|Year : 2020 | Volume
| Issue : 3 | Page : 156-163
The Development of a Fuzzy Logic Model-Based Suicide Risk Assessment Tool
Fatma Ayhan1, Besti Üstün2, Türker Tekin Ergüze3
1 Batman University, Health Sciences School, Psychiatric Nursing Department, Batman, Turkey
2 Faculty of Health Sciences, Psychiatric Nursing Department, Uskudar University, Istanbul, Turkey
3 Department of Software Engineering, Uskudar University, İstanbul, Turkey
|Date of Submission||29-Oct-2020|
|Date of Decision||11-Nov-2020|
|Date of Acceptance||13-Nov-2020|
|Date of Web Publication||25-Dec-2020|
Department of Psychiatric Nursing, Batman University, Health Sciences School, Batman 72200
Source of Support: None, Conflict of Interest: None
Aim: The purpose of the research was to develop a fuzzy logic model-based risk assessment tool and to determine the views of health professionals working in community mental health centers concerning this. Introduction: Retrospective and psychologic autopsy researches have shown that a detectable psychological disease is available in at least 90% of all completed suicides. A methodological method was employed in developing the fuzzy logic model-based risk assessment tool, and the qualitative research method was used to elicit attitudes toward it. Materials and Methods: In the first stage, the suicide risk assessment tool was developed. In the second stage, the health professionals used the suicide risk assessment tool for 4 weeks. Feedback regarding health professionals' opinions of the suicide risk assessment tool was then elicited handling a semi-structured reportage form, and data were investigated handling descriptive analysis. Statistical Analysis: Health professionals reported powerful aspects making the suicide risk assessment tool a functional, practical, comprehensive, and highly applicable guideline. Results: Participants also recommended the addition to the results screen of a chart showing the course of the suicide risk assessment and that the reliability and validity of the tool be confirmed. Conclusion: In conclusion, health professionals expressed positive opinions regarding the scope of the suicide risk assessment tool and its measurement of that risk. Following confirmation of its reliability and validity, it may be useful for the fuzzy logic-based suicide risk assessment form to be integrated into and applied in the community health centers system.
Keywords: Community mental health, fuzzy logic, suicide, suicide risk assessment
|How to cite this article:|
Ayhan F, Üstün B, Ergüze TT. The Development of a Fuzzy Logic Model-Based Suicide Risk Assessment Tool. J Neurobehav Sci 2020;7:156-63
| Introduction|| |
Around 800,000 people globally lose their lives because of suicide every year World Health Organization (WHO, 2018). In 2018, 3161 individuals in Turkey died from suicide. The World Health Organization (WHO) 2013–2020 Mental Health Action Plan, adopted by the World Health Assembly in 2013, targeted a 10% decrease in national suicide levels by 2020.
Retrospective and psychologic autopsy researches have shown that a detectable psychological disease is available in at least 90% of all completed suicides. Half to two-thirds of psychiatric diagnoses involve mood disorders. Other research findings concerning the relationship between suicide and mental disorders show that the highest suicide-related mortality rates occur in individuals diagnosed with substance misuse and eating disorders, and indicate moderate suicide levels for mood and personality disorders, and relatively low proportions for anxiety disorders. The risk of suicide also increases with the use of more than one substance, and in case of depression, eating disorders, and personality disorder. The groups with the highest risk of suicide are those with borderline personality disorder, and antisocial personality disorder.
Community Mental Health Centers (CMHCs) in Turkey provide psychosocial support services, follow-up, and treatment within the framework of a community-based mental health model for individuals with chronic mental disorders such as schizophrenia, other psychotic disorders, and bipolar disorder. Twenty to fifty percent of schizophrenia patients attempt suicide. Individuals with bipolar disorder also frequently attempt suicide during depressive attacks, and the risk of suicide increases still further in case of substance use disorder, anxiety disorder, eating disorder, and personality disorder accompanying bipolar disorder (Hansson et al., 2018). Due to their characteristics, it is particularly important to assess the suicide risks of the individuals being followed up by CMHCs.
There is no method capable of completely accurately showing the risk of suicide. However, the probability of risk can be predicted by means of risk assessment. Health professionals should trust their own clinical experience in risk assessment but must also confirm their assessments on the basis of evidence-based findings. Important factors determining suicidal behavior are interactions between suicide risk and protective factors. Factors protecting against suicide include a sense of belonging, a good cognitive level, useful and good developed coping capabilities and the fact that religious faiths disapprove of suicide, being married and having children, powerful family bonds, and positive supportive relations between the patient and clinician.
