Are New Standards for Assessing and Managing Suicidal Patients Needed in Canada?

Simulated Case Vignette

A recently divorced middle-aged man presents to the emergency department acutely intoxicated with suicide ideation. The patient is held overnight and then assessed by both the crisis team and staff psychiatrist. As the patient has no apparent psychiatric disorder and now denies suicide ideation, he is released from the emergency department. Seventy-two hours later, he dies by suicide after a self-inflicted gunshot wound. A quality improvement review of the case inquires about whether the psychiatrist and the hospital service were using standardized suicide risk assessment measures, routinely creating safety plans and completing caring contacts.

Introduction

The case vignette illustrates that “new standards” for the assessment and management of suicidal patients are expected to affect clinical practice in the next year or two. These standards will influence the delivery of care in psychiatric inpatient and outpatient services, emergency departments, and primary care settings and will directly affect psychiatrists, emergency physicians, family practitioners, and other mental health professionals. This analysis introduces the rationale and the nature of the new practices, as well as the controversy related to the new approaches, and recommends that Canadian psychiatric practice needs to be in the forefront of adopting new standards for caring for suicidal patients.

In the United States, Surgeon General David Satcher issued the first National Strategy for Suicide Prevention in 2001, and the strategy was revised in 2012 and included a new goal for “promoting suicide prevention as a core component of health care services.” ¹ This new goal was required because of the rise in suicide rates between 2001 and 2014 ² and the success of the Zero Suicide model that had been initiated at the Henry Ford Health System in Detroit. The Henry Ford Health System was the first to develop and launch Perfect Depression Care that included the Zero Suicide approach, resulting in a 75% drop in their suicide rate over 4 years after implementation. ³

The Zero Suicide approach aims to improve care for individuals at risk of suicide in health care systems and relies on a systemwide tactic to improve outcomes and close gaps rather than on the heroic efforts of individual health professionals. ^3,4 This quality improvement initiative has core components at 3 levels: a direct practice level that focuses on identifying the risk of suicidal behaviour and necessitates treating the risk of suicide behaviour as a distinct syndrome using best practice interventions; a process level related to quality and safety improvement to provide accessible, reliable, and continuous care to patients; and an organizational level that promotes a safety culture and a systemwide commitment to the aspirational goal of zero suicides. ⁵

Although Zero Suicide has captured much attention and serves as a stimulus for this review, a recent systematic review of clinical practice guidelines in suicide prevention illustrates that many organizations and clinical specialties are working to inform emerging standards related to clinical practice, research, and training in this area. ⁶ Accreditation bodies have promoted these new approaches to care; for example, the Joint Commission that accredits and certifies the quality of nearly 21,000 US health care organizations and programs in the United States released the Sentinel Event Alert (Issue 56, February 24, 2016) ⁷ calling for all US outpatient and inpatient health care settings to improve their assessment and care practices for patients at risk for suicide. The US Suicide Prevention Resource Centre developed the Zero Suicide Toolkit based on the essential elements of the Zero Suicide approach, ^{2 –4} and more recently, the National Action Alliance for Suicide Prevention in the United States has published the “Recommended Standard Care for People with Suicide Risk: Making Health Care Suicide Safe” ⁸ that outlines specific standard care elements for inpatient mental health settings, outpatient mental health settings, emergency departments, and primary care settings. The Zero Suicide model has been implemented in Canada, Australia, New Zealand, the United Kingdom, and Netherlands. ⁵ In Ontario, the Ontario Hospital Association’s “Strengthening Suicide Prevention in Ontario Hospitals” report ⁹ calls for the Ontario hospitals to adopt the Zero Suicide approach.

Our own 3-year case series of consecutive deaths by suicide (from 2012-2014) in southwestern Ontario that combined information from confidential health care provider questionnaires and the Ontario coroners’ files demonstrated that two-thirds (67.4%, n = 246/365) of individuals who died by suicide had consulted their primary care physicians prior to their deaths. Almost two-thirds (72.2%; n = 114/158) of these consultations were routine scheduled visits for nonurgent purposes; however, our respondent health professionals reported having completed suicide risk assessments in 108 (87%) of these visits. ¹⁰ Similarly, Ahmedani et al. ¹¹ found that suicide decedents were seen within a month prior to their death but were not seen related to a mental health diagnosis. These findings suggest that primary care and other clinicians must be guided to shift their suicide risk assessment from a focus on prediction to prevention. ¹² The Zero Suicide model has been adopted by primary care settings, ¹³ but this change of practice is in “its infancy.” ² The key elements to adopt a more preventive focus are described next.

