Evaluation of official procedures for suicide prevention in hospital from a forensic psychiatric and a risk management perspective

Background: Suicide is a severe public health problem, in 2008 the Italian ministerial recommendation n 4 on the management of suicide defined key areas for the identification of suicidal risk in hospital wards. The guidelines are important in defining professional liability issues, in line with Law 24 of 8/3/ 2017 ‘Gelli-Bianco’. Our study aimed to investigate the appropriateness of the official documents on suicide prevention delivered by Italian hospitals and their compliance with the ministerial recommendation. Methods: The Italian hospitals’ public procedures on suicide prevention issued between 2008 and 2019 (n1⁄4 33) were retrieved thorough web search and further evaluated according to their compliance with the 2008 Italian ministerial recommendations. Results: The guidelines documents were generally in line with the ministerial recommendation. However, we found a lack of implementation in the specific training of health professionals. Most guidelines provided no risk stratification, nor specific procedures for different risk degrees or diagnoses. More than half of the documents did not report standardised tools for the assessment of suicidal risk. Conclusions: The public procedures on suicide prevention in Italian hospitals present general indications, leaving room for interpretation. Public procedures should be implemented with greater attention to the elements of judgement in the assessment of suicidal risk.


Introduction
The suicide of a patient in a hospital ward poses notable problems and challenges. Besides the emotional impact on relatives and friends (Bellini et al. 2018), suicide is a stressful event that creates burden also on the healthcare workers, with possible consequences at managerial and organisational levels in the care team (Kaye and Soreff 1991;Knoll 2012). Inpatient suicide occurring in the hospital is considered an adverse event, and also carries a high probability of generating professional liability issues (Franchitto et al. 2007).
In 2008, the Italian Ministry of Health issued recommendation 4 ('Prevention of suicide of patients in hospitals') (Ministero della Salute, 2008), that provided measures a nd training processes to be applied in case of suspect suicidality and indicated the structural characteristics to be respected for suicide prevention in hospitals. This recommendation constitutes the most authoritative directive to date for the management of suicide risk in Italian hospitals. Italian hospitals were required to transpose the ministerial recommendation, by implementing their own official documents (e.g., guidelines, recommendations, protocols, regulations) fitted on the bases of local features. After tracking sentinel events for eighteen months in the National Health System in Italy, Cardone et al. (2009) noted that suicide was the most common sentinel event. When analysing causes and contributing factors of suicides, the authors noted a substantial lack of official documents and appropriate procedures for the management of suicidal risk in 34% of cases, thus proposing the development of recommendations as a possible solution.
The Italian Ministry of Health's implementation of a monitoring system for the reporting of sentinel events (Information System for the Monitoring of Health Errors -SIMES) highlighted suicide as the second reported adverse event in hospitals. The fifth ministerial report on the monitoring of sentinel events, analysed the data from 2005 to 2012, and reported 295 cases of suicide or attempted patient suicide in hospital, which corresponds to 15.4% of total sentinel events. Communication difficulties among personnel have also been acknowledged as a risk factor for inpatient hospital suicide, having been reported in 31% of cases.
Suicide is more prevalent in patients admitted for respiratory, cardiovascular, endocrine, hematological and renal diseases than in those patients admitted for a mental disorder (Matandela 2017). To illustrate this point, 40% of suicide cases in hospital occur in a medical unit, while only 5% in a psychiatric unit (Cardone et al. 2009).
In Italy most cases of suicide in hospital forecast the opening of a criminal proceeding against the medical staff, even if most of these cases do not come to trial. Even so, in some cases the doctors are alleged to have committed manslaughter or, in more recent years, for culpable liability for death or personal injury in the health setting. The cases are usually conducted in a civil court because, under Italian jurisdiction, the identification of criminal liability in a penal court would require more stringent evidence (Fiori and Marchetti 2009). Medical responsibility for patient suicide is frequently linked to a lack of surveillance conduct (Terranova and Sartore 2013). This juridical definition might recall anachronistic custodial aspects.
The global risk for professional liability proceedings in psychiatry is generally low (Martin-Fumad o et al. 2015), although in Italy several judgements of the Supreme Court have progressively extended the psychiatrists' duty of care. These judgements have defined new perspectives in the field of professional liability in psychiatry (Catanesi et al. 2016), and have increased the possible risk of legal proceedings following an adverse event, such as suicide of a patient or a violent act committed by a patient for whom one held a duty of care (Terranova and Rocca 2016).