Factors reported to involve a high risk of suicide include suicidal ideations, preparatory actions, stressful life events and cognitive/emotional conditions, extensive suicidal behaviors and objective signs of suicidal attainment (such as specific planning, access to lethal means, and the occasion to set these into action), stressful life events admission to and discharge from hospital, important physical diseases, chronic painful diseases impairing functioning and external appearance, diseases rendering the individual dependent on others or affecting the ability to see and hear, and chronic diseases, hopelessness, and social isolation.
The risk of suicide is generally assessed using scales, but these may be limited in terms of the extent to which the assessment logic actually reflects human ways of thinking, the inclusion in the assessment process of health professionals' knowledge and experience and these being supported with proven data, and taking into account factors protecting the individual against suicide. The use of existing suicide risk assessment scales is, for this reason, insufficient in terms of health professionals confirming their own clinical experience with evidence-based information. A single numerical value obtained from scales may also result in important risk-related situations being missed. A risk evaluation and suicide risk assessment appear in the official CMHC directive, but no structured, objective measurement tool is employed, and risk assessment relies more on the individual's worker's own knowledge and experience.
One of the most appropriate procedures for assessing nonlinear, multi-dimensional, complex, and imprecise subjects such as suicide is fuzzy logic. This was invented in 1965 by Lotfi A. Zadeh under the name of fuzzy logic or the fuzzy setting theory. The greatest advantage of fuzzy logic is that it very closely matches human reasoning. In the classic set theory, also known as binary logic, an element either belongs to a set or else does not. However, there are several conditions, in which an element's membership of a set is uncertain or indefinite. These are frequently seen in the sphere of health. When classic logic is used in suicide risk assessment, binary results are elicited for the risk of suicide, in the form of low or high, or present or absent. The production of the absolute results regarding the risk of suicide can result in health professionals missing risky or borderline situations. In fuzzy logic theory, an element may have various degrees of membership or may belong to more than one set. The degree of membership of elements in a fuzzy set ranges between 0 and 1, and these values show partial membership of a set. For instance, when an individual's suicide risk is assessed using fuzzy logic, risk status may be present in a low-risk group with one particular degree of membership and at the same time in a high-risk group with another particular degree of membership. In this way, health professionals can see the extent of the individual's membership of different groups and can at the same time include their own knowledge and experimentation in the suicide risk assessment process. There is no tool and/or scale assessing the risk of suicide using the fuzzy logic method in Turkey. In international terms, the UK Galatean Risk and Safety Tool program is based on the fuzzy logic model.
If effective suicide risk assessment is not performed, the individual may experience various risks, ranging from mild damage to potentially fatal situations. This makes accurate risk assessment essential. The fuzzy logic-based suicide risk assessment tool developed in the scope of the present research is practical and user-friendly and may be expected to contribute to suicide risk assessment depending on health professionals' knowledge and experience. The aim of this study is to develop a fuzzy logic model-based suicide risk assessment tool and to establish the views of health professionals working in CMHCs regarding that tool.
| Materials and Methods|| |
Ethics committee approval
Approval for the research was granted by the Üsküdar University Non-Interventional Research Ethical Committee (No.B.08.6.YÖK.2.ÜS.0.05.0.06/2017/327).
Type of research
A methodological design was employed in the development of a fuzzy logic model-based suicide risk assessment tool, and a qualitative method was used to collect opinions concerning the tool developed.
The application was performed in two phases. In the first phase, a fuzzy logic model-based suicide risk assessment tool was developed, while in the second, user views concerning the tool were elicited.
