Analysis of suicides reported as adverse events in psychiatry

Article in Review: Mackenhauer, J., Winsløv, J-H., Holmskov, J., Inger Brødsgaard, I., Gram Larsen, T.,& and Mainz, J. (2021). Analysis of Suicides Reported as Adverse Events in Psychiatry Resulted in Nine Quality Improvement Initiatives. Crisis, http://dx.doi.org/10.1027/0227-5910/a000787

___________________________________________________________________

Summary

This research has reviewed records of people who had died by suicide while in the care of a local hospital. An assessment was undertaken to look for breaches in clinical care, for the purpose of enabling quality improvement. Across the cases, it was found that documentation related to most cases was vague and inadequate. Insufficient risk assessments had also been carried out, which neglected to properly consider previous suicide attempts, substance use, illness or job loss. Protective factors and the insight of relatives were also largely neglected.

Setting

This study was carried out in Denmark, a country where the suicide rate declined from 38 deaths per 100,000 in 1980 to 11.4 deaths per 100,000 in 2000. By 2016, the suicide rate in Denmark had stabilised at 12.8 deaths per 100,000. As a nation with significant public health subsidies, most people with severe mental illness are treated in public hospitals.

Study inclusion and method

All instances of suicide reported as an adverse event between 2012 and 2016 were eligible for assessment. This did not include suicides occurring post-discharge or after the cessation of care. A total of 35 classes were analysed.

A medical chart review was undertaken, to look at:

• Demographics
• Time of suicide assessment
• Documentation of involvement of relatives
• Coding and descriptions of recent or previous suicidal behaviours
• Records of adherence to therapy
• Suicide risk factors
• Suicide protective factors

Most patients were well known to the institutions and had been receiving inpatient and outpatient treatment for spans of months to years. A few patients had only a single contact with the psychotic emergency department. A panel of physicians and psychiatrists reviewed the cases and determined if the main medical diagnoses was correct, incorrect or investigations were insufficient.

The panel found that for seven of the 35 patients (20%), the main diagnosis was correct. In 15 cases (43%) the diagnosis was unclear due to insufficient diagnostic assessment. Adequate suicide risk assessments and documentation were found to have occurred in just six of the 35 cases. Across all cases, the risk assessments tended to emphasise suicide ideation rather than other, well-known risk factors such as previous attempts, substance abuse, physical illness and job loss.

In many cases, the criteria for the main diagnosis and clinical details were vague and had inadequate medical chart documentation.

Improvement opportunities identified

Suicidal behaviour is multifactorial and can be triggered by multiple and simultaneous factors. An adequate suicide risk assessment is a critical way in which risk can be assessed and managed. A suitable suicide risk assessment should include:

• Description of the patient’s suicidal idealisation or self-harm
• Their clinical state
• The presence of permanent, temporary or potential risk factors
• Any protective factors

In a majority of these cases, there was an inadequate description of suicidal behaviours and risks. The risk assessments tended to focus on suicidal ideation and provided less information related to known risk factors such as physical illness, personal conflicts or job losses. Protective factors were rarely considered. Relatives were involved in only four of the 35 risk assessments, and yet relatives can often provide critical insight into risk, by reporting behaviours such as the individual describing feelings of hopelessness, stockpiling pills or changing their will.

The study found that there is a need for improvement related to documentation of psychosocial considerations, clinical reasoning and correct registration of additional diagnoses. Other quality improvement suggestions resulting from the study included:

• Training in the process of the assessment of suicide risks
• Improved internal and external collaboration
• The sharing of patient data across institutions

This research has demonstrated that suicide prevention and the completeness of assessments and documentation must be a focus for anyone who works with at-risk patients, in both primary and hospital care settings.