Medical Care for Patients With Mental Health and/or Substance‐Use Disorders: A Qualitative Investigation of Emergency Department Patient Experiences and Recommendations

ABSTRACT

Objectives

(1) To investigate the healthcare experiences of individuals with mental health and/or substance use disorders (SUDs) who seek medical care in the emergency department (ED) for physical health concerns (e.g., abdominal pain), and (2) to explore recommendations for improving these patients' care experiences. Although this population suffers from a high disease burden and disproportionately seeks care in the ED, surprisingly little research has examined their experiences.

Study Setting and Design

Qualitative study employing grounded theory and semi‐structured interviews with patients with mental health and/or SUD(s) (identified via health records) seeking care for a physical health concern during an ED visit to an academic medical center in the Northeastern USA. Longer follow‐up interviews were conducted via telephone. Interviews occurred between February 2018 and January 2019.

Data Sources and Analytical Sample

We transcribed and coded digital recordings of interviews with 50 patients at two time periods. We used constant comparative analysis throughout interviewing, coding, and analysis.

Principal Findings

Most participants were White (80%), non‐Hispanic (84%), and female (72%). Three key themes emerged, revealing a broad range of negative and positive patient care experiences that are influenced by provider, treatment, and healthcare system factors. Findings demonstrate that stigma and perceived biases due to mental health and/or SUDs extend to medical care experiences not directly related to these conditions (e.g., dismissive, rushed, unprofessional care, medical errors). Participants also identified positive care experiences (e.g., attentive, communicative, efficient, quality care), which informed recommendations for improving care.

Conclusions

Understanding care experiences of vulnerable patient populations is critically important for informing interventions to improve their healthcare quality and safety. Findings underscore the need for additional education and training for providers, integrated behavioral healthcare, and widespread changes to the healthcare system. Localized interventions that target stigma and mismatches between patients' care expectations and ED capabilities are also needed.

Summary.

• What is known on this topic

  • ○
    Patients with mental health and/or substance use disorders (SUDs) comprise a vulnerable population at risk for compromised healthcare quality and safety.
  • ○
    Elevated risks in this population extend to the medical care they receive for physical health concerns and contribute to healthcare disparities.
  • ○
    Little is known about these patients' experiences seeking medical care in the ED (where they disproportionately seek care) and how their lived experiences can inform recommendations for improving care.
• What this study adds

  • ○
    We identified risks to healthcare quality and safety for patients with mental health and/or SUDs seeking medical care in the ED. Risks emerged across multiple healthcare domains (provider, treatment, system).
  • ○
    In addition to negative ED experiences, many patients identified positive experiences, which informed recommendations for improving the quality and safety of care.
  • ○
    Patients' healthcare experiences and recommendations underscore the critical need for education, training, evidence‐based interventions, and widespread healthcare reform.

1. Introduction

Relative to the general population, individuals with mental health conditions and/or substance use disorder (SUD) are more likely to experience physical illnesses and symptoms [1, 2, 3], which contribute to healthcare disparities [1, 4, 5, 6] and increased mortality rates [7, 8, 9, 10, 11]. Research has begun to investigate the specific ways these patients receive medical care for physical health concerns (e.g., chest pain), finding barriers at the patient (e.g., financial circumstances), provider (e.g., stigma), and system (e.g., poor service coordination) levels [12, 13, 14]. Yet, little is known about how these patients' mental health and/or SUDs influence their medical care experiences in emergency departments (ED), where they disproportionately seek care [15, 16, 17, 18, 19].

The ED is a high‐risk practice environment [20] with constant medical demands, often operating with insufficient resources, and without established patient‐provider relationships [21, 22]. These factors may compromise care, especially for vulnerable patients, including those with mental health and/or SUDs [14]. Stigma can also threaten the care these patients receive [23, 24], as providers may hold negative attitudes and experience negative emotions in response to them, which can impair clinical decision‐making and care [12, 14, 25, 26, 27]. Further, stigma can contribute to diagnostic overshadowing [28, 29] leading providers to misattribute a patient's physical symptoms to their mental health and/or SUDs, which is prevalent in primary [12, 30] and emergency care [24, 31].

Despite that many patients with mental health and/or SUDs seek medical care in the ED, and the challenges of providing such care, little is known about these patients' care experiences. Research has examined providers' perspectives [14, 22, 31, 32], but with the exception of a few recent studies conducted in Canada [24, 33] and Australia [34], patient voices are overlooked [35]. Further, these investigations primarily focused on patients seeking ED care for mental health concerns [24] or patients from specific populations (i.e., those with opioid use disorders) [33], and consist of small patient samples [24, 33]. Data from one large survey in Australia found that patients with mental health conditions who utilized the ED for any reason rated their care experiences more negatively than patients without such a condition [34].