- Stage 1: The development of a fuzzy logic model-based suicide risk assessment tool
The procedures performed during the development of the fuzzy logic model-based suicide risk assessment tool are listed, in order, below
- Determination of suicide risk factors: Evidence-based study findings in systematic reviews and meta-analyses were examined to identify factors increasing and protecting against the risk of suicide. Fourteen risk areas and eight protective factors were identified,,,
- The collection of specialist opinions to determine the effect levels of risk factors: Specialist opinions were elicited to determine the effect levels of factors increasing or protecting against the risk of suicide. Views were collected from 19 specialists working in psychiatric clinics (nine physicians, four nurses, two social services experts, and four psychologists). These specialists had been working in psychiatric clinics for between 10 and 42 years. They were asked to score each factor reducing or increasing the risk of suicide between “0 and 1” with 1% sensitivity. These numerical values were transferred onto Excel, and mean values were determined [Table 1]
- Calculating the risk of suicide with a fuzzy logic model: An Excel file based on probability density function was prepared to calculate Gaussian distribution for the calculation of suicide risk with a fuzzy logic model. The model functions through the formula:
when the risk assessment tool items increasing the risk of suicide are completed, the maximum possible score is 9.41 and the minimum score is 2.16. Risk levels are assessed as low, moderate, or high. The high-risk rate rises the closer scores approach to 9.41, the moderate risk rate the closer they approach to 5.78, and the low risk rate the closer they approach to 2.16. Factors protecting the individual against suicide on the risk assessment tool are evaluated separately from the fuzzy logic model. The software produces a result by adding the effect levels of each protective factor. The maximum possible score from the factors protecting the individual against suicide is 2.25, and the minimum possible score is 0.26. Higher scores indicate a greater level of protection against suicide. After determining an individual's suicide risk level, health professionals can form an opinion concerning suicide risk status in the light of their clinical knowledge and experience and of the score obtained from protective factors
- Conversion of the fuzzy logic model-based suicide risk assessment tool into an online format: The suicide risk assessment tool was converted into an online format to enhance various facilitating features, such as ease of access, maintenance of confidentiality, and remote access (www.ufalt.net)
- Stage 2: Determination of the views of health professionals working in CMHCs concerning the fuzzy logic model-based suicide risk assessment tool
- Sample: The research sample consisted of health professionals working in two CMHCs affiliated to the Istanbul Provincial Health Directorate and one affiliated to the Karaman Provincial Health Directorate. A social services specialist, two nurses and two psychologists were working in the first CMHC, and a psychiatrist, two nurses, a psychologist, a social services specialist and an ergotherapist in the second, and a psychiatrist, two nurses, a psychologist, and a social services specialist in the third. Fifteen health professionals in CMHCs agreeing to take part in the study were enrolled
- Data collection tool: Health professionals' opinions were elicited using a semi-structured reportage form developed by the authors. This consisted of two parts. The first section contained eight questions designed to elicit characteristics such as health professionals' age, sex, work experience, and receipt of occupational and risk assessment training. The second section contained one question regarding how health professionals' use of the suicide risk assessment tool affected their evaluations, and two others concerning the powerful aspects of the tool and those requiring further development
- Application: Health professionals in the three CMHCs were informed about the suicide risk assessment tool developed, which was made available for their use. They used the suicide risk assessment tool for 4 weeks. Feedback concerning their opinions of the tool was obtained through a semi-structured reportage form. Interviews were held face to face with each health professional, in their own offices, and lasted between 15 and 35 min. The interviews were also recorded
- Data reliability and validity: The principles of credibility, transmissibility, consistency, and confirmability in qualitative inquiry were employed in the establishment of data reliability and validity
- Data evaluation: The study data were subjected to descriptive evaluation. This involves data being summarized and interpreted according to previously determined themes.
| Results|| |
Fifteen health professionals were interviewed in the scope of the research. The nurses participating in the research were numbered from N1 to N6, the psychiatric specialists were numbered PS1 and PS2, the psychologists were numbered from P1 to P3, the social services specialists were numbered from SS1 to SS3, and the ergotherapist was coded Ergotherapist (ERG). The health professionals enrolled in the research and some of their characteristics are shown in [Table 2].
|Table 2: Characteristics of the health professionals who provided views concerning the suicide risk assessment tool|
Click here to view
The data obtained from the interviews were subjected to descriptive analysis, and themes and subthemes were identified. Themes refer to powerful aspects of the tool and those requiring development, while subthemes refer to scope, applicability, and suicide risk measurement.
Theme 1: Powerful aspects
The health professionals participating in the research reported powerful aspects concerning the scope of the fuzzy logic model-based suicide risk assessment tool, its applicability, and suicide risk assessment status.
