000000222 [DOI] [PubMed] [Google Scholar]
  16. Frahm Laursen M, Valentin JB, Licht RW, Correll CU, Nielsen RE, 2019. Longitudinal outcomes in pediatric- and adult-onset bipolar patients compared to healthy and schizophrenia controls. Bipolar Disord. 21(6), 514–524. 10.1111/bdi.12793 [DOI] [PubMed] [Google Scholar]
  17. Frueh BC, Knapp RG, Cusack KJ, Grubaugh AL, Sauvageot JA, Cousins VC et al. , 2005. Patients' reports of traumatic or harmful experiences within the psychiatric setting. Psychiatr Serv. 56(9), 1123–33. 10.1176/appi.ps.56.9.1123 [DOI] [PubMed] [Google Scholar]
  18. Gonçalves-Pinho M, Freitas A, von Doellinger O, Ribeiro JP, 2022. Bipolar disorder related hospitalizations - A descriptive nationwide study using a big data approach. Psychiatr Q. 93(1), 325–333. 10.1007/s11126-021-09951-6 [DOI] [PubMed] [Google Scholar]
  19. Hariman K, Cheng KM, Lam J, Leung SK, Lui SSY, 2020. Clinical risk model to predict 28-day unplanned readmission via the accident and emergency department after discharge from acute psychiatric units for patients with psychotic spectrum disorders. BJPsych Open. 6(1), e13. 10.1192/bjo.2019.97 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hooker EA, Mallow PJ, Oglesby MM, 2019. Characteristics and trends of emergency department visits in the United States (2010–2014). J Emerg Med. 56(3), 344–351. 10.1016/j.jemermed.2018.12.025 [DOI] [PubMed] [Google Scholar]
  21. Huff ED, 2000. Outpatient utilization patterns and quality outcomes after first acute episode of mental health hospitalization: Is some better than none, and is more service associated with better outcomes? Eval Health Prof. 23(4), 441–456. 10.1177/01632780022034714 [DOI] [PubMed] [Google Scholar]
  22. Hunt KA, Weber EJ, Showstack JA, Colby DC, Callaham ML, 2006. Characteristics of frequent users of emergency departments. Ann Emerg Med. 48(1), 1–8. 10.1016/j.annemergmed.2005.12.030 [DOI] [PubMed] [Google Scholar]
  23. Innes H, Lewsey J, Smith DJ, 2015. Predictors of admission and readmission to hospital for major depression: A community cohort study of 52,990 individuals. J Affect Disord. 183, 10–4. 10.1016/j.jad.2015.04.019 [DOI] [PubMed] [Google Scholar]
  24. Irmiter C, McCarthy JF, Barry KL, Soliman S, Blow FC, 2007. Reinstitutionalization following psychiatric discharge among VA patients with serious mental illness: A national longitudinal study. Psychatr Q. 78(4), 279–286. 10.1007/s11126-007-9046-y [DOI] [Google Scholar]
  25. Kessler T, Lev-Ran S, 2019. The association between comorbid psychiatric diagnoses and hospitalization-related factors among individuals with schizophrenia. Compr Psychiatry. 89, 7–15. 10.1016/j.comppsych.2018.12.004 [DOI] [PubMed] [Google Scholar]
  26. Killaspy H, Banerjee S, King M, Lloyd M, 2000. Prospective controlled study of psychiatric out-patient non-attendance. Characteristics and outcome. Br J Psychiatry. 176, 160–5. 10.1192/bjp.176.2.160 [DOI] [PubMed] [Google Scholar]
  27. Lin D, Kim H, Wada K, Aboumrad M, Powell E, Zwain G et al. , 2022. Unemployment, homelessness, and other societal outcomes among US veterans with schizophrenia relapse: A retrospective cohort study. Prim Care Companion CNS Disord. 24(5). 10.4088/PCC.21m03173 [DOI] [Google Scholar]
  28. Lorine K, Goenjian H, Kim S, Steinberg AM, Schmidt K, Goenjian AK, 2015. Risk factors associated with psychiatric readmission. J Nerv Ment Dis. 203(6), 425–430. 10.1097/nmd.0000000000000305 [DOI] [PubMed] [Google Scholar]
  29. Marcus SC, Chuang CC, Ng-Mak DS, Olfson M, 2017. Outpatient follow-up care and risk of hospital readmission in schizophrenia and bipolar disorder. Psychiatr Serv. 68(12), 1239–1246. 10.1176/appi.ps.201600498 [DOI] [PubMed] [Google Scholar]
  30. McAlpine DD, Mechanic D, 2000. Utilization of specialty mental health care among persons with severe mental illness: the roles of demographics, need, insurance, and risk. Health Serv Res. 35(1 Pt 2), 277–92. [PMC free article] [PubMed] [Google Scholar]
  31. Miller IW, Camargo CA Jr., Arias SA, Sullivan AF, Allen MH, Manton AB et al. , 2017. Suicide prevention in an emergency department population: The ED-SAFE study. JAMA Psychiatry. 74(6), 563–570. 10.1001/jamapsychiatry.2017.0678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Myers N, Bhatty S, Broussard B, Compton MT, 2017. Clinical correlates of initial treatment disengagement in first-episode psychosis. Clin Schizophr Relat Psychoses. 11(2), 95–102. 10.3371/csrp.Mybh.103114 [DOI] [PubMed] [Google Scholar]
  33. Niedzwiecki MJ, Sharma PJ, Kanzaria HK, McConville S, & Hsia RY, 2018. Factors associated with emergency department use by patients with and without mental health diagnoses. JAMA Netw Open. 1(6), e183528–e183528. 10.1001/jamanetworkopen.2018.3528 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Oexle N, Rüsch N, Viering S, Wyss C, Seifritz E, Xu Z et al. , 2017. Self-stigma and suicidality: a longitudinal study. Eur Arch Psychiatry Clin Neurosci. 267(4), 359–361. 10.1007/s00406-016-0698-1 [DOI] [PubMed] [Google Scholar]
