Article Text
PDF
Abstract
Objective To identify and describe evidence on brief emergency department (ED)-delivered behavioural and care process interventions among patients presenting with suicide attempt or acute ideation, substance overdose or psychosis.
Design We employed a scoping review design and searched multiple data sources, clinical trial registries and references lists through March 2023. We included English-language trials and rigorously designed observational studies. In alignment with scoping review guidelines, we did not assess the quality of included studies or rate the strength of evidence of intervention effectiveness.
Population Our population of interest was adults presenting to the ED with suicidality (eg, attempt or acute ideation), any substance overdose or acute psychosis from a primary mental health condition.
Intervention We included studies of brief behavioural or care process interventions delivered in the ED.
Outcome measures Health outcomes (eg, symptom reduction), healthcare utilisation and harms.
Results Our search identified 2034 potentially relevant articles. We included 40 studies: 3 systematic reviews and 39 primary studies. Most studies (n=34) examined ED interventions in patients with suicide attempt or suicidal ideation, while eight studies examined interventions in patients with opioid overdose. No studies examined ED interventions in patients with acute psychosis. Most suicide prevention studies reported that brief psychological, psychosocial or screening and triage interventions reduce suicide and suicide attempt following an ED visit. Most clinical trial interventions were multicomponent and included at least one follow-up. All substance overdose studies focused on opioids. These studies often contained medication and referral or consultation components. Multiple studies reported increases in substance use disorder treatment utilisation; evidence on repeat overdose events was limited.
Conclusions A wide range of multicomponent ED-delivered behavioural health interventions for suicidality and opioid use disorder show short-term improvement on primary outcomes such as suicide reattempt. Few studies on non-opioid substances and psychosis are available.
- Systematic Review
- ACCIDENT & EMERGENCY MEDICINE
- Schizophrenia & psychotic disorders
- Substance misuse
- Suicide & self-harm
- Behavior
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
- Systematic Review
- ACCIDENT & EMERGENCY MEDICINE
- Schizophrenia & psychotic disorders
- Substance misuse
- Suicide & self-harm
- Behavior
STRENGTHS AND LIMITATIONS OF THIS STUDY
This scoping review summarises behavioural and care process emergency department (ED) interventions for acute mental health events not covered in previous systematic or scoping reviews.
This scoping review updates evidence summaries on brief ED suicide interventions from outdated systematic reviews.
Practising ED clinicians guided the research questions so that this review could inform ongoing care improvement.
The broad range of populations, interventions and outcomes included in this scoping review allows us to identify opportunities for targeted systematic reviews and knowledge gaps for future primary studies.
The scoping review design did not include risk of bias assessment, preventing us from synthesising the reported effectiveness of relevant interventions.
Background
Suicide and substance overdose are major public health concerns,1 2 for which care from the emergency department (ED) is frequently sought. As of 2020, a mental health emergency was among the ten most common reason for an ED visit in the USA.3 In 2022, the rates of non-fatal overdose presenting the ED were 180 per 100 000 and the rate of suicidal ideation was 400 per 100 000.4 5 In addition to suicide and substance overdose, psychosis is an acute mental health event for which emergency services are often needed.6 The ED is a critical healthcare touchpoint for individuals experiencing these events, and for some, it serves as their initial contact with healthcare services.
Discharge from the ED following an acute mental health event is considered a high-risk period during which symptoms may reemerge.7 To limit this risk assessment, brief screening and/or referral interventions delivered or initiated in the ED are used to more effectively link individuals from inpatient to outpatient services.8 9 Given the high priority of suicide prevention and substance use disorder (SUD) treatment in many health systems, including the US Department of Veterans Affairs (VA),10 11 and the dire public health need to provide both medical and mental health services in the ED context, we aimed to identify and describe available evidence on brief ED-delivered interventions among patients presenting with suicide attempt or acute ideation, substance overdose or psychosis. Because there is a large and methodologically complex evidence based on this topic and a need to understand the landscape of emergency behavioural health, we chose to conduct a scoping review that could identify opportunities for targeted systematic reviews, as well as knowledge gaps to be addressed with future research efforts.
Methods
This review is derived from a report produced by the VA Evidence Synthesis Programme that was intended to inform VA policy-making in the area of emergency medicine.12 The following questions, which were developed in consultation with a panel of clinicians and researchers in emergency medicine and mental health, were the focus of the review:
What studies have examined the effectiveness and harms of ED and urgent care centre screening, referral and management interventions for adults with acute mental health events (eg, suicidality, substance overdose or psychosis)?