Legislation in the USA [36] and evolving economic, clinical, and political pressures have resulted in the ED serving as a “safety net of the safety net” [37] by providing care to everyone regardless of medical condition or ability to pay. Thus, research examining care experiences of vulnerable populations who disproportionately rely on the ED is urgently needed. This study aimed to enrich our understanding of the experiences of patients with mental health and/or SUDs and gather recommendations for improving care. We sought to (1) inform researchers' efforts to develop and implement interventions, (2) raise providers' awareness of factors that influence patient experiences to inform education, training, and clinical practice, and (3) highlight aspects of the healthcare system that policymakers should target for reform.

2. Methods

2.1. Study Design

We conducted semi‐structured interviews with patients who presented to an ED in the Northeastern USA with a physical health chief complaint (e.g., chest pain) and had a mental health and/or SUD(s) documented in their medical record. We conducted two interviews with each patient. The first was conducted face‐to‐face at each patient's bedside between February and December 2018. This interview allowed us to establish contact with patients and gather preliminary information about their visit and their mental health and/or SUDs (see supplement for interview guide). To characterize our sample, we also collected demographic information (e.g., race, ethnicity, gender). Given that patients were frequently in pain and currently receiving care, the first interview was designed to be brief (M = 16.35 min; SD = 9.42) and followed by a longer interview outside of the ED.

We conducted follow‐up interviews on the telephone to explore patients' ED experiences in greater depth. We expected patients to be more comfortable at this time, able to reflect and elaborate on their ED experience, and respond candidly. During this interview (M = 31.40 min; SD = 12.02), participants described their care experiences in detail, reflected on how their mental health and/or SUD(s) may have impacted their care, and made recommendations to improve patient experiences (see supplement for interview guide). Participants did not receive questions in advance for either interview. Follow‐up interviews were completed an average of 13.31 days after the first interview (SD = 8.92). We aimed to conduct follow‐up interviews within 2 weeks of initial interviews, but if we could not reach participants, we made up to three attempts to contact them. All interviews were completed by January 2019.

Interview topics and questions were informed by both a literature review and interviews being conducted with emergency physicians and nurses for a related study [14, 22]. We employed a grounded theory approach [38, 39] throughout interviewing and analysis. This rigorous method allows researchers to generate an understanding of a phenomenon through constant comparative analysis (CCA) in which new data is continuously analyzed and compared to earlier data to identify similarities, differences, and recurrent concepts in the data set. Thus, data collection and analysis occurred concurrently. In this inductive approach, data drives findings rather than being used to test or confirm hypotheses, which makes this method valuable for exploring the broad range of participants' ED experiences.

Interviews were conducted by our six‐person research assistant (RA) team (including JT and KB), which included five White females and one Persian male. Interviewers were rigorously trained by qualitative research experts EK and LMI, who are experienced researchers with PhDs in education (EK) and psychology (LMI). LMI and EK closely supervised data collection and coding. To enhance interviewer reflexivity, expert debriefing and memos were used. Whenever possible, the same interviewer was employed for both initial and follow‐up interviews (72% of patients); interviewers changed only if scheduling necessitated it.

Most patients (66%) were alone during hospital interviews; some (34%) had others present (e.g., family). If someone from the patient's care team entered during the interview, we paused until this person left. Occasionally, we ended the interview and completed it via telephone. To our knowledge, all follow‐up interviews were completed without others present. Participants received a $5 gift card for the first interview and a $40 gift card for the second one. Participants were not contacted after their second interview and did not review transcripts or provide feedback on results.

This study was approved by the University of Massachusetts Amherst institutional review board (Protocol #1285). To guide data collection, analysis, and reporting, we used the Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines and checklist. This study was funded by a grant from the Agency for Healthcare Research and Quality, which played no role in carrying out this research.

2.2. Sampling and Recruitment

Patients were eligible if they presented to the ED with a chief complaint relevant to a medical concern (e.g., chest pain), had a mental health and/or SUD(s) documented in their medical record, and had reliable telephone service. To identify patients, we obtained a Health Insurance Portability and Accountability Act (HIPPA) Waiver of Authorization that allowed us to review patients' health records during their visits. When RAs identified an eligible patient, they sought the patient's nurse (blinded to study) to enquire if the patient could be approached for a research study. If the nurse said yes (indicating the patient was stable and capable of interacting), an RA went to the patient's bedside, introduced themselves, and provided study information (if interested). Patients were told the study would help us “gain a better understanding of the experiences that patients have when seeking emergency medical care,” and that we are “particularly interested in how prior health conditions such as mental health concerns or substance use shape patient experiences.” Patients were informed they would be asked about their care experiences over two interviews—first in the ED and later via telephone. No additional information was provided. All participants provided written consent.