Subtheme 1: Scope
The participants declared that the number and content of the questions in the tool were sufficient to prevent questions being missed in the assessment and for the consideration of protective factors: “PS 1: In my opinion, the scope is sufficient… I consider it ideal in terms of the number of questions… We must obtain maximum benefit with a minimum number of questions, particularly when interviewing cases of this type …” “N5: I think that the questions fully meet requirements…”
Subtheme 2: Applicability
Some health professionals reported that the applicability of the suicide risk assessment took was practical and facilitated suicide risk evaluation. The participants who described the developed suicide risk assessment tool as practical ascribed this to its not being time-consuming in nature, its being simple and very easy to apply, its being based on information elicited when speaking to the patient, its being capable of application within the interview, its containing few areas requiring interpretation, and its being useful in case of uncertain assessments and highly practical. The fact that it could be easily completed by anyone with psychiatric training was also described as a powerful aspect of the tool: “P1: …It is very simple. I mean it is very simple to apply. It is easy to apply it verbally… It did not take much time, and represents an excellent conversation in suicide assessment. It is something I can apply within the interview …” “SSS 4: …Our having such a form available may be something that will enhance the quality of our interviews with patients. I, therefore, consider it functional…”
The way the tool facilitated suicide risk assessment was an element emphasized by almost all participants: “PS1: It facilitates our work because the questions involve are basic parameters we need to know about the individual being assessed, whom we think maybe a suicide risk. You have in all likelihood selected the most rational or valuable items from among these, and this facilitates our work in that sense. In other words, it prevents us skipping any question and also considers protective factors right from the beginning, and that makes our job easier. It serves us as a guide.” “SSS1: The questions are definitely comprehensible. Patients definitely respond to these questions because there is nothing that any patient cannot understand.”
Other powerful aspects cited by health professionals include the easy accessibility of the suicide assessment risk tool due to its availability in an electronic environment, the fact it provides a numerical figure at the end of the assessment, that it permits more professional assessment, its enhancement of the task being performed, the fact it provides a common language, and the way it permits official documentation: “PS1: It made things easier. I can make a more professional assessment. In addition, it also enhanced the quality of my assessments.” “P1:… The risk assessment tool provides results along the lines of a 10% low risk, a 50% moderate risk, and a 20% high risk. This enables us to say something about the risk and contributes to the analysis.” “N5: I think this should be applied to all patients. Then we can proceed based on the results… There will still be situations that are missed. But at least it will have been applied. Then we will have documentation to say that “We applied it, and there was no such ideation then.”
Subtheme 3: Suicide risk measurement status
Health professionals reported that the assessment result they obtained using the suicide risk assessment tool was similar to that they estimated themselves, and that the tool provided clarity in case of uncertainty: “N4: The assessments are exactly what we think ourselves. For example, it provides immediate clarity about things we are uncertain of…”
One opinion to the effect that the suicide risk assessment tool is consistent and suited to evidential study was expressed as follows: “P2:… In my view, the system is consistent internally and with its results. Of course, something more empirical would be needed to determine how useful this is. But I think the results are consistent and suitable for use …”
Theme 2: Aspects requiring development
Subtheme 1: Applicability
In terms of the applicability of the suicide risk assessment scale, some participants stated that the application of the tool could be enhanced if it were completed by a counselor well acquainted with the patient, if it were integrated into the system employed in CMHCs, and if graphics were added to the result screen: “P2: The risk assessment could easily be completed if the person applying the test is sufficiently acquainted with the patient.” “PS1: …if it were integrated into systems used in our daily procedures, for example, there is a program on the web containing the forms used in our CMHCs, and if it were integrated with that, then we could apply it very quickly…Graphics could be added to the results screen, and every graphic we produce could be given separately when we enter the patient's file. Let us assume that we apply this scale every 2 weeks, if there were a graphic showing the course, then that would be very useful very quickly.”
| Discussion|| |
The views of health professionals in CMHCs regarding the fuzzy logic model-based suicide risk assessment tool are discussed below under the themes identified.
Theme 1: Powerful aspects
Participants stated that the items in the suicide risk assessment tool can ensure that various questions that should be asked during the assessment are not overlooked. They also considered that it contains questions that assess existing emotions, ideas, and behaviors that can affect the individual's suicide risk while considering dynamic or state changes, that the tool can provide a common language. Participants also felt that the calculation of a number at the end of the analysis permits a more professional assessment, that the tool professionalizes application, is very simple and functional, and also permits official documentation. Reasons cited in studies for health professionals expressing positive views of risk assessment tools include the presence of questions that are very useful in evaluating previous psychological history and that might not otherwise come to mind, their facilitation of discussions about risk on the part of care providers, and their facilitation of professional decision-making. Causes for criticism of risk assessment tools include lack of attention to dynamic or state changes since they tend to focus on previous (static) risk factors, and their inability to consider individual factors concerning the patient, their being highly mechanical and behaviorally reductive, dehumanizing, and their failure to include professional intuition in the assessment process. We attribute the generally positive opinions of our suicide risk assessment tool on the part of participants to the questions evaluating both the individual's previous (static) and variable (dynamic) characteristics. We also think that participants approved the tool due to its providing information about risk group membership percentages as a result of the fuzzy logic model employed, rather than providing definite values as in classic logic, thus including health professionals' views and experience in the decision-making process.