The new standards incorporate the following in the care of patients seen in emergency departments, mental health services, and primary care. The Joint Commission called for screening all patients for suicide ideation, using a brief, standardized, evidence-based screening tool. ⁷ The National Action Alliance for Suicide Prevention’s “Recommended Standard Care for People with Suicide Risk: Making Health Care Suicide Safe” calls for primary care to screen all patients with mental illness and/or substance use disorders while recommending that emergency departments identify and assess all patients who have harmed themselves and those with mental illness and/or substance use disorders. ⁸ If the patient screens positive because of elevated risk, a more comprehensive assessment by a mental health professional using a standardized suicide risk assessment tool should follow. The tools to use may vary by clinical settings, but 2 examples of widely adopted and freely available tools are the Patient Health Questionnaire–9 (PHQ-9) and the Columbia Suicide Severity Rating Scale (C-SSRS). The PHQ-9 is a screening tool used in primary care that asks about suicide, and positive responses to the ninth item can identify patients having suicide rates 6 to 10 times higher in the subsequent year than patients with negative responses. ¹⁴ The C-SSRS has been adapted for use in a variety of settings and by various users (e.g., family, friends, first responders, researchers, and health care providers) and in many different countries. The scale can be used as a screening or assessment tool and is a reliable and validated measure that captures current and past suicide ideation, suicide attempts, preparatory behaviour, and nonsuicidal self-injury. ¹⁵ In addition, the necessary training to use the C-SSRS is available free of charge.

To manage patients at risk for suicide, according to the new standards, physicians should ensure that a safety plan intervention is carried out. A safety plan is a brief intervention collaboratively completed with the patient to develop a plan to recognize suicidal thoughts and to manage them safely. The plan includes 5 questions:

What are your warning signs that you are going into a crisis?

What coping strategies such as distraction or soothing techniques have you used successfully in the past?

What social situations and/or people can help distract you when you are in crisis?

Who can you ask for help when you are in crisis (or note if a person is unhelpful when you are in crisis)?

What professionals or agencies can you contact during a crisis? (adapted from Stanley and Brown ¹⁶ )

A crucial part of the safety plan is working with the patient and family to remove access to lethal means of suicide (e.g., guns, large quantities of medication). The physician should work out the plan, ensure that the agreed-upon actions have been taken, and document these actions in the health record.

The new standards call for providing “caring contacts” to patients with significant risk after hospital discharge or emergency department discharge, as well as follow-up in primary care settings and when care is interrupted (e.g., missed appointments, transfers of care). “Caring contacts” encompass contacts via phone calls, texts, or email as preferred by the patient and simply provide messages of support and encouragement. ¹⁴ The National Action Alliance for Suicide Prevention’s “Recommended Standard Care for People with Suicide Risk: Making Health Care Suicide Safe” provides specific guidelines for caring contacts by setting. ⁸ For example, for emergency departments, 1 caring contact should be completed within 48 hours of the visit and a second caring contact within 7 days of the visit. Similarly, following discharge from an inpatient psychiatric service, the recommended standard of care calls for caring contacts within 48 hours of discharge and within 7 days of discharge.

The Zero Suicide model also calls for adopting evidence-based interventions that directly address the patient’s risk for suicide. These approaches require that staff be trained in evidence-based psychotherapies that target individuals’ risk of suicide (e.g., dialectical behaviour therapy [DBT], cognitive behaviour therapy, Collaborative Assessment and Management of Suicidality, and Attempted Suicide Short Intervention Program; see Mokkenstorm et al. ⁵ ). These various psychotherapies for patients at risk for recurrent suicide behaviour have decreased the risk of future suicide attempts, lessened the medical risk from subsequent suicidal behaviour, and decreased the likelihood of emergency department visits for suicidal behaviour. These findings have led experts to extract a limited number of psychotherapy principles such as balancing validation with the need for change, fostering self-agency, and having access to supervision that may be effective in reducing the risk of future suicide behaviour in a variety of patients. ¹⁷ However, the National Action Alliance for Suicide Prevention has not included these treatment approaches as part of the “basic elements of suicide care” and, at this time, are considered additional “promising and desirable” elements. ⁸

As with any call for change, the Zero Suicide approach has been controversial. Many professionals have argued that setting a goal of zero suicides is inappropriate and creates additional distress for care providers working with patients at risk. ⁵ Critics argue that Zero Suicide will enhance the guilt felt by those bereaved by suicide, as the expectation of zero suicides is unrealistic. The adoption of the approach is also argued against because the evidence for these practices is only just emerging. To counter these criticisms, physicians should recognize that these approaches could be initiated without needing to set the aspirational goal of zero suicides. Other wording such as strategies to mitigate suicide risk or optimize suicide prevention has been recommended to support undertaking these changes in practice. ⁵ However, some leaders will argue that physicians typically resist recommended improvements and change ¹⁸ and that retaining the aspirational goal of zero suicides is needed. These different organizational approaches can lead to testable research questions.