We attempted to approach all eligible patients (n = 227) during RA shifts. Our final sample consists of 50 patients (see Figure 1). In grounded theory, sample size is determined by achieving theoretical saturation, which occurs when data collection no longer yields new information, themes, or insights. Although sample size cannot be determined a priori [38], research suggests saturation typically occurs with 20–30 participants [40]; thus, we anticipated a sample of approximately this size would likely be necessary. After achieving saturation, we continued sampling to ensure we recruited patients with a broad range of health conditions for comparison purposes. When our final set of interviews yielded no new information, we ended data collection. All participants were recruited from a major academic hospital, which is well established, has a major trauma center, is in a large city, and trains medical students and residents.

FIGURE 1.

Flow diagram of patients included in the analysis. EHR, Electronic Health Record. *Two of these participants were inadvertently enrolled in the study despite not meeting eligibility criteria. For the third participant, the digital voice recorder malfunctioned during the follow‐up telephone interview, and very little information was obtained from this participant's first interview.**The final sample includes (a) two participants whose ED interviews were not recorded due to recorder malfunction, and (b) two participants who were interrupted for medical care after consenting to participate (and could not complete the first interview). All four participants completed the full interview on the phone and were retained in the sample.

2.3. Data Processing and Analysis

Digitally recorded interviews were transcribed verbatim by trained undergraduate RAs. We combined each patient's first and second interview and created one transcript; however, interviews were also analyzed separately. Each transcript was reviewed for accuracy by two other RAs. Coding proceeded under CCA guidelines employed in grounded theory [39, 41]. We first identified open codes, which broadly categorize conceptually similar ideas and assign labels to them (e.g., “stigma”). We then created axial codes in which we aggregated open codes to create larger categories of related concepts (e.g., “experiences with healthcare providers”). We continuously developed, refined, and tested open and axial codes throughout data collection and analysis to ensure they accurately reflected the data across all transcripts. Finally, we developed selective codes that integrated axial codes into overarching themes, which are presented in the results.

Transcripts were coded using NVivo qualitative software. Three trained RAs (including JT and KB) independently coded two transcripts and then met to review coding and resolve discrepancies. KB and the third RA then coded the remaining transcripts using a shared codebook. Discrepancies were resolved and new codes were developed during regular meetings and via email discussions. JT reviewed all coding for consistency.

3. Results

3.1. Sample Characteristics

See Table 1 for participants' demographic information and Table 2 for their chief complaints, mental health conditions, and SUDs.

TABLE 1.

Participant demographic information.

Mean age (SD)	49.62 (14.01)
Age range (median)	18–69 (54)
Gender	n (%)
Female	36 (72%)
Male	12 (24%)
Transgender	1 (2%)
No information provided	1 (2%)
Race
American Indian/Alaska Native, White	2 (4%)
Biracial or multiracial	2 (4%)
White	40 (80%)
No information provided	6 (12%)
Ethnicity
Hispanic/Latino/Spanish origin	6 (12%)
Not Hispanic/Latino/Spanish origin	42 (84%)
No information provided	2 (4%)
Highest level of education
Some high school	6 (12%)
High school diploma	14 (28%)
Some college	15 (30%)
Associate's degree	6 (12%)
Bachelor's degree	5 (10%)
Graduate or professional degree	4 (8%)
Employment status
Employed	18 (36%)
Unemployed	32 (64%)
Subjective financial situation
“I struggle to make ends meet”	18 (36%)
“I have enough money to get by but no more”	20 (40%)
“I have enough money to live well”	9 (18%)
No information provided	3 (6%)

TABLE 2.

Participants' chief medical, mental health, and substance use conditions.

	n (%)
Chief medical complaint
Chest pain	13 (26%)
Abdominal pain	6 12 (%)
Dyspnea	5 (10%)
Headache/migraine	3 (6%)
Back pain	3 (6%)
Kidney failure/pain	2 (4%)
Injuries from fall	2 (4%)
Other a	16 (32%)
Mental health conditions
Depression	35 (70%)
Anxiety	33 (66%)
Post‐traumatic stress disorder	12 (24%)
Bipolar disorder	11 (22%)
Panic disorder	7 (14%)
Eating disorder	5 (10%)
Other b	13 (26%)
Substance‐use disorder
Alcohol	18 (36%)
Cannabis	6 (12%)
Prescription opioids	3 (6%)
Cocaine	4 (8%)
Heroin	4 (8%)
Unspecified substance	10 (20%)
Mental health and SUD conditions
Mental health condition and SUD	25 (50%)
Mental health condition only	21 (42%)
SUD only	4 (8%)

Note: The percentage of participants with mental health conditions and substance‐use disorders reflects those who have multiple conditions (e.g., depression and anxiety). Thus, total percentages can exceed 100%.

Abbreviation: SUD, substance use disorder.

^a “Other” reflects chief complaints that only appeared once in the sample and includes, for example, rash, broken ankle, dysphagia, seizure, syncope, nausea/vomiting, respiratory illness, hiccups, and various pain complaints (e.g., leg, neck, foot, wrist pain).

