National Confidential Inquiry into Suicide and Homicide by People with Mental Illness: new directions

Absconding from Public Mental Health Inpatient Units - Who Does it, and why?

Patients who abscond from acute inpatient psychiatric wards put themselves and others at risks of variable nature and severity. There is a limited understanding of what motivates them to do so. The research in this area is limited and predominantly outdated. To better understand the characteristics, patterns and predictors of people who abscond from hospital, this study compared absconding and non-absconding patients in a tertiary public hospital. Demographic and clinical characteristics were audited for patients who absconded (n = 159) and who did not abscond (n = 100) throughout a 2-year period. Patients who absconded in the public settings were more likely to be male, have suffered from a psychotic illness, had an increased number of psychiatric comorbidities, had a history of aggression, and used greater numbers of substances. They also had experienced homelessness, were case managed, and were managed as involuntary patients. Predictors were similar, and included involuntary legal status, greater number of substances used, greater numbers of comorbid psychiatric illnesses, polysubstance use, an absence of self-harm history, shorter duration of admission and male sex. This study has implications for the identification of those who are at greater risk of absconding from an inpatient unit. Through enhanced understanding, greater measures can be taken to minimise absconding and its associated risks.

Testing Suicide Risk Prediction Algorithms Using Phone Measurements With Patients in Acute Mental Health Settings: Feasibility Study

BACKGROUND

Digital phenotyping and machine learning are currently being used to augment or even replace traditional analytic procedures in many domains, including health care. Given the heavy reliance on smartphones and mobile devices around the world, this readily available source of data is an important and highly underutilized source that has the potential to improve mental health risk prediction and prevention and advance mental health globally.

OBJECTIVE

This study aimed to apply machine learning in an acute mental health setting for suicide risk prediction. This study uses a nascent approach, adding to existing knowledge by using data collected through a smartphone in place of clinical data, which have typically been collected from health care records.

METHODS

We created a smartphone app called Strength Within Me, which was linked to Fitbit, Apple Health kit, and Facebook, to collect salient clinical information such as sleep behavior and mood, step frequency and count, and engagement patterns with the phone from a cohort of inpatients with acute mental health (n=66). In addition, clinical research interviews were used to assess mood, sleep, and suicide risk. Multiple machine learning algorithms were tested to determine the best fit.

RESULTS

K-nearest neighbors (KNN; k=2) with uniform weighting and the Euclidean distance metric emerged as the most promising algorithm, with 68% mean accuracy (averaged over 10,000 simulations of splitting the training and testing data via 10-fold cross-validation) and an average area under the curve of 0.65. We applied a combined 5×2 F test to test the model performance of KNN against the baseline classifier that guesses training majority, random forest, support vector machine and logistic regression, and achieved F statistics of 10.7 (P=.009) and 17.6 (P=.003) for training majority and random forest, respectively, rejecting the null of performance being the same. Therefore, we have taken the first steps in prototyping a system that could continuously and accurately assess the risk of suicide via mobile devices.

CONCLUSIONS

Predicting for suicidality is an underaddressed area of research to which this paper makes a useful contribution. This is part of the first generation of studies to suggest that it is feasible to utilize smartphone-generated user input and passive sensor data to generate a risk algorithm among inpatients at suicide risk. The model reveals fair concordance between phone-derived and research-generated clinical data, and with iterative development, it has the potential for accurate discriminant risk prediction. However, although full automation and independence of clinical judgment or input would be a worthy development for those individuals who are less likely to access specialist mental health services, and for providing a timely response in a crisis situation, the ethical and legal implications of such advances in the field of psychiatry need to be acknowledged.

Suicides in Psychiatric Patients: Identifying Health Care-Related Factors through Clinical Practice Reviews

The objective of this study was to identify health care-related factors associated with death by suicide in psychiatric patients and to gain insight into clinician views on how to deal with suicidality. The study material derived from a clinician committee in a psychiatric department reviewing every outpatient and inpatient suicide in a standardized way. Reports' conclusions and corresponding plenary discussion minutes regarding 94 suicides were analyzed using inductive thematic content analysis. Health care-related factors were categorized into 4 themes: patient evaluation, patient management, clinician training, and involvement of relevant non-clinical partners. Clinician views on the themes were expressed through statements (i) promoting or restricting an aspect of care (here called recommendations), which mainly followed existing guidelines and were consensual and (ii) without precise indication (here called comments), which departed from mainstream opinions or addressed topics not covered by existing policy. Involvement of non-clinical partners emerged as a new key issue for suicide prevention in psychiatric departments and should be openly discussed with patients. Clinicians preferred balanced conclusions when they reviewed suicide cases.