The need for official documents adapted to actual clinical practice became even more crucial after the Italian government issued Law 8/3/2017 n. 24, also known as the Gelli-Bianco Law. This law poses the problem of occupational accidents, which include patient suicide, as part of the wider issue of the safety of care. Among the innovations introduced, Law 24/2017 established regional ombudsmen, as guarantors for the protection of the right to security of care, who should intervene, free of charge, at the direct request of the injured party, or his delegate.
The Law 24/2017 also established in each region a centre for the management of health risk and patient safety, an office responsible for collecting data on risks, adverse events and legal disputes from public and private health and social care. The data collected by the centres for the management of health risk and patient safety should be electronically transmitted on an annual basis to the newly-formed National Observatory of Good Practices on Health Safety. Each healthcare facility is then obliged to publish annually on its website a final report on the adverse events which have occurred, specifying their causes and any subsequent legal action. After acquiring the abovementioned data from regional centres, the National Observatory of Good Practices on Health Safety prepares guidelines and issues appropriate measures for the prevention of health risks, monitors good practices for safety and periodically updates the collected data.
Healthcare facilities also have an obligation of transparency for which they are obliged to accomplish the requests of those subjects who have the right and interest to obtain health documentation concerning the possible medical responsibility, within seven days, with possible integration within a maximum of thirty days. Healthcare facilities are also required to publish on their Internet website the data relating to compensation they have paid in the last 5 years. Article 5 of the Gelli-Bianco Law is relevant for professional liability issues as it states that health professionals shall comply, with the exception of specific cases, with the recommendations developed by scientific societies included in a special list of the Ministry of Health. Article 6 of the Gelli-Bianco Law provides that there is no criminal responsibility for the health professional if the event occurred due to his/her inexperience, as long as the guidelines had been followed.
In these cases, as stated by article 15 of Law 24/2017, every forensic evaluation must be performed by a medico-legal expert and a doctor with specific skills in the field. The law has provisions aimed at reducing civil disputes, including a mandatory preliminary mediation.
There is general agreement that the guidelines are an effective tool for establishing quality of care and patient safety and, that they play a significant role in establishing a standard of judgement in the evaluation of the conduct of health care professionals (Albolino et al. 2019). The guidelines set out decision points to be integrated with clinical judgement and, in a wider perspective, could become an element of health policy evaluation (Lohr et al. 1998).
Nonetheless, it can be argued that some problems still remain unsolved. The law provision for the establishment of national guidelines can be perceived as excessively centralising, especially in a country like Italy, where regional and sub-regional differences are significant. This is, however, in line with the policies, for example, of the essential levels of healthcare.
However, the point remains that the guidelines, have become the test bench for the professional evaluation of health workers and the Court of Cassation, with its judgement in Joint Sections of 21.12.2017 no. 8770, has reaffirmed their centrality.
We, therefore, wanted to verify the current state of free and transparent access to official hospital documents on the prevention of suicide in Italy and the appropriateness or otherwise of the same to the 'Recommendations' of the Ministry of Health of 2008.

Materials and methods
The official documents of the healthcare facilities, published between 1 January 2008 and 31 March 2019, were searched on the World Wide Web using the Google search engine. Researchers selected guidelines on the following predefined criteria: (1) the title includes the terms 'guideline' (linea-guida), 'company procedure' (procedura aziendale), 'prevention' (prevenzione) 'suicide' (suicidio) 'hospital' (ospedale) and the names of regions, and (2) guidelines were of public access.
On the basis of the ministerial recommendation N. 4 of 2008, we developed a survey sheet to be used to assess the compliance of the procedures we retrieved with the main requirements provided by the recommendations. The ministerial recommendation N. 4 of 2008 states that suicide of a patient in the hospital is a severe event, and its prevention is based on an appropriate clinical evaluation. Some areas inside the hospital are at a greater risk, including psychiatry, oncology, emergency room, obstetrics and common areas. An accurate personal history of the patient must be collected, especially previous suicide attempts, substance consumption and abuse, social and working characteristics, patients' statements concerning loss of hope and personal meaning, thought suicide and recent stressful life events. Special attention should be given to persons who have a clear psychiatric disorder, including delirium and other somatic diseases with cognitive or mental impact, to those who have suicidal ideation or a recent suicide attempt, or those with recent bereavement. Care includes psychological support, involvement of family members, friends and volunteers, adequate communication between staff members, and personalised pharmacological and non-pharmacological therapies. Those patients who are at-risk should not be left alone. The environment should be equipped with adequate protection measures, which are listed in the ministerial recommendation. A procedure for communicating the death of the patient at the family members must be established by the hospital. Continuous education of the staff is considered paramount. The recommendations also include a list of questions that may be included in the clinical examination (Guide for the interview of the patent), depending on the specific situation.