What patient factors (eg, demographics, comorbidities, symptom severity), intervention features (eg, modality, duration, family involvement), or ED and urgent care centre setting characteristics have been identified as moderators of intervention effectiveness?
As a scoping review, the review protocol was not registered. We followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guidelines (see online supplemental table S1) for a checklist of reporting items.
Supplemental material
Patient and public involvement
No patients or members of the public were involved in the conception or completion of this review.
Study eligibility criteria
We included studies enrolling adults presenting to the ED or urgent care centre with suicidality (ie, attempt or acute ideation), substance overdose (opioids, alcohol, methamphetamine, kratom or benzodiazepines), or acute psychotic symptoms for conditions which psychosis which was primary diagnosis and was not secondary or due to another medical condition or medication. We included brief mental health intervention that included screening or risk assessment; triage; referral to inpatient, residential or outpatient settings; behavioural interventions (eg, safety planning, lethal means counselling, caring contacts); or treatment of agitation related to substance withdrawal. Included studies must have reported on engagement in outpatient, residential or inpatient mental healthcare; severity of acute symptoms (suicidality, severity of psychosis, etc); ED or urgent care centre outcomes (eg, boarding times, ED utilisation), patient or staff safety outcomes (eg, self-directed violence attempts in ED) or adverse events or harms of interventions. We included studies of randomised clinical trials and observational research which examined the aforementioned interventions and outcomes.
Legal hold interventions, medication comparative effectiveness trials, primary medical interventions and cardiopulmonary stabilisation (eg, airway management), and critical care management of use of reversal agents (eg, naltrexone) were excluded because they were outside the scope of this behavioural and care process focused review.
Data sources and searches
To identify articles relevant to the key questions, a research librarian searched Ovid MEDLINE, Ovid PsycINFO, CINAHL and ClinicalTrials.gov, as well as Agency for Healthcare Research and Quality, Cochrane Database of Systematic Reviews and Health Services Research and Development databases through March 2023 using terms for emergency department, suicide, overdose and psychosis (see online supplemental table S2 for complete search strategies). We limited the search to published and indexed articles involving human subjects available in the English language. We limited the search of systematic reviews within the last 7 years, while the primary literature was unlimited by publication date. Study selection was based on the eligibility criteria described above. Titles, abstracts and full-text articles were reviewed by one investigator and checked by another.
Supplemental material
Data abstraction and synthesis
All data abstraction was first completed by one reviewer and then checked by another; disagreements were resolved by consensus or discussion with a third reviewer. We used a standardised format to abstract relevant data, including details on the study population (ie, mental health condition, treatment setting, timing and demographics), intervention (ie, description of experimental practice, mode of delivery and interventionist/provider), comparator and outcomes (ie, description of outcomes reported and results for studies with comparators). Because we did not synthesise the findings of included studies, studies were not assessed for risks of bias. We analysed evidence narratively and in tables, describing patterns in available evidence and knowledge gaps to be filled by future research.
Results
The literature flow diagram (online supplemental figure S1) summarises the results of the study selection process.
Supplemental material
Literature overview
Our search identified 2042 potentially relevant articles. We included 42 studies: 3 systematic reviews and 39 primary studies. Most studies (N=34) examined ED interventions in patients with suicide attempt or suicidal ideation, while eight studies examined interventions in patients with opioid overdose. No studies examined ED behavioural or care process interventions in patients with acute psychosis. We identified six underway studies: four in patients with suicide attempt or suicidal ideation, one in patients with opioid dependence and one in patients with psychosis.
Suicide
We identified three systematic reviews that examined interventions among patients admitted to the ED following a suicide attempt (table 1). The first review examined the effects of interventions aimed at preventing repeat suicidal behaviour.13 This meta-analysis included 24 randomised controlled trials (RCTs) and found that active contact and follow-up interventions (ie, attempts by providers to stay in contact with patients after ED visits through home visits, telephone calls, postcards or letters) were effective in preventing a repeat suicide within 12 months (N=5319; pooled risk ratio (RR) = 0.83, 95% CI (0.71 to 0.97)). However, the effect did not appear to be maintained at 24 months (N=925; pooled RR=0.98, 95 CI (0.76 to 1.22)), although considerably fewer studies followed patients to this time point.
Systematic reviews on ED interventions for suicide attempt
Another systematic review more broadly examined the effectiveness of brief psychological interventions delivered in the ED in reducing suicidal thoughts and behaviour.14 This review defined brief interventions as ‘up to three sessions delivered in or soon after the presenting episode which can be supplemented by further follow-up contact’, and included two RCTs, one pilot RCT and one quasi-experimental study. The authors concluded that brief psychological interventions appear to be effective in reducing suicide and suicide attempt, though they could not determine the number of sessions or interventional components that were most effective. An additional review15 that, in part, investigated whether risk assessment tools can identify patients who are safe to be discharged, concluded that risk assessment tools should not be used in isolation but instead in conjunction with clinical judgement, communication with the patient and their family and awareness of community context (eg, availability of postdischarge care resources).