^b “Other” reflects mental health conditions that appeared fewer than five times in the sample, and includes suicidal ideation (n = 4), borderline personality disorder (n = 3), adjustment disorder (n = 2), dissociative identity disorder (n = 2), agoraphobia (n = 1), and postpartum depression (n = 1).

3.2. Themes Reflecting Patient Care Experiences and Recommendations

Three key themes emerged revealing that patient care experiences are influenced by factors related to (1) healthcare providers (i.e., interpersonal interactions reflecting providers' attitudes and behaviors), (2) treatment (i.e., clinical treatment and diagnosis), and (3) the healthcare system. Patients reported negative and positive care experiences for each theme and made recommendations for improvement. Few differences emerged between patients with mental health and/or SUDs, or between in‐person and follow‐up interviews. Therefore, our results reflect combined interviews. Representative quotes appear in the text; additional representative treatment and system‐related quotes are in the S1. For each quote, the participant identifier appears in parentheses followed by whether the participant had a mental health condition (MH), an SUD (SUD) or both (MH + SUD).

3.3. Patient Care Experiences and Recommendations Related to Healthcare Providers

3.3.1. Negative Experiences

Many patients believed that when providers became aware of their mental health and/or SUDs either through their medical record or conversation, providers treated them differently, reflecting stigma and discrimination. For example, one noted:

Before they knew, they treated me so good. And then the minute they… saw the suboxone…they treated me like shit. (ID#17, MH)

Reflecting on the perceived intensity of discrimination and stigma, this patient later denied a SUD despite being on his chart: “I was pissed that they put that down on my papers. Cause I am not a substance abuse user.” Others minimized their conditions, perhaps to avoid stigma: “I don't consider it depression…I consider it frustration and disappointment.” (ID#38, MH + SUD).

Numerous patients reported that providers were dismissive of their concerns or failed to listen to their intuition about their symptoms even when a medical concern was considered, as one frustrated patient described:

I said ‘It's got nothing to do with my heart, please, would you please listen to me? My heart is fine. It's something else.’ And they don't listen. (ID#36, MH+SUD)

Concerns about providers not listening reflected a general lack of communication that permeated many patient experiences.

Many patients reported a lack of professionalism characterized by non‐serious or jocular interactions among staff:

If you're laying there sick and you're not happy…and all these doctors and nurses are smiling and joking around and laughing with each other. It's gonna…bother people. (ID#30, MH+SUD)

These behaviors contributed to patients feeling ignored.

Table 3 contains additional representative quotes.

TABLE 3.

Representative quotes reflecting positive and negative patient care experiences and recommendations related to healthcare providers.

1. Negative experiences

A. Stigma and discrimination

“So, if I never said anything, I wouldn't be flagged. So, if I ever go in with an ailment to where say pain medication would be a treatment, it might wreck it and I'm flagged that I was an addict, so that's automatically one strike against, you know?” (ID#15, MH + SUD) “I think when it comes to dealing with, with heroin addicts, I think it's just that there's already a stigma on you anyways… So, I think they kinda look at you a little different anyways. You know? They don't look at you like a normal human being who had made a mistake or, or something like that…” (ID#16, SUD)

B. Dismissive of patient concerns

“They didn't bother really looking or listening or doing any kind of diagnostics…” (ID#31, MH) “They kept walking past me and I'm having like a lot of chest pain and I'm like okay, this‐it didn't seem busy.” (ID#48, MH)

C. Lack of communication with providers

“I did the stress test, and you know I was supposed to go back for the second half with pictures, I guess. And the nurse came in and said that the doctor said it's not necessary. But [it] really wasn't explained why it wasn't necessary.” (ID#18, MH) “Well, I still… I think that if they told me that the doctor would be down here in about an hour, I would be expecting that. Rather than me thinking, “Oh he's going to come any minute now.” (ID#47, MH + SUD)

D. Lack of professionalism

“Well, a lot of times, no matter where you go a lot of them sit at the nurse's station giggling, laughing, you know, rolling on the seats and da‐da‐da‐da, I said, I yelled out, ‘Instead of sitting at the desk why don't you help patients!’” (ID#46, MH + SUD) “The nurses didn't really seem to care like one way or the other just would walk around bitching about their shifts right in front of me.” (ID#3, MH + SUD)

2. Positive experiences

A. Respectful and attentive to patient concerns

“They tried to calm me down because I feel panicked more when they said that there's something wrong with the heart, I feel more panicked. You know, I feel more upset. I told them, I said, ‘You make me more upset.’ No, they answer, ‘You are lucky you are here. We'll take good care of you.’” (ID#33, MH) “Well, I think they took care of me pretty well. Um you know any time that I had any uh concerns or anything like that there they were ready to address it.” (ID#26, MH + SUD)