  35. Olfson M, Wall M, Wang S, Crystal S, Liu SM, Gerhard T et al. , 2016. Short-term suicide risk after psychiatric hospital discharge. JAMA Psychiatry. 73(11), 1119–1126. 10.1001/jamapsychiatry.2016.2035 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Painter JM, Malte CA, Rubinsky AD, Campellone TR, Gilmore AK, Baer JS, 2018. High inpatient utilization among Veterans Health Administration patients with substance-use disorders and co-occurring mental health conditions. Am J Drug Alcohol Abuse. 44(3), 386–394. 10.1080/00952990.2017.1381701 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Perich T, Mitchell PB, Vilus B, 2022. Stigma in bipolar disorder: A current review of the literature. Aust N Z J Psychiatry. 56(9), 1060–1064. 10.1177/00048674221080708 [DOI] [PubMed] [Google Scholar]
  38. Pickens G, McDermott K, Smith M, Mummert A, Karaca Z, 2022. Trends in treatment costs of U.S. emergency department visits. Am J Emerg Med. 58, 89–94. 10.1016/j.ajem.2022.05.035 [DOI] [PubMed] [Google Scholar]
  39. Posner K, Brown GK, Stanley B, Brent DA, Yershova KV, Oquendo MA et al. , 2011. The Columbia-Suicide Severity Rating Scale: Initial validity and internal consistency findings from three multisite studies with adolescents and adults. Am J Psychiatry. 168(12), 1266–1277. 10.1176/appi.ajp.2011.10111704 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Rabasco A, Arias S, Benz MB, Weinstock LM, Miller I, Boudreaux ED et al. , 2024. Longitudinal risk of suicide outcomes in people with severe mental illness following an emergency department visit and the effects of suicide prevention treatment. J Affect Disord. 347, 477–485. 10.1016/j.jad.2023.12.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Ronaldson A, Elton L, Jayakumar S, Jieman A, Halvorsrud K, Bhui K, 2020. Severe mental illness and health service utilisation for nonpsychiatric medical disorders: A systematic review and meta-analysis. PLOS Med. 17(9), e1003284. 10.1371/journal.pmed.1003284 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Sickel AE, Seacat JD, Nabors NA, 2014. Mental health stigma update: A review of consequences. Adv Ment Health. 12(3), 202–215. 10.1080/18374905.2014.11081898 [DOI] [Google Scholar]
  43. Sirotich F, Durbin A, Durbin J, 2016. Examining the need profiles of patients with multiple emergency department visits for mental health reasons: a cross-sectional study. Soc Psychiatry Psychiatr Epidemiol. 51(5), 777–786. 10.1007/s00127-016-1188-5 [DOI] [PubMed] [Google Scholar]
  44. Stensland M, Watson PR, Grazier KL, 2012. An examination of costs, charges, and payments for inpatient psychiatric treatment in community hospitals. Psychiatr Serv. 63(7), 666–71. 10.1176/appi.ps.201100402 [DOI] [PubMed] [Google Scholar]
  45. Sullivan G, Young AS, Morgenstern H, 1997. Behaviors as risk factors for rehospitalization: implications for predicting and preventing admissions among the seriously mentally ill. Soc Psychiatry Psychiatr Epidemiol. 32(4), 185–190. 10.1007/BF00788237 [DOI] [PubMed] [Google Scholar]
  46. Suzuki Y, Yasumura S, Fukao A, Otani K, 2003. Associated factors of rehospitalization among schizophrenic patients. PCN. 57(6), 555–561. 10.1046/j.1440-1819.2003.01167.x [DOI] [Google Scholar]
  47. Thornicroft G, Tansella M, Becker T, Knapp M, Leese M, Schene A et al. , 2004. The personal impact of schizophrenia in Europe. Schizophr Res. 69(2), 125–132. 10.1016/S0920-9964(03)00191-9 [DOI] [PubMed] [Google Scholar]
  48. Tibirna A, Petrescu C, Ciobanu CA, Cercel GE, Manea MC, Ciobanu AM et al. , 2024. Hospitalization costs and mental health: Challenges and solutions from recent research - A narrative review. Preprints. 10.20944/preprints202406.1341.v1 [DOI] [Google Scholar]
  49. Velelekou A, Papathanasiou IV, Alikari V, Papagiannis D, Tsaras K, Fradelos EC, 2022. Factors influencing the duration of hospitalization of patients with schizophrenia. Med Pharm Rep. 95(3), 290–299. 10.15386/mpr-2144 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Vreeker A, Boks MPM, Abramovic L, Verkooijen S, van Bergen AH, Hillegers MHJ et al. , 2016. High educational performance is a distinctive feature of bipolar disorder: a study on cognition in bipolar disorder, schizophrenia patients, relatives and controls. Psychol Med. 46(4), 807–818. 10.1017/S0033291715002299 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Ward-Ciesielski EF, Rizv SL, 2021. The potential iatrogenic effects of psychiatric hospitalization for suicidal behavior: A critical review and recommendations for research. Clin Psychol. 28(1), 60–71. 10.1111/cpsp.12332 [DOI] [Google Scholar]
  52. Yan T, Greene M, Chang E, Broder MS, Touya M, Munday J, Hartry A, 2019. Hospitalization risk factors in antipsychotic-treated schizophrenia, bipolar I disorder or major depressive disorder. J Comp Eff Res. 8(4), 217–227. 10.2217/cer-2018-0090 [DOI] [PubMed] [Google Scholar]

RESOURCES