Learning from critical incidents

Critical incident reviews are an integral part of modern psychiatric practice. The issue is central to the clinical governance agenda in the UK, yet there is widespread debate about their usefulness. There is a lack of systematic research into their impact on clinical outcomes, with most authors commenting on their form, their political implications, and whether they should exist at all. This article explores the historical basis to incident investigation, outlines an ‘ideal’ method of review and discusses the concepts of the learning organisation and root cause analysis. Further discussion focuses on what the objectives of critical incident review might be and whether organisations as a whole can learn from them.

Psychiatric and psycho-social characteristics of suicide completers: a comprehensive evaluation of psychiatric case records and postmortem findings

Objectives

To explore the demographic, psychosocial and clinical characteristics of individuals known to the mental health services, who died by probable suicide in the West of Ireland.

Methods

Postmortem reports between January 2006 and May 2012 detailed 153 individuals who died by probable suicide, 58 of whom attended the mental health services. Relevant socio-demographic and clinical data was extracted from individuals’ lifetime case notes.

Results

Recurrent depressive disorder (44%) was the most common diagnosis and hanging the most common method of death (58%). Of individuals who died by hanging, 79% previously attempted suicide by the same method. For individuals with a documented history of depression, only 32% had antidepressants detected in their toxicology reports. Similarly, only one individual (20%) with schizophrenia had antipsychotics detected in their toxicology report.

Conclusions

Individuals who died by probable suicide, most commonly died by hanging and drowning; with previous attempts of hanging particularly prevalent in the group who subsequently died by hanging. At the time of death, less than one-third of individuals according to toxicology reports were taking the medication that was last prescribed to them by the mental health services suggesting a high rate of treatment non-concordance in individuals who died by probable suicide.

Can we risk recovery? A grounded theory of clinical psychologists' perceptions of risk and recovery-oriented mental health services

OBJECTIVES

This study sought to explore the views of clinical psychologists towards the concepts of 'risk' and 'recovery' and to set those views against the context of mental health services.

DESIGN

An exploratory, social constructionist grounded theory methodology was adopted.

METHODS

Eleven clinical psychologists working in adult mental health services each participated in one individual semistructured interview.

RESULTS

The clinical psychologists studied were aware of the emergence of recovery-oriented approaches, but felt unable to incorporate them in practice because of perceptions of being bound by both their own limitations and those of their circumstances, including issues of risk, thus giving rise to dilemmas in professional practice. Narrow definitions of risk as equated to danger dominated over broader conceptualizations of risk with positive consequences. The existing culture of mental health services was seen as emphasizing the need to avoid harmful consequences of taking risks, which in turn was seen to limit innovations in implementing recovery-oriented approaches.

CONCLUSIONS

Participants' ability to work in a recovery-oriented manner seemed to be limited by the way in which services perceived and responded to risk. Participants did not discuss risks arising from stigma, social exclusion, racism, sexism, or iatrogenic effects of psychiatric treatment. Narrow conceptualizations of risk as related to harm and danger seen in this study contribute to a sense of needing to be risk averse. However, the implications for practice included ideas about what might increase the possibilities for adopting recovery approaches across disciplines.

Characteristics of suicide attempts in young people undergoing treatment for first episode psychosis

OBJECTIVES

Understanding the characteristics of suicide attempts in people undergoing treatment for first episode psychosis (FEP) may have implications for risk management at a service level and local suicide prevention strategies. Although studies have focused on identifying individual-level risk factors for suicide attempts in this patient group, none have yet conducted an in-depth profile of suicide attempts. The aim of the present study was to examine the characteristics of suicide attempts in young people during the initial 18 months of treatment for FEP.

METHOD

A retrospective medical record audit study of a cohort of patients accepted for treatment at a specialist FEP service between 1/12/2002 and 30/11/2005.