We also identified 10 RCTs of interventions delivered or initiated in the ED to reduce suicide attempts and other related outcomes among those who presented with a suicide attempt (tables 2 and 3).16–26 Across most studies, interventions were multicomponent, consisting of assessment and/or intervention and follow-up. Interventions included a single-session web-based lethal means counselling decision aid focused on reducing access to firearms and medications,18 safety planning in the ED with a peer recovery specialist,26 a single telemedicine mental health consultation for risk stratification and connection to services,27 a single telephone contact by a trained psychiatrist,25 a 10-session cognitive–behavioural therapy (CBT) intervention,19 a 6-session CBT intervention,24 a 6-session abandonment psychotherapy programme with an antidepressant medication protocol,17 mobile crisis team assessment and triage,20 5 structured consultations with the patient’s general practitioner,22 a single brief information session followed by telephone follow-up21 and 3 brief therapy sessions followed by regular personalised letters over 24 months.23 Interventions with a postdischarge component varied greatly in timing, delivering this component at 48 hours20 or 24 months23 after discharge. Nine17–25 out of 10 RCTs compared intervention effectiveness to that of usual care, while the remaining RCTs compared safety planning with a peer specialist to safety planning with an ED clinician.26 Of the six trials that examined suicide reattempt, which was the most common primary outcome assessed, five found a treatment effect19 21 23–25 and the other found no difference between the intervention and treatment as usual (see tables 2 and 3).17 Treatment effect on suicide reattempt over time were mixed: one study24 no longer detected a difference at 12 months, while another19 still detected an effect at 18 months. Additional outcomes assessed across the studies were related to healthcare utilisation (eg, ED readmission and mental health consultation), symptom severity (eg, depression) and feasibility (table 2).
Characteristics of included randomised controlled trials for suicide attempt
Intervention details and findings of included randomised control trials for suicide attempt
Finally, we identified 21 relevant observational studies (online supplemental table S3). Two larger comparative cohort studies measuring the same outcomes, and with similar findings, are notable. The first28 (N=1376) consisted of a universal suicide risk screening, coupled with provision of resources and post-ED telephone calls delivered by a physician or psychologist, and found a reduction in suicide attempts at 1-month postintervention compared with usual care (12% vs 22%, p=0.03). The second study,29 which included 1640 Veterans and was composed of a safety planning behavioural intervention in the ED with 2 follow-up monitoring calls completed by social workers or psychologists, also found a reduction in suicidal behaviour over 6 months (OR=0.56, 95% CI (0.33 to 0.95), p=0.03)
Supplemental material
Among the observational studies, there were many different interventions tested (see online supplemental table S3). Although some consisted of a single encounter such as a psychiatry ED consultation,29–32 psychosocial assessment32 or observation in the ED using a mobile audio-visual device,33 most were more involved with multiple telephone or in-person risk assessment and service linkage follow-ups over many months postdischarge.27–30 34–44 Several of the multicomponent interventions included a safety planning or lethal means training component delivered by trained mental health professionals.29 39 43 44 Other examples of interventions included brief solution focused therapy38 and a ‘caring contacts’ intervention that sent personalised expressions of kindness over the course of 12 months postdischarge.37 There was a large range of follow-up times from time of ED discharge29 to 10 years postdischarge.30 The outcomes assessed across studies also varied and included intervention feasibility and acceptability, suicide reattempt, suicidal ideation, knowledge of safety plan, depression, inpatient and outpatient mental health treatment utilisation and ED-related outcomes such as length of stay, return to ED and wait times (see online supplemental table S4). Across all included primary studies, harms or adverse events of interventions were either not reported or not assessed. Of note, several observational studies were focused on Veterans.29 37 43 44
Supplemental material
Among the four underway studies identified on ED-based interventions for individuals with suicide attempt or suicidal ideation, two studies examine psychotherapy or psychosocial interventions, one study examines a mobile phone application and one investigates a safety planning intervention. Two of these studies are completed with results posted to ClinicalTrials.gov but no identified publication, one is completed with no posted or published results, and another is in progress (see online supplemental table S5).