B. Effective communication with providers

“… They explained why they were going to do a test that I needed to get done. They, you know, I was able to really sit and ask questions and they explained everything in full details.” (ID#1, MH) “They're…very very helpful and…they help you to understand what they what they doin, they don't just come in and do stuff, they explain it to you.” (ID#27, MH)

C. Professional and responsive care providers

“I mean I've always found, you know, I've always found that the nurses were always professional, they always, you know, even like something as simple as putting in‐ not that it's simple because I couldn't do it‐ but, you know, something, you know as simple for them as putting in an IV or anything like that, you know, I've never‐ never had any issues with the level of care.” (ID#11, MH) “And they take the time, they take the time and they courteous, they real courteous with me.” (ID#25, MH) “They were very responsive, they were very good, they tried to get me not in the hallways as soon as possible. And left me in a room and I had that red button thing that if I was having an emergency and if I had to go to the bathroom, they were willing to come help me.” (ID#37, MH + SUD) “They kept me warm, there was always a nurse or somebody checking to see if I'm alright and stuff like that.” (ID#44, MH + SUD)

3. Care recommendations

A. Greater attention and respect for patient concerns

“I hope that, you know… they can change a little bit and pay a little more attention to what the‐why the patient is there and listen to them! Listen! It's their body, they know what they feel and what's goin on.” (ID#41, MH) “I just think that there needs to be more listening to like the patient even though, obviously, the patient doesn't know more than the doctors but also it's just like… how do I say this? They definitely need to be someone there to kind of look in and advocate for what a patient is saying and actually like follow through on what a patient is asking for…” (ID#31, MH)

B. Greater communication and explanations from providers

“I don't even know if that's something that can be taken care of. I think the specialists need to take a little more time explaining the actual condition. And not just worry about coming in to do the quick assessments.” (ID#52, MH + SUD) “Yeah I mean I think its I definitely think it's helpful if they can give you an idea of what to expect. Like you know am I gunna be here for an hour am I gunna be here for five hours I mean am I going to be here all night…” (ID#50, MH)

Note: Numbers indicate the participant identification number, followed by whether the participant had only a mental health condition (MH), only an SUD (SUD), or both (MH + SUD).

3.3.2. Positive Experiences

Positive experiences with providers were characterized by listening to concerns, informing patients of aspects of care, and showing respect (see Table 3). Some patients identified the importance of clear, effective communication:

They wanted me to know exactly what was gonna happen and they gave me choices, and so, you know, I could make an informed decision, and I was…very happy. (ID#14, MH)

Professional and responsive attention also improved perceived care quality. Patients were grateful for consistent, comforting care (“She came in and checked on things, made sure I was comfortable.” [ID#20, MH + SUD]), and close attention:

They wanted to listen. They wanted to get as much information to try to diagnose you—as much as they could just by listening to you instead of just by the bloodwork and all that. (ID#40, MH+SUD)

3.3.3. Recommendations

Many patients wanted providers to attend to and respect their opinions and concerns to a greater extent (See Table 3):

I think doctors should listen to their patients, they can tell if somebody's off their rocker or if somebody's on drugs. I think [doctors] need to listen to what the patient has to say because a patient only knows their body… (ID#13, MH+SUD)

Patients also preferred greater communication and explanation, particularly about clinical conditions and treatment options, which may alleviate worries:

There was one four‐hour stretch where nobody came to talk to me. So, if there was… a way for a doctor or nurse to come in to say, ‘Hey, here's where we are. Processing things.’ Just because I was nervous… (ID#7, MS+SUD)

3.4. Patient Care Experiences and Recommendations Related to Treatment

3.4.1. Negative Experiences

Long delays throughout treatment, often coupled with very brief encounters with providers, frequently upset patients. Although this sometimes resulted from system‐related factors (discussed below), patients often lamented other causes. Many patients described rushed, hurried, or frantic treatment. Some believed providers were trying to see as many patients as possible, (“Let's see how many we can get in and out in as quick a time as we can” [ID#49, SUD]), or rushing to keep up with demand (“It felt very much like she was trying to get in, put [in] the IV and get out…it seemed like she had a lot of other things to do.” [ID#7, MH + SUD]). Some noted that they expected fast‐paced care, but nonetheless felt unheard:

Everybody seems so rushed…It's called emergency so I expect it to be that way…but I just feel like sometimes they don't give you…two seconds to get it out of your mouth. (ID#8, MH+SUD)

Patients were particularly frustrated when offered services that never arrived, (“They kept saying, ‘Well would you like some Tylenol or something?’ and I said ‘yeah that'd be great’ and they never gave it to me” [ID#32, MH]), or when requests were unfulfilled (“I asked one of ‘em for a pillow… I never got the pillow.” [ID#41, MH]).