In terms of the applicability of the suicide risk assessment tool, the participants stated that it contained questions requiring consideration and professionalized application, was highly functional and easy to administer, and permitted official documentation. The provision of official documentation is also cited as the most important reason for positive attitudes toward the using of risk assessment tools on the part of health professionals in previous similar research.,, Some researches of attitudes toward risk assessment tools have reported that health professionals are undecided regarding the use of such tools, and that they may regard them as bureaucratic instruments of no value or purpose or else as psychiatric technology developed to reduce the role of clinical expertise., Studies have also suggested that health professionals consider them to be useful guidelines capable of informing and guiding less able or less experienced applicators, and that they would employ a risk assessment tool if it was appropriate and user-friendly. We attribute the positive approach to the application of the risk assessment tool in the present study to the fuzzy logic model, an artificial intelligence program, very closely reflecting human thinking, and to participants being aware of the need for a suicide risk assessment tool.
While the result yielded by the tool was consistent with health professionals' own estimations of the individual's risk status, they also reported positive views concerning the fact that it additionally provided clarity in uncertain situations. More than one in four of the nurses participating in Holley et al. research believed that risk assessment tools were capable of predicting the probability of risk, but they also did not object to the statement that risk could not be predicted.
Theme 2: Aspects requiring development
The health professionals participating in this research also recommended that the suicide risk assessment tool be integrated with the CMHC system and suggested that a graphic demonstrating the route of the patient's suicide risk assessment on the outcome monitor would also be useful. This finding shows that the participants in our study held positive opinions regarding the routine use of the suicide risk assessment tool. In contrast to these findings, one study involving community mental health nurses reported that some participants felt that the using of a standard risk assessment tool who a totally unnecessary bureaucratic imposition, while the majority felt that imposing the use of a standardized risk assessment tool would act as an obstacle to creativity in their work. We think that the integration of our suicide assessment risk tool into the CMHC system will not create an additional workload since it will become part of existing practices. The addition to the results screen of a graphic demonstrating the route of the patient's suicide risk assessment will also be useful in terms of evaluating change in the patient's suicide risk.
Longitudinal researches are needed to define the effectiveness of the suicide risk assessment tool developed in this study in measuring suicide. The lack of data regarding the validity of the scale is the limitation of the study.
| Conclusion|| |
The health professionals in this research held positive views concerning the scope of the suicide risk assessment tool we developed. In the light of health professionals' views regarding the questions in the suicide risk assessment tool and their suggestions for how it might be improved, we may conclude that some aspects require further development.
Patient informed consent
Informed consent was obtained.
Ethics committee approval
Approval for the research was granted by the Üsküdar University Non-Interventional Research Ethical Committee (No. B.08.6.YÖK.2.ÜS.0.05.0.06/2017/327).
Financial support and sponsorship
No funding was received.
Conflicts of interest
There is no conflict of interest to declare.
Author contribution area and rate
Fatma AYHAN (%45), Besti ÜSTÜN (%35) and Türker
Tekin ERGÜZEL (%20) contributed to the conception and design of this study.
Fatma AYHAN (%45) and Besti ÜSTÜN (%35) performed the descriptive analysis and drafted the manuscript.
Türker Tekin ERGÜZEL (%20) prepared the fuzzy logic software and critically reviewed the manuscript and supervised the whole study process. All authors read and approved the final manuscript.
| References|| |
Conwell Y, Duberstein PR, Cox C, Herrmann JH, Forbes NT, Caine ED. Relationships of age and axis I diagnoses in victims of completed suicide: A psychological autopsy study. Am J Psychiatry 1996;153:1001-8. https://doi.org/10.1176/ajp.153.8.1001
Herrera M. Mood disorders and suicide. In: Falcone T, Mitchell TJ, editors. Suicide Prevention a Practical Guide for the Practitioner. US: Springer International Publishing; 2018. p. 23-53.
Malone KM, Oquendo MA, Haas GL, Ellis SP, Li S, Mann JJ. Protective factors against suicidal acts in major depression: reasons for living. Am J Psychiatry 2000;157:1084-8. https://doi.org/10.1176/appi.ajp.157.7.1084
Harris EC, Barraclough B. Suicide as an outcome for mental disorders. A meta-analysis. Br J Psychiatry 1997;170:205-28.