Supplemental material
Overdose
All studies examining ED interventions for overdose included patients with opioid overdose, and we did not find anything specific to alcohol, methamphetamine, benzodiazepines, nor cannabis. We identified two RCTs45 46 addressing behavioural interventions for this population. One 2019 RCT (N=256)45 examined the effects of a multicomponent intervention featuring one brief behavioural change counselling session (using motivational interviewing), overdose education, and provision of a naloxone kit compared with an informational flyer. The second RCT (N=648)46 compared ED behavioural interventions and follow-up care (daily for 10 days and then weekly for 3 months) delivered by a peer recovery specialist to a brief ED intervention delivered by a licensed clinical social worker (table 4). The 2019 study45 examined subsequent ED visits, subsequent overdose event, and time to first overdose event for 12 months following the intervention, while the 2022 study46 measured enrolment in treatment within 30 days of their ED visit. Neither study found significant group differences for any of the primary outcomes (table 5).
Characteristics of included randomised controlled trials for overdose
Intervention details and findings of included randomised controlled trials for overdose
The remaining six studies were case series47–50 or interrupted time series (online supplemental tables S6 and S7)51 52 using multicomponent interventions. A medication component (eg, take-home naloxone,51 52 naloxone training,47 in-ED naloxone50 or a buprenorphine prescription48) was common, often accompanied by a referral to consultation with a peer recovery specialist47 48 51 52 and/or and referral for substance use treatment.48 49 51 52 Interventions were delivered by a range of ED providers (eg, physicians) or other staff (nurses, peer recovery specialists and social workers) who had training in addiction counselling and/or motivational interviewing. Commonly assessed outcomes across observational studies included receipt of take-home naloxone, agreement to being connected to recovery support, referral to substance use treatment and participation in substance use treatment. Only one study assessed hospital admission and death.50 Four studies reported only postintervention outcome data, while two included pre–post intervention data51 52 (online supplemental table S7). Of studies with pre–post data, one reported51 an increase in patients receiving take-home naloxone, consultation with recovery coach, and discharge with referral to treatment after implementation of programme. The second52 reported an increase in number of patients discharged with naloxone after implementation of the intervention while fewer patients received behavioural counselling or referral to treatment after SUD treatment programme implementation. Again, no studies reported on or assessed adverse events from the interventions.
Supplemental material
Supplemental material
Finally, one observational study examined a combination of buprenorphine and referral for substance use treatment among individuals who had co-occurring suicidal ideation and opioid use disorder (OUD), and found it was feasible to initiate buprenorphine and have patients remained engaged in outpatients OUD treatment for 30 days.31 53 A single underway study was identified on an ED-based case management intervention with screening, assessment and referral for opioid dependent patients presenting to the ED. This study was completed in 2020 with results posted to ClinicalTrials.gov but no identified publication (see online supplemental table S5).
Psychosis
We did not identify any eligible literature examining brief interventions for psychosis that were delivered or initiated in the ED. A single underway study was identified, focusing on ED-based text messaging to engage young people with psychosis while waiting for an ED consultation with a non-ED psychiatrist after the patient has been discharged from the ED. All patients were then referred to an early psychosis intervention programme. This study appears to be in progress (see online supplemental table S5).
Intervention effect moderators
Suicide
A single study compared intervention effects of a manualised six-session abandonment therapy intervention delivered by a nurse to the same intervention delivered by a psychotherapist, finding no significant differences in rates of suicide attempt, suicidal relapse, and suicidal ideation between groups 3 months after the intervention.17 Both the nurse-delivered and psychotherapist-delivered intervention groups showed improvements on the aforementioned outcomes compared with treatment as usual (table 3). Another observational study of a four-session case management programme to link patients to services compared completers (defined as those who completed all four sessions) to non-completers (those who completed fewer than four sessions) and found that completers were more likely to have decreased suicide risk (65.3% vs 46.9%, Adjusted OR (AOR) 2.13 (95% CI 1.42 to 3.20)) and fewer untreated stressors (49.8% vs 61.1%, AOR 0.64 (95% CI 0.43 to 0.96), but no difference in lack of a support system (35.4% vs 45.7%, AOR 0.68 (95% CI 0.45 to 1.03).40
No other studies examined patient, intervention or setting characteristics as moderators of intervention effectiveness. Across studies Veterans were a commonly represented population, though no study compared effectiveness of intervention in Veterans compared with non-Veterans.
Overdose and psychosis
We did not find any studies in overdose or acute psychosis populations that directly examined moderation effects.