Patients were highly distressed by clinical errors. Some reported being misdiagnosed or receiving no diagnosis, which they attributed to inadequate testing. For one patient, this manifested as a sense that providers performed unnecessary or incorrect tests and then dismissed their needs as non‐emergent:

I think that one major flaw is that they use unnecessary tests and saw nothing was wrong and was like, ‘You're okay.’ Like, nothing's an emergency. (ID#51, MH+SUD)

Patients also described diagnostic overshadowing:

They would have realized that there was something wrong ‐ they wouldn't just assume…If they weren't able to just assume anxiety, they would've…realized that this was an actual issue. (ID#5, MH)

Others reported receiving incorrect medications or doses, including a patient who described over‐prescription resulting in non‐compliance:

They gave me a ton of medicine. So much stuff, I couldn't even function… So, I said, ‘Screw this. I ain't taking this stuff anymore.’ And I stopped taking it. But then they had me so drugged. (ID#35, MH+SUD)

3.4.2. Positive Experiences

Several patients appreciated rapid treatment (“The doctors [are] very good there…they are fast, they are efficient, and the nurses are great” [ID#18, MH]), and many appreciated careful attention to their chief complaint, thorough procedures, adequate testing, and accurate diagnoses. One remarked that despite delays, adequate testing provided relief:

Even though I was [there] for about 10 or 11 hours, it still ended up going good. They did the tests that I needed…and when I brought up any sort of concern, the doctor ended up…doing other tests [to] ease my mind. (ID#2, MH)

Some patients reported feeling comfortable and respected when providers acknowledged their mental health and/or SUDs but did not disproportionately focus on these conditions. For example, a patient who disclosed anxiety reported a positive experience when providers carefully explained care:

They were very supportive, they explained stuff to me more to get a full understanding of what's going on…so, they had consideration for anxiety. (ID#1, MH)

Another patient appreciated providers checking‐in regarding mood (“I think they see the chart and they bring it up to me, and…basically ask me how I'm doing or whatever…how my mood is” [ID#23, MH + SUD]). Providers' willingness to discuss these conditions sometimes resulted in service initiation:

They knew [about my alcoholism] and I was very honest. And they were helpful and giving me resources and…that's how I'd gotten into this outreach program… There was a social worker at the hospital…[she] started…the intake process…and wanted me to get better. Like didn't judge me or anything like that. (ID#3, MH+SUD)

Conversely, a minority of patients preferred and appreciated medical treatment that did not address their mental health and/or SUDs, which these patients believed were irrelevant to their medical care:

They didn't talk to me about [my mental health] at all because it wasn't contributing to anything that was going on. Yeah so, they take real good care of me no matter what. (ID#27, MH)

3.4.3. Recommendations

Patients desired thorough treatment and careful consideration of their needs. For some, this included further diagnostic tests and not disregarding symptoms: “If one test doesn't show something, they should do something else because you're not there for nothing.” This was grounded in a belief that “They need to treat you the same regardless of if it's an emergency or not” (ID#41, MH). Patients also suggested providers view mental health and SUDs as part of the clinical background, but not without critical examination. One patient with chronic pain suggested that providers holistically assess drug history before negatively labelling patients:

So, when I keep going in there trying to get pills, I mean it's not our fault! We didn't ask for this. And if they could like swab our mouth…or [test urine] when we first go in for the pain, they'll see we're not like usual pain pill poppers. (ID#48, MH)

Additional representative quotes concerning treatment‐related experiences and recommendations are in the Table S1.

3.5. Patient Care Experiences and Recommendations Related to the Healthcare System

3.5.1. Negative Experiences

Key features of ED systems, resources, infrastructure, and policies adversely impacted patient care experiences. Some patients thought the prioritization system contributed to long waits to be seen, which were especially negative when unexpected (“Yesterday they said…the longest wait was 3½ hours, well 8½ hours later I was brought in” [ID#39, SUD]) or the patient was in pain (“I have a chest pain, and my face was swelling and they put me waiting!” [ID#43, MH]).

Limited resources also made many emergency visits unpleasant and slow. These included understaffing (“They were understaffed…so it took longer to get any kind of treatment” [ID#37, MH + SUD]) and perceptions of insufficient amenities (“I asked for something to eat a couple times, and they [weren't] able to get me things like that” [ID#26, MH + SUD]).

Overcrowding distressed many patients, including one who described her experience in a hallway:

It was so crowded, there was a whole mob. The lobby, the hallways, and patients crowding the hallway… It's a mess. (ID#42, SUD)

Crowded, exposed settings engendered patient concerns about the privacy of their medical information (“You can't have patients coming in…then talking about your medical histories in the hallway with everybody else” [ID#14, MH]) and others' (“So I'm sitting there listening to the nurses talk about this guy's medical stuff and you can't just go out” [ID#17, MH]).