The presence/absence of the following parameters was evaluated: a. Full patient's history: family history, previous suicide attempts, suicide ideation, a questionnaire proposed by the Ministry ('Guide for the interview with the patient') included in the recommendation, concomitant diseases, socio-cultural aspects, substance abuse, status. b. Care pathways: use of psychological and psychiatric counselling, involvement of general practitioners, specialists, families, volunteers, adequate communication, personalised therapies, protected discharge, hospital referent and connection with local services. c. Organisational processes: information for health personnel, specific procedures based on guidelines, transfers, patient surveillance, supervision. d. Structural characteristics: security devices, security frames, equipment that does not allow improper use, measures aimed at limiting access to lethal means. e. Training: further staff training for both patients understanding and safety measures with periodic updating.
Each document included in the study was evaluated for each of the 5 points listed above, assessing whether they were present, mentioned without adequate explanation or implementation (partial compliance), or absent.
Two examiners who competed the ad-hoc survey sheet independently evaluated the selected documents. The coefficient of concordance in the evaluation proved to be over 95% on all the selected items.
Twenty-one documents presented complete compliance with the 2008 ministerial recommendation, n ¼ 11 had partial compliance and only 1 was non-compliant. The methods of collecting the full patient's history were provided in a specific form in 25 procedures. In 6 procedures the need for collecting anamnesis was indicated but it was not specified the type of data to be collected, while in two cases it was not indicated at all.
The structural characteristics were specified in 28 public guidelines. In 3 cases they were indicated but not specified, and in 2 cases they were not indicated. The organisational processes for patient management were adequately and specifically indicated in 27 cases. In 5 cases they were not indicated, and in 1 case they were indicated, but not specified. The need for implementation of staff training was absent in 12 public documents, present although not further specified in 7 public documents, and specified and implemented in 14 public documents.
Risk stratification was indicated, with a corresponding assessment tool, in 5 cases, was merely mentioned in 5 cases, and was not present in 23 cases. Only seven have different procedures depending on the risk level. One procedure was stratified by diagnosis, one cited stratification by diagnosis without giving any indication, and 31 had no stratification by diagnosis.
The ministerial recommendation of 2008 does not report the need to use tools for the assessment of suicidal risk which accordingly are not provided by 22 of the guidelines that we examined. In the remaining 11 guidelines, 21 different instruments were proposed, with no specifications on their validity.
The need for a patient's daily observation was provided in 14 of the documents examined. The reference bibliography was usually the same reported in the ministerial recommendation. In 31 public documents there were no indications on how to deal with the media in case of patient suicide. In 28 cases there were indications of possible professional responsibility profiles. A connection with local services was present in 25 documents, while it was absent in 8 cases. Tables 1 and 2 resume the findings.

Discussion
The Italian procedures for patient suicide in hospital that we examined appear to be substantially in line with the ministerial recommendation. Nonetheless the lack of a need for a specific training of health professionals is a shortcoming, as more than half of the examined documents do not provide for it, nor do they specify how it should be implemented. This result deserves attention especially since the Italian civil jurisprudence has often affirmed the concept of team responsibility. The Gelli-Bianco Law focuses on the responsibility of health workers, a concept also taken from the judgement of the Joint Sections of the Supreme Court No. 8770 of 21.12.2017. The involvement of nurses, psychologists and psychiatric rehabilitators has become central and it seems necessary to adapt the hospital documents to this perspective.
The substantial conformity of the documents examined with the ministerial recommendation must however be compared to the absence of specific implementations for each healthcare facilities. A possible criticality lies in the failure of the ministerial recommendation to provide indication on how to apply them to the local settings, or to address the specific territorial and assistance needs. The health authorities seem to have not taken the chance  25  6  2  Structuralcharacteristics a  28  3  2  Organisational processes a  27  5  1  Staff training a  14  7  12  Risk stratification b  5  5  2 3  Stratification by diagnosis b  1  1 3 1 a Indications provided by the ministerial recommendation. b indications deemed necessary by the authors. to implement regional variances to the ministerial recommendation, proposing for almost all the guidelines the same issues as the above-mentioned document. Another issue from a medical legal perspective is that only one-third of the analysed procedures envisage the use of standardised tools for suicide risk assessment, while the others only provide the anamnestic interview that is attached to the ministerial recommendation, a method that is flawed by intrinsic limits when assessing suicide risk. A further evidence that is clinically controversial (Roaten et al. 2016;Stanley et al. 2019) but is relevant for forensic psychiatric purposes, is the lack of risk stratification, which is provided for in only one-third of the selected documents. Moreover, out of 10 guidelines that provide for a stratification of risk, only 7 identify a specific pathway according to the severity of the risk.