RESULTS

Of 607 patients, 73 (12%) attempted suicide during treatment. Of these 73, most (72.6%) attempted suicide on one occasion. The majority of attempts (85.3%) occurred when patients were treated as outpatients and were in regular contact with the service. Suicide attempts tended to be impulsive (77.6%), triggered by interpersonal conflict or distress due to psychotic symptoms. Two thirds involved self-poisoning, usually by overdose of prescribed medications. All inpatient suicide attempts were by hanging or strangulation. Individuals infrequently sought help immediately before or after the attempt; if help-seeking occurred, informal sources of support were contacted.

CONCLUSIONS

To reduce the number of suicide attempts among individuals treated for FEP, psychiatric services could consider: restricting the amount of medication prescribed per purchase; individualised suicide risk management plans for all newly admitted patients, including those who do not appear to be at risk; stringent reviews of inpatient psychiatric units for potential ligature points; providing information and psycho-education for significant others in recognition and response to suicide risk; fostering patients' problem solving and conflict resolution skills; and regular risk assessment and close monitoring of patients, particularly during the high risk period of 3 months after a suicide attempt.

Economic factors and suicide rates: associations over time in four countries

OBJECTIVE

Suicides account for more than 30,000 deaths per year in the US alone. Suicide rates change over time, and the factors influencing them remain poorly understood. Economic factors, in particular unemployment, have been suggested as a major influence. However, the evidence for this has been inconsistent, which may be partly explained by shortcomings of the statistical methods used.

METHODS

Time series analytical techniques (unit root and co-integration tests) were applied to test the associations over time between economic factors, i.e. unemployment, real gross domestic product per capita (RGDP) and the consumer price index (CPI) and death rates by suicide as collected by national agencies in the UK (1901-2006), US (1900-1997), France (1970-2004) and Italy (1970-2001). Traditional correlation analyses were used when appropriate.

RESULTS

Co-integration and correlation tests showed a long-run association between economic factors and suicide rates. Increase/decrease of unemployment predicted an increase/decrease of suicide rates over long historical periods and in different nations. RGDP and the CPI were also linked with suicide rates, but this was not consistently so and the direction of the association varied.

CONCLUSIONS

Unemployment is a major factor influencing suicide rates over long periods of time and in different national contexts. It needs to be considered as a confounding factor in evaluations of suicide prevention strategies.

Suicide trends in an expanding mental health service in Auckland

OBJECTIVE

Suicide is a global health priority. It is important to analyse the effects of investment in mental health services on suicide rates. This paper presents and discusses recent changes in suicide rates and diagnostic mix among clients of the mental health service in Auckland during a time of service growth.

METHOD

Details of suicides among current clients were analysed for the years 1993 to 2006, comparing the periods 1993-1999 and 2000-2006. These periods were compared with the local population and national figures.

RESULTS

The local population suicide rate decreased significantly with the rate among clients increasing slightly so the proportion occurring among clients increased (up to 31%). There was an increase in non-psychotic diagnoses.

CONCLUSIONS

As access to services expands the proportion of suicides that occur in known clients will paradoxically increase. Service expansion may have contributed to the falling population suicide rate in Auckland.

A comparison of suicide and undetermined deaths in Cornwall across national and local agencies

We looked to evaluate the reporting pathways and statistics between relevant stakeholders for reporting suicides and undetermined deaths in Cornwall. We checked on the correlation of the numbers provided for suicides and open verdicts given by different national and regional agencies including the local Coroner's office, Office of National Statistics (ONS), Cornwall Partnership NHS Trust (CPT) and The National Confidential Inquiry (NCI) into suicides for Cornwall. We examined the data of suicides and open verdicts presented by these agencies between the years 2000 and 2002. Our study shows that the data transfer between the relevant agencies (local Coroner, ONS, CPT and NCI) can be inaccurate. These inaccuracies not only prevent establishment of clear records but could also impact on policy decisions. It is also important to recognize that NHS trusts involved in mental health are being evaluated on their record on suicide prevention which in some senses is the most important outcome measure, thus making it important for inspecting bodies to look out for and take account of different figures from different agencies.

Are psychiatric wards becoming safer?

Recent increasing recognition of the global burden and extent of suicide has led to worldwide suicide prevention strategies, such as the National Suicide Prevention Strategy for England (Department of Health, 2002). This identified those in contact with mental health services as a priority group and evidence indicates that psychiatric inpatients are at particular risk (Qin and Nordentoft, 2005). Inpatients have been identified as a priority group by the National Confidential Inquiry (Appleby et al, 2006). Given the close proximity to services, they are clearly a group who are potentially amenable to intervention.