Routine care processes (e.g., shift changes, handoffs, referrals) sometimes resulted in miscommunication among providers within the system, leading some patients to fall through the cracks, as one described:

This really got me mad, too…the nurse would say…‘I'm leaving shift now and there's another nurse coming on. I'll make her aware that you want to take a shower.’…’Never happened. (ID#9, MH)

Miscommunication among providers also resulted in patients receiving inappropriate referrals (They referred me out to a physician's assistant who…specializes in back fractures. That's not what I have.” [ID#52, MH + SUD]), or unnecessary restrictions:

They told me I couldn't eat or drink or go to the bathroom… Then we just like asked another nurse and she was like, ‘Oh I don't know why they told you [that] you couldn't eat or drink. Like that's weird.’ So, I could've eaten or drank. (ID#51, MH+SUD)

3.5.2. Positive Experiences

Satisfying system‐related experiences included brief waiting times, adequate privacy, and sufficient resources and amenities. For example, one patient reflected on their care, saying “It moves right along. The doctors are in…you're not waiting for much.” (ID#47, MH + SUD). Another felt pleased to receive their own room, noting “…it was just nice.” (ID#21, MH + SUD).

Several patients expressed gratitude for outpatient resources and referrals, with one discussing resources that matched their needs and preferences:

I still had the 24‐hour supervision, but [someone from behavioral health] came to visit me…she asked if I wanted to be in their resource program. It's music therapy and I love music. (ID#50, MH)

3.5.3. Recommendations

Patients largely focused on shorter waiting times to be seen (“Just make the waiting time for a bed shorter” [ID#23, MH + SUD]), greater privacy in waiting areas (“Put some type of screen block on there…” [ID#14, MH]), and more staff (“There's always a need for more staff…more help…more help and everything else” [ID#2, MH]). Additionally, some recommended handoffs that better inform providers, including direct handoffs from primary care doctors (“I wish there was a way to accommodate folks that have been instructed by their doctors to go the emergency room.” [ID#24, MH + SUD]).

Additional representative quotes concerning system‐related experiences and recommendations are in the Table S2.

4. Discussion

Patients with mental health and/or SUDs described negative and positive care experiences in the ED. Stigmatizing, dismissive, opaque, and unprofessional provider behaviors created highly negative experiences, whereas attentive, communicative, responsive, and professional behaviors engendered positive experiences. Long delays during treatment, rushed treatment, and clinical errors led to patient frustration; efficient, thorough, and appropriate treatment contributed to positive experiences. Negative system‐related experiences were characterized by overcrowding, lack of privacy, limited resources and amenities, and miscommunication among providers within the system; positive experiences were characterized by short waiting room times, privacy, sufficient amenities, and appropriate outpatient resources and referrals. These varied care experiences can have cumulative and interactive effects to shape patients' overall ED experience. Although many experiences we uncovered may overlap with those of patients without mental health and/or SUDs, the magnitude and impact of the negative experiences may be greater for patients with these conditions, as suggested in recent studies [14, 22, 42].

Stigma and perceived discrimination were particularly negative experiences for patients, consistent with research investigating emergency care for psychiatric conditions [24, 31] and SUDs [33]. Unsurprisingly, this stigma extends to medical care. Diagnostic overshadowing [28] and premature closure [43] are common among vulnerable patients across medicine and are often driven by clinician mistrust of patients' accounts of symptoms [44]. Our participants' negative experiences coincide with emergency physicians' and nurses' experiences caring for this patient population, captured in a qualitative study conducted at the same time as the current one with physicians and nurses from the same hospital (and seven others) [14, 22]. Many reported having negative attitudes toward these patients and engaging in the negative behaviors our participants reported [14, 42]. For example, providers described how mental health and/or SUDs can set up negative expectations about a patient, which may lead them to anchor on diagnoses (i.e., premature closure), be less likely to see medical illnesses (i.e., diagnostic overshadowing), be dismissive, spend less time with patients, and conduct less extensive workups. These findings validate our participants' experiences, highlight patients' awareness of discrepant care, and suggest negative experiences common to general ED patient populations may be exacerbated for patients with mental health and/or SUDs.

Participants identified system‐related factors that adversely impact patients and providers, trigger negative emotions, and contribute to patient safety risks [14, 22]. ED crowding, which is associated with long wait times, care in hallways, and privacy concerns, is exacerbated by limited community resources to treat patients with nonemergent needs—including many with primary mental health and/or SUDs. Challenges and frustration associated with treating this population in the ED reinforce provider stigma, which can adversely impact care when these patients present with medical concerns [14]. Broad healthcare reform, including increased funding and resources, is needed to address system‐level deficiencies. Integrated care, in which mental health professionals work alongside traditionally trained emergency staff [45, 46], should also be widely implemented. Such care reduces waiting times [47], patient distress [47], stigma [24], cost [47], and improves overall patient throughput [47, 48], signaling significant benefits for emergency healthcare systems.