Even more problematic is the lack of procedural differentiation by patients' diagnosis. In the field of forensics, the suicide of a person suffering from a mental disorder that implies a higher risk of suicide, such as borderline personality disorder or bipolar disorder, is an element of differentiation of judgement. The diagnosis is a crucial step in medical care and decisions. Producing procedures that do not focus on the different needs and risks among diagnostic groups is a practice that deserves further study. A possible explanation for the omission of diagnosis-related procedures might reside in the attempt to reduce stigma.
The lack of differentiation by diagnosis means that specific pharmacological therapies are not indicated for certain diagnostic categories. For example, there is extensive bibliography on the usefulness of lithium as an antisuicidal drug in affective disorders, especially in bipolar disorder (among many: Benard et al. 2016, Haußmann et al. 2016Smith and Cipriani 2017;Tondo and Baldessarini 2018). In terms of forensic perspective and considering the increasingly complex international literature, it is peculiar not to emphasise a differentiation of suicidal risk based also on diagnosis and the corresponding specific indications of treatment.
Furthermore, suicidal risk is a usually dynamic phenomenon (Pompili et al. 2004;Solano et al. 2018) and requires repeated and constant evaluations, especially in the period following a possible suicide attempt or when suicidal ideation is particularly intense. When healthcare professionals come into contact with a potential suicide risk, there must be a need to stabilise the patient's clinical condition in order to define an effective improvement in the prognosis.
In this context, daily reassessment is certainly an important clinical and organisational aspect. However, only 14 of the 33 examined documents provided the need for daily assessment of patients. All the documents examined provide for the use of third-party care to be placed alongside the patient, in accordance with the ministerial recommendation. It is never specified how this third party should be identified, whether they should have specific expertise, or to what extent they should be informed of the condition of the person they are supervising. Moreover, no local public health company has implemented the possibility of using information technology to follow patients once discharged, a mode of protection that seems promising (Falcone et al. 2017).
It should be noted that two documents examined determine the need for a stratification of risk and describe the processes of patient management in detail, indicating times of greatest risk, specific allocation of rooms, requirements for staff who carry out surveillance and psychiatric advice.
The lack of focus by the documents on how to consider the method used in previous suicide attempts is another issue that deserves attention. A patient's risk profiles and implications in his clinical management are evidently different for a patient who has drunk bleach and has gastro-oesophageal lesions, for example. They also do not provide guidance on how to assess the level of environmental support, the degree of awareness of disease and the assessment of patients' capacity to provide informed consent to treatment (Mandarelli et al. 2014), represent two other aspects worthy of consideration. We believe this because in judicial proceedings, psychiatrists can be considered responsible for not having implemented coercive measures, in particular compulsory treatments, and there is usually a lack of effective documentation about the person and his ability to express an autonomous decision-making and competency.
From an organisational point of view, no hospital guide documents indicate how to enter the hospital with a rapid assessment of suicidal risk and attribute a specific emergency code already in the triage procedure. There are also no plans to provide education to the patient and his family.

Conclusions
The documents examined are almost formally in line with the 2008 ministerial recommendation. On the other hand, on a deeper examination, they are usually generic and leave wide margins of risk for users and, consequently, also for the care teams. The introduction of the Gelli-Bianco Law places the development of appropriate and evidence-based documents at the centre of risk prevention.
The risk of committing suicide, even though it is a substantially unpredictable condition for which the forecasting capabilities are limited (Franklin et al. 2017;Belsher et al. 2019) can be managed in a rational and professional manner once the risk factors have been effectively identified. The hospital guide documents in this field should more rigorously respect the possible elements of judgement within the decision-making chain of these complex cases, possibly including an assessment of the ability to give consent to treatment. The total number of procedures which were public available online was limited, this can be considered a limit of the present study due to the possible incomplete finding of documents as we had no access to intranet documents. If this represents a limit, the perspective of transparency required by the Gelli-Bianco Law justifies this type of choice.
Since the suicidal risk assessment and prevention in hospital and its possible consequences in terms of professional liability are central in the clinical and forensic psychiatry practices, it would be necessary to compare international guidelines on such topics. It would be also useful to guarantee international shared standards to allow equality in access and right to treatment, as well as freedom of choice, also for those patients who are at risk of suicide.

Disclosure statement
In accordance with Taylor & Francis policy and my ethical obligation as a researcher, I am reporting that I and the other researches involved are not members or consultants of the public health companies whose suicide guidelines have been evaluated in this study, nor we have any criminal o civil proceedings pending with any of these public health companies.