Introduction
As of 2014, there were over 170 000 residents in inpatient and other 24-hour residential treatment beds on any given night, an average of over 53.6 patients per 100 000 population. In 2018, the National Mental Health Services Survey estimated that the mean population rate for beds in mental hospitals in the United States was 39.0 per 100 000, with a median of 31.4.
Safety is paramount in the inpatient psychiatric settings since the criterion for admission is primarily based on acuity, severity, and danger to self or others.Reference Sakinofsky 1 Therefore, it is both a standard of care and a key measure of quality and safety per The Joint Commission (TJC) guidelines.
Suicide is the 11th major cause of death in the United States, of which inpatient suicides comprise a relatively small, yet clinically significant fraction, accounting for approximately 1500 cases annually. Furthermore, it accounts for the second most common sentinel event (SE), accounting for nearly 12% of all SEs.Reference Huang, Hu, Han, Lu and Liu 2 Nearly one-third of deaths by suicide occur while the patient is on 15-minute observations.Reference Mills, DeRosier, Ballot, Shepherd and Bagian 3 These data raise many scientific doubts including its association with separate risk factors, as suggested by some studies.Reference Agerbo 4 , Reference Høyer, Licht and Mortensen 5 There is marked variability in the use of universal screening tools, structured risk assessments, and clinical practices that also inspire further scientific inquiries.Reference Agerbo 4 The inconsistent use of risk assessment tools, ambiguity about protective observations, and flaws in the structural designs were thought to be plausible explanations for the increased risk for inpatient suicides.Reference Nelson, Denneson and Low 6
However, empirical evidence about specific risk factors for the inpatient population, and preventive and mitigation strategies are sparse, scattered, and on many critical issues not available.
Identifying individuals at risk
The rate of death by suicide among inpatients has been estimated much higher ie 600 to 800 suicides per 100 000 patient-years, which is nearly 50 to 72 times greater than the general population.Reference Steeg, Kapur and Webb 7 , Reference Madsen, Agerbo, Mortensen and Nordentoft 8 Therefore, methods of identifying high-risk patients were based on several reported factors like being away without leave at any time during the index admission, akathisia/extrapyramidal side effects at the time of suicide, and family history of suicide.Reference Sakinofsky 1 , Reference Sharma, Persad and Kueneman 9 -Reference Dong, Ho and Kan 11 Dong et alReference Dong, Ho and Kan 11 highlighted that most inpatient deaths by suicide occurred at a time when the patient was considered at no or low risk for suicide. The risk for death by suicide peaks immediately after admission or discharge. The first week of inpatient care is considered critical, and as much as 77% of deaths by suicide have been reported during this phase.Reference Madsen, Agerbo, Mortensen and Nordentoft 8 , Reference Dong, Ho and Kan 11 -Reference Qin and Nordentoft 13 A systematic review suggested a higher probability of death by suicide in inpatients with schizophrenia when on leave compared with patients with affective disorder. Furthermore, since agreed leaves were usually given later on during the admission, patients with affective disorders have a greater risk earlier in the hospitalization. Since the emergence of these data, the practice of leave during inpatient stays has been discontinued. Furthermore, death-by-suicide rates have also been correlated with admission numbers, and previous suicidal behaviors could indicate future risk for suicide.Reference Bowers, Banda and Nijman 14 These risk factors are highly correlated but do not have a cumulative effect on suicide risk, with studies reporting less than 2% predictive value for identifying high-risk individuals.Reference Large, Smith, Sharma, Nielssen and Singh 10
Overall, no group of psychiatric patients could be considered at lower suicide risk. Variations in findings have been observed for association with age, gender, marital status, employment, and educational qualifications.Reference Bowers, Banda and Nijman 14 Furthermore, data on association with religion, ethnicity, living alone, and forensic history are inconclusive.Reference Bowers, Banda and Nijman 14 This could be due to the differential presence of different groups in the samples, and therefore a separate evaluation of different groups could provide a clearer conclusion.Reference Bowers, Banda and Nijman 14 Although assessment of inpatient suicide risk can include several false positives, it is, therefore, crucial to avoid exclusion of individuals not at high risk.
Screening
Suicide risk assessment is a continual process, and its utility primarily rests on 4 principles: therapeutic relationship, communication and collaboration, documentation of the assessment process, and cultural awareness, with special considerations for given care settings, life span, and traumatic life experiences.Reference Holleran, Baker and Cheng 15 Screening inpatients at a higher suicide risk relies on evaluating underlying factors, such as risk factors, protective factors, and warning signs. Several inpatient suicide risk assessment screening tools and prediction models have been developed, with strengths and limitations. Overall, the sensitivity or Area Under The Curve (AUC) for most tools is nearly 80% or 0.7, respectively, highlighting fair or better risk discrimination in patients; however, their clinical use is limited by the frequency of false positives, suggesting the need for further wide-scale studies to refine these methods. The Columbia-Suicide Severity Rating Scale is a commonly used suicide screening tool that has shown good reliability and sensitivity.Reference Posner, Brown and Stanley 16 Although a UK advisory body and the National Institute for Health and Care Excellence recommended that “assessment tools and scales designed to give a crude indication of the level of risk (eg, high or low) of suicide” should not be used.