Soloff PH, Lis JA, Kelly T, Cornelius J, Ulrich R. Risk factors for suicidal behavior in borderline personality disorder. Am J Psychiatry 1994;151:1316-23.
Black DW, Blum N, Pfohl B, Hale N. Suicidal behavior in borderline personality disorder: prevalence, risk factors, prediction, and prevention. J Personality Disord 2004;18:226-39. https://doi.org/10.1521/pedi.18.104.22.168445
Lu L, Dong M, Zhang L, Zhu XM, Ungvari GS, Ng CH, et al
. Prevalence of suicide attempts in individuals with schizophrenia: A meta-analysis of observational studies. Epidemiol Psychiatr Sci 2019;29:e39. https://doi.org/10.1017/S2045796019000313
American Psychiatric Association. Practice guidelines for the assessment and treatment of patients with suicidal behaviours. Am J Psychiatry 2003;160:1-60.
Fawcett J. Depressive disorders. In: Simon RI, Hales RE, editors. Textbook of Suicide Assessment and Management. Washington, DC: American Psychiatric Publishing; 2006. p. 255-75.
Joiner TE, Hollar D, van Orden K. On buckeyes, gators, super bowl Sunday, and the miracle on ice: pulling together is associated with lower suicide rates. J Soc Clin Psychol 2006;25:179-95. https://doi.org/10.1521/jscp.2006.25.2.179
Samra J, White J, Goldner E. Working with the Client who is Suicidal: A Tool for Adult Mental Health and Addiction Services. Canada: Library and Archives Canada Cataloguing in Publication Data; 2007.
Rudd MD, Berman AL, Joiner TE, Nock MK, Silverman MM, Mandrusiak M, et al
. Warning signs for suicide: Theory, research, and clinical applications. Suicide Life Threat Behav 2006;36:255-62. https://doi.org/10.1521/suli.2006.36.3.255
Walsh G, Sara G, Ryan CJ, Large M. Meta-analysis of suicide rates among psychiatric in-patients. Acta Psychiaty. Scand 2015;131:174-84. https://doi.org/10.1111/acps.12383
Chattopadhyay S, Pratihar DK, De Sarkar SC. Statistical modelling of psychoses data. Comput Methods Programs Biomed 2010;100:222-36.
Zadeh LA. Fuzzy algorithms. Informat Control 1965;12:94-102.
Buckıngham CD. Psychological cue use and implications for a clinical decision support system. Med Informmatic 2002;27:237-51. DOI: 10.1080/1463923031000063342.
Franklin JC, Ribeiro JD, Fox KR, Bentley KH, Kleiman EM, Huang X, et al
. Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychol Bull 2017;143:187-232. doi: 10.1037/bul0000084.
Ribeiro JD, Franklin JC, Fox KR, Bentley KH, Kleiman EM, Chang BP, et al
. Self-injurious thoughts and behaviors as risk factors for future suicide ideation, attempts, and death: A meta-analysis of longitudinal studies. Psychol Med 2016;46:225-36. https://doi.org/10.1017/S0033291715001804
Runeson B, Odeberg J, Pettersson A, Edbom T, Adamsson JI, Waern M. Instruments for the assessment of suicide risk: A systematic review evaluating the certainty of the evidence. Plos One 2017;12:1-13. doi: 10.1371/journal.pone.0180292.
Muir-Cochrane E, Gerace A, Mosel K, O'kane D, Barkway P, Curren D, et al
. Managing risk: Clinical decision-making in mental health services. Issues Mental Health Nurs 2011;32:726-34. https://doi.org/10.3109/01612840.2011.603880
Holley J, Chambers M, Gillard S. The impact of risk management practice upon the ımplementation of recovery-oriented care in community mental health services: A qualitative investigation. J Mental Health 2016;25:315-22. https://doi.org/10.3109/09638237.2015.1124402
Doyle M, Dolan M. 'Violence risk assessment: Combining actuarial and clinical information to structure clinical judgements for the formulation and management of risk'. J Psychiatric Mental Health Nurs 2002;9:649-57. https://doi.org/10.1046/j.1365-2850.2002.00535.x
Gerace A, Curren D, Muir-Cochrane E. 'Multidisciplinary health professionals' assessments of risk: How are tools used to reach consensus about risk assessment and management? J Psychiatric Mental Health Nurs 2013;20:557-63. https://doi.org/10.1111/jpm.12026
[Table 1], [Table 2]