Discussion
The present scoping review identified a wide range of behavioural and care process interventions for suicidality, ranging from interventions delivered only during the ED stay to more complex, multicomponent interventions that were initiated during the ED stay and continued for several visits after discharge. Common components of the interventions included risk assessment, psychiatric consultation, safety planning and lethal means counselling. Most available RCTs reported a treatment effect in the form of decreased rates of suicide reattempt within a short time following an intervention. There were also several interventions that did not involve in-person contact (eg, were delivered via telephone or web-based video); most commonly this applied to follow-up visits rather than the ED-delivered intervention component. Although outcomes of various modalities were not directly compared, this flexibility in intervention delivery may assist in reaching a broader population and the ability to be implemented in EDs where trained mental health professionals may not be present.
Importantly, we also identified several qualitative studies that explored barriers and facilitators to implementing ED suicide prevention interventions including screening/risk assessment,43 54 55 psychiatry consultation and referral,56 and lethal means counselling.57 Across these studies, common themes of successful implementation of these strategies included the clinician having sufficient time and privacy, the screening protocol or intervention being integrated into the ED workflow, and the provider possessing rapport with patients and collaborating with colleagues. This final implementation strategy—collaboration with colleagues—should not be overlooked in its significance, as lack of trained mental health providers is a barrier across EDs to the provision of behavioural interventions.58 Additionally, to our knowledge, there are at least three prior systematic reviews focused exclusively on ED-delivered behavioural interventions for suicidality,13–15 the most recent of which was conducted in 2018. A main finding across the studies examined in these reviews were that repeated contacts with someone trained in suicide prevention decreased post-ED discharge suicidality.13 14 This kind of high touch intervention requires the support of additional resources or better integration with mental health, that are not readily available in many EDs, yet are needed.
Several behavioural interventions for opioid overdose were associated with successful linkage to specialty outpatient care. Most interventions studies were multicomponent, with the most common components being provision of naloxone and referral to SUD treatment. Although several interventions appeared to have promising effects on outcomes, only one RCT was identified that investigated these interventions, suggesting the need for additional rigorous studies in this population. There appeared to be several other gaps, including a need for additional research on brief ED-delivered interventions for non-opioid substances, such as alcohol.
We did not identify any primary studies that met our criteria for psychosis-focused behavioural or care process interventions delivered in the ED. Recommended practices of care for emergency management of psychosis are available, though they are based on primary and community care research. These guidelines state that safety and violence assessment is the first priority, followed by provision of medication (antipsychotic or others, depending on aetiology of psychosis).59 These guidelines also emphasise the importance of early intervention during the first psychotic episode to improve clinical outcomes.60 61
Limitations
Overall, this scoping review adds to previously conducted reviews by including a broader array of acute mental health events and intervention types. Nonetheless, this review is limited in that it does not offer a synthesis of available evidence, including medication only interventions and psychosis related to medication or other medical conditions. This was largely due to inconsistency in intervention components and characteristics and in the timing of outcome assessment across included studies. Studies themselves also had important limitations, particularly infrequent use of randomisation and comparison groups.
Conclusions
Existing evidence on the effectiveness of brief ED-delivered behavioural or care processes interventions among patients presenting with suicide attempt or acute ideation, substance overdose or psychosis is limited by methodological inconsistency, ethical challenges related to randomisation and setting level barriers to implementation. Future studies in suicide could examine differences in effectiveness by patient sociodemographic and clinical factors, intervention features (eg, modality such as telemedicine-delivered vs in-person, duration of intervention, family involvement), and ED setting characteristics (eg, rural vs urban, staffing models, bed capacity). Future research in opioid overdose should employ more rigorous research designs that feature comparison groups, such as treatment as usual. When a comparator is not feasible or ethical, studies should compare outcomes before and after the intervention is delivered (ie, a pre–post design). More generally, consistent assessment and reporting of adverse events is recommended. Finally, there appears to be a paucity of research on ED behavioural and care process interventions for psychosis and non-opioid substances, such as alcohol. Future research could examine brief interventions for these conditions that are initiated in the ED and continued in outpatient or community mental healthcare settings.
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
Ethics statements
Patient consent for publication
Ethics approval
Given the use of published aggregate data, ethics approval or oversight was not required.
Acknowledgments
We are grateful to Kathryn Vela for literature searching, Payten Sonnen for editorial and citation management support. We would also like to acknowledge the following contributors whose consultation during the topic development and report revision stages ensured relevance of this review to the VA: David Atkins, MD, MPH; Karen Bossi, MA; Chad Kessler, MD, MHPE; Christine Timko, PhD; and Jason Chen, PhD.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors SMN: study concept and design, acquisition of the data, analysis and interpretation of the data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content; JA: acquisition of the data, study design, and critical revision of the manuscript for important intellectual content; SKY: acquisition of the data, critical revision of the manuscript for important intellectual content. SMN, guarantor.
Funding Funding was provided by the US Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service.
Disclaimer The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.