Prevention strategies
Suicide prevention strategies could be divided into 3 core components: (1) improving detection and awareness; (2) improving response refinement and standardization; and (3) improving patient-focused care. Table 1 summarizes risk factors and prevention strategies.
Summary of Studies Evaluating the Risk of Inpatient Suicides and Prevention Strategies
Abbreviation: CAMS, collaborative assessment and management of suicidality; HFMEA, health care failure mode and effective analysis; MHEOCC, Mental Health Environment of Care Checklist; PACT, post-admission cognitive therapy; VA, Veterans Affairs.
Improving detection and awareness
Suicide-risk detection should be improvised by decreasing the variability in risk screening protocol by incorporating screening questionnaires that take less time and effort for the patient. Suicide screening should also be developed specifically for the inpatient population, for example, suicidal risks for inpatients with autism spectrum disorder are less studied; therefore, identifying those individuals with unknown risks is critical.Reference Horowitz, Thurm and Farmer 17 The use of suicide risk assessments is now been recommended for all inpatients by TJC.Reference Dahale, Sherine and Chaturvedi 12 The staff training from all disciplines is encouraged to educate about newer evidence related to suicide risk. The poster campaigns and mandatory lectures and training have proved beneficial in identifying and responding to high-risk patients.Reference van Landschoot, Portzky and van Heeringen 18 -Reference Ramberg, Di Lucca and Hadlaczky 20
Standardizing safety protocols and suicide-proof architecture
Hospital safety measures and designs, as well as the availability of resources, should be regularly monitored. Since hanging is the most common method of suicide attempt and completion in inpatient units, lanyards and anchor points should be removed.Reference Mills, DeRosier, Ballot, Shepherd and Bagian 3 , Reference Gupta, Moll and Gupta 21 The Mental Health Environment of Care Checklist (MHEOCC) could be followed in the hospital setting as these have been successfully implemented in some studies, resulting in decreased suicide rate from 2.64 to 0.087/100 000 admissions, and 4.2 to 0.74/100 000 admissions in another study.Reference Watts, Young-Xu and Mills 22 , Reference Watts, Shiner, Young-Xu and Mills 23 Similarly, the health care failure mode and effective analysis can help establish a comprehensive inpatient suicide prevention network.Reference Changchien, Yen and Wang 24 Once a patient is identified as a suicide risk, the process of transferring to the MH unit should be streamlined without prolonged waiting. This must include effective functioning of these facilities on all days, throughout the year, including weekends and holidays. A reduction in waiting time in the Emergency Room and faster patient transition from a suboptimal environment to an appropriate MH facility would also be recommended.
Patient-focused care
The role of patient-centric care and understanding the perspectives of the patient and the caregiver is another key strategy in mitigating suicide risk. This also includes educating them about the treatment process and encouraging shared decision-making. Communication and/or coordination between the at-risk patients and their care providers is paramount to alleviating suicide risk.Reference Dahale, Sherine and Chaturvedi 12 Psychiatric consultation should be encouraged and set up with appropriate follow-up care is considered an established standard of practice. Increasing length of hospital stay and readmission have shown promising results in preventing suicide risk in psychiatric inpatients.Reference Agerbo 4 Pharmacological interventions like clozapine administration for 6 weeks have been shown to decrease the rate of suicidal behaviors from 28% (pre-clozapine) to 3% during the administration period, followed by 18% in the post-clozapine period. Psychological interventions such as post-admission cognitive therapy or collaborative assessment and management of suicidality have proved beneficial in reducing depression, hopelessness, and suicidal ideation in most cases.Reference Ghahramanlou-Holloway, Cox and Greene 25 -Reference Ellis, Rufino, Allen, Fowler and Jobes 27 The use of lithium in treatment-resistant affective illness is an evidence-based treatment modality to reduce the risk of suicide.Reference Guzzetta, Tondo, Centorrino and Baldessarini 28
Conclusion
One death by suicide is too many, and such an event while undergoing inpatient treatment raises many critical questions. Among many challenges, and a lack of strong empirical support, several measures could be included in clinical practice. First, studies suggest no group of patients considered at low risk but recommend using protective observations during the entire stay, more specific close observations, and precautions for an acutely suicidal patient. The protective observation intervals must change randomly, ideally less than 15 minutes to make the time interval less predictable. The first week of the inpatient stay has reported the highest deaths by suicide. Secondly, an emphasis on a specifically tailored treatment plan focused on the individual needs of the patients by trained, informed, and educated mental health professionals. Lastly, a systematic process in developing a suicide-proof architecture of the mental health facilities. The use of risk assessment tools is helpful; however, merely relying on high- and low-risk scores is been discouraged and must be complemented with evidence-based treatment.
Disclosures
The authors do not have anything to disclose.