Our data also suggest that system‐related factors can interact with other factors to shape patient experiences. Some of the negative care experiences reported by our participants may have resulted from a mismatch between their expectations and the capabilities and purpose of the ED. For example, participants reported negative experiences when providers failed to arrive at a definitive diagnosis after ruling out serious possibilities. Although patient dissatisfaction in such cases is understandable, greater public education about the role of the ED may increase patient understanding of the capabilities and limits of ED care. Mismatches in expectations also trigger negative emotions among providers, which can compromise care [14, 22, 42]. Efforts to reduce these mismatches should improve ED experiences for patients and providers.

Although negative experiences were particularly potent, many participants also described positive experiences (e.g., short wait time), some of which resulted from factors that are difficult to control (e.g., ED census; patient symptom severity). However, many resulted from respectful, attentive, and responsive providers, and careful clinical assessments and diagnosis, all of which can have cumulative effects on patient experiences. In related work, a small subset of ED physicians and nurses reported little to no bias in caring for patients with mental health and/or SUDs [14], greater empathy, advocacy, and time spent with patients [42]—some of which may be cultivated via training. Stigma‐reduction efforts [24, 49] may narrow the gap between the care experiences participants desired and the care they reported experiencing. Critically, anti‐stigma interventions must attend to structural processes that are key drivers of stigma [24]. Addressing how institutionalized structures, policies, and “cultures of caring” may generate stigma [24] should occur in tandem with other evidence‐based interventions [50, 51].

Whether and how providers should engage with patients about their mental health and/or SUDs when they seek medical care is important. Some participants praised providers who delivered high‐quality care without discussing these conditions, whereas others appreciated it when providers discussed symptoms. Participants generally felt that providers should respectfully note, and be willing to thoughtfully discuss, mental health and/or SUD symptoms. This preference aligns with the practice of narrative medicine, in which providers aim to recognize, absorb, and interpret patients' illness narratives and develop care plans together [52, 53, 54, 55]. Education and training to incorporate narrative methods into clinical practice can increase provider understanding, empathy, perspective taking, and compassion, and improve communication [52, 53, 54, 55], thereby enhancing providers' ability to understand and respect patients' views concerning their mental health and/or SUDs. Other interventions designed to enhance emotional intelligence can improve self‐awareness, self‐regulation, empathy, and perspective taking [56, 57, 58], and should also be considered.

4.1. Strengths and Limitations

Recruiting and interviewing a large sample of patients experiencing a broad range of health conditions during their ED visits, and again later, is a key strength of this work. However, participants were recruited from one ED in the Northeastern USA and are predominantly female, White, and non‐Hispanic, which may reduce the transferability of our findings to other populations. Individuals who belong to marginalized groups (e.g., racial and ethnic minorities, transgender individuals) face greater stigma in healthcare (and elsewhere) and may have more negative experiences than those we uncovered. Research is needed to investigate similarities and differences in ED experiences across diverse populations. Further, the transferability of our findings to other hospitals and geographic locations is unknown; however, interviews conducted with emergency physicians and nurses at the same hospital (and seven others) revealed almost no differences in how patients with mental health and/or SUDs are treated across hospitals [14, 22]. Thus, there may be a similar lack of variability in patient experiences across hospitals, potentially mitigating concerns about reliance on one site.

Many eligible patients could not be approached or declined to participate, and some could not be reached for follow‐up interviews. This may have introduced selection bias in which patients who participated in our study may differ from those who did not (e.g., by clinical presentation, race). Although we are unable to evaluate this possibility, selection bias may have resulted in an underrepresentation of the range of patient experiences and recommendations. Finally, self‐report or recall biases may have influenced our results. We attempted to minimize this by interviewing patients twice in different settings, emphasizing that interviewers were unaffiliated with the hospital, and assuring participants of confidentiality.

5. Conclusions

Patients with mental health and/or SUDs disproportionately seek medical care in EDs, yet research rarely examines first‐hand accounts of their experiences. This study uncovered negative and positive provider, treatment, and system‐related factors that contribute to these patients' experiences and offered recommendations for improving care experiences. Healthcare disparities and increased morbidity and mortality for this vulnerable population are well‐established; however, widespread changes to alter their healthcare outcomes have been slow to follow. Changes are urgently needed and will require localized, evidence‐based interventions and broad systemic changes to ensure safe, high quality, and equitable care [14, 22, 42].

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1.

Isbell L. M., Le V., Huff N. R., et al., “Medical Care for Patients With Mental Health and/or Substance‐Use Disorders: A Qualitative Investigation of Emergency Department Patient Experiences and Recommendations,” Health Services Research 60, no. 4 (2025): e14617, 10.1111/1475-6773.14617.

Data Availability Statement

No data are available. The datagenerated in this study are confidential interview transcriptsthat are not available for sharing.