It must be acknowledged that screening for the risk of suicide remains controversial; for example, the US Preventive Services Task Force did not recommend screening for suicide risk in adolescents, adults, and older adults in primary care. ¹⁹ Also, there is no consensus about what are the best screening and assessment tools to recommend. ⁸ Simon and colleagues ²⁰ have recently demonstrated the value of combining electronic health record data and self-report tools such as the PHQ-9 questionnaire to enhance prediction models for subsequent suicide attempts and suicide deaths following outpatient visits. With the broad adoption of electronic medical records, these novel prediction models can become standard decision support tools in the future.

The research support for these practices is accumulating. A recent randomized controlled trial demonstrated that soldiers seen for emergency appointments receiving crisis response planning (like safety planning; although safety planning includes a restricting access to means component) versus contracting for safety had on 6-month follow-up less likelihood of making a suicide attempt, experienced significantly faster reductions in suicide ideation, and had significantly fewer inpatient days. ²¹ Gysin-Maillart et al. ²² demonstrated the efficacy of the Attempted Suicide Short Intervention Program (ASSIP) versus treatment as usual in patients who had recently attempted suicide by finding a 18.4% lower rate of participants making at least 1 repeated suicide attempt compared to the comparison group at 24 months (8.3% vs. 26.7%). The ASSIP includes safety planning and follow-up personalized caring contacts as part of the intervention. Reducing access to means was felt to be a critical factor in the success of the Zero Suicide program at the Henry Ford Health System, ²³ and a recent review of existing clinical practice guidelines in suicide prevention indicated that almost all include the need to restrict access to means. ⁶ In addition, caring contacts following discharge from an inpatient service or after emergency department visits have been shown to reduce suicide and suicide attempts in the high-risk periods following these episodes. ²⁴

Over the next several years, the research evidence related to Zero Suicide will advance significantly because of specific funding competitions supported by the National Institute of Mental Health in the United States. ²⁵ For example, Labouliere et al. ²⁶ are operationalizing the Zero Suicide approach with the New York State Office of Mental Health involving almost 180 clinics servicing 86,000 Medicaid-enrolled patients per year, and their project will be the largest implementation and evaluation of the Zero Suicide approach ever conducted in outpatient behavioral health. Boudreaux and colleagues ²⁵ are testing the extension of their Emergency Department Safety Assessment and Followup Evaluation (ED-SAFE) study to 6 other emergency departments, 25 inpatient units, and 8 primary care clinics. Their suicide prevention approach used in the ED-SAFE study called for improved brief suicide risk screening, provision of outpatient suicide prevention discharge resources, and follow-up telephone counselling for the patient and a significant other. For youth aged 12 to 24 years, Asarnow and colleagues ²⁵ will test the value for reducing future suicide attempts of the Zero Suicide practices versus a stepped-care treatment approach, which includes Zero Suicide practices but then also matches intensity of treatment to severity of risk.

Conclusion

There is a growing momentum to apply these new standards to the care of patients at risk for suicide. Waiting on proof regarding suicide prevention is problematic as suicide is an extremely complex and rare phenomenon to study. Ethically, many true experiments related to suicidal individuals are not appropriate, and often it is impossible to adequately demonstrate if our interventions truly protect against suicide. However, if the Zero Suicide approach is implemented, ongoing evaluation should be included. For example, in the United Kingdom, the National Health Service Plan and the National Suicide Prevention Strategy for England and Wales recommended that crisis resolution teams be established across England, assertive outreach services were to be expanded, and a new legislative framework of care and treatment for mental health care patients was to be created. ²⁷ While and colleagues ²⁷ examined the impact of implementing these recommendations by comparing the before-and-after suicide rates of services that adopted most of the recommendations versus services that adopted fewer recommendations. Implementation of recommendations was associated with lower suicide rates in both cross-sectional and pre-post analyses; for example, the provision of 24-hour crisis care was associated with the biggest reduction in suicide rates: from 11.44 per 10,000 patient contacts per year (95% confidence interval, 11.12 to 11.77) before implementation to 9.32 (8.99 to 9.67) after implementation (P < 0.0001). Services that did not implement the recommendations had little reduction in suicide. We recommend that Canadian professions and organizations embrace new standards that include these clinical approaches while the necessary research continues. Hopefully, this analysis will help psychiatrists prepare for the upcoming changes in practice to prevent suicide that could enhance the care of patients within our health services.