- Jane L. Pearson, Ph.D.,
- National Institute of Mental Health
- Barbara Stanley, Ph.D.,
- Columbia University
- Cheryl King, Ph.D.,
- University of Michigan
- Celia Fisher, Ph.D.,
- Fordham University
Prepared under NIMH Contract Numbers 263-MD 004930 (Dr. Stanley), 263-MD-004928 (Dr. King), 263-MD-004929 (Dr. Fisher). Jane Pearson, Project Officer. January 2001.
- Current Federal Initiatives
- Considertion of Safety Efforts in Tandem with Research Evidence
Factors Associated with Increased Risk for Suicidal Behavior
- Design Considerations
- Treatment Comparison Conditions
- Treatment commensurate with Study participant Risk Status
Points to Consider in Planning an Intervention Trial with Suicidal Study Participants
- Monitoring and Risk Management Protocols
- Risk Management Protocols
- Increased Monitoring and Supervision
- Research Clinician Competencies
Legal Risk to Investigators and Institutions Conducting Research with Persons at High Risk for Suicidal Behavior
Purpose: In 1997 approximately 30,000 people died by suicide in the United States, making suicide the 8th leading cause of death in the U.S. (Hoyert, Kochanek, & Murphy, 1999). There are an estimated 8 to 25 attempted suicides for every completion. Persons with mental disorders are at increased risk for suicidality and death by suicide. NIMH is providing this guidance for those investigators conducting research on interventions to reduce suicidality, as well as for investigators likely to encounter persons at risk for suicidality in intervention trials involving persons with mental disorders. This document focuses on issues most relevant to adult study participants. For study participants considered special populations, such as children and prisoners, see information about required additional safeguards and procedures at http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.html.
In June 1998, the National Institutes of Health (NIH) issued a policy on data and safety monitoring (http://grants.nih.gov/grants/guide/notice-files/not98-084.html ) that requires oversight and monitoring of all intervention studies to ensure the safety of participants and the validity and integrity of the data. The policy further elaborates that monitoring should be commensurate with risks and with the size and complexity of the trials. The NIH already requires data and safety monitoring, generally in the form of the Data and Safety Monitoring Boards (DSMBs) for phase III clinical trials. Beginning with the October 2000 receipt date, investigators must submit a monitoring plan for phase I and II clinical trials to the funding Institute and Center (IC) before the trial begins (http://grants.nih.gov/grants/guide/notice-files/NOT-OD-00-038.html ). All institutions carrying out an NIMH-funded phase I or II clinical trial must establish a data monitoring system, which may include a DSMB. Interventions focused on reducing suicidality are considered high risk by NIMH.
The intent of instituting a DSMB is to provide oversight and monitoring of an intervention study to ensure the safety of participants and validity and integrity of the data. This document describes the safety challenges and potential solutions for consideration by researchers conducting intervention research involving persons that are, or may become suicidal. They are intended to be used by investigators in the development of intervention trials and by the Institutional Review Boards (IRBs) and DSMBs that must review and monitor these studies. The intent is to propose ways to minimize risk in approaches to conducting research on suicidality, in order to assist in facilitating research on the nature of suicidality and on the efficacy of interventions proposed to reduce suicidality. Interventions of interest include psychosocial, pharmacological, somatic, and their combination, or system interventions that change typical practice or treatment. The approaches reviewed here include study design considerations, increased monitoring, and implementation of risk management protocols. A checklist of informed consent issues is also included.
Definitions: Suicidality and suicidal behaviors includes completed suicide, attempted suicide, and suicidal ideation (each defined below). Completed suicide is considered a serious and unexpected adverse event; suicidal ideation in certain samples (e.g., major depression) is less likely to be considered so. Whether attempted suicide and suicidal ideation are considered unexpected adverse events will depend on the particular trial. As described by NIH guidelines
Federal regulations (45 CFR Part 46, Subpart A), shared by 17 Departments and Agencies as the Common Rule, require written procedures and policies for ensuring reporting of "unanticipated problems" involving risks to participants to the IRB, appropriate institutional officials, and the Department or Agency Head. Generally, the funding Institutes and Centers establish operational definitions of adverse events that apply to the particular trial. In the case of study participants who are suicidal, the management of serious suicide attempts and facilitation of inpatient hospitalization should be addressed and planned for. In working with their IRB, funding agency, and DSMB when appropriate, investigators should describe how likely serious attempts are to occur, how frequent inpatient hospitalization may be, and their protocols for managing these events.
Completed suicide refers to death from self-inflicted injury where there is evidence that the decedent intended to kill himself/herself. On U.S. death certificates, this is coded as E950 through E959 in the International Classification of Disease-9th Revision (World Health Organization, 1977). Suicide attempt refers to a behavior with a nonfatal outcome, for which there is evidence (either explicit or implicit) that the person intended at some (nonzero) level to kill himself/herself. A suicide attempt may or may not result in injuries. Suicidal ideation refers to any self-reported thoughts of engaging in suicide-related behavior. Some investigators also consider thoughts that are less explicit in terms of wanting to take one's life (wanting to be dead, not wanting to awake) as indications of "passive" suicide ideation.
Scope: This document applies to all intervention studies where reduction in suicidality is a primary or secondary outcome and to intervention studies aimed at reducing symptomatology and associated disability in disorders known to be associated with an increased risk of suicidality.
Need: Despite the public health and personal burden associated with suicidality, the empirically validated knowledge base is limited. Clinical wisdom and empirical evidence have minimal overlap when it comes to intervention with persons at high risk for suicidality. Interventions such as "no suicide contracts" or hospitalization have little or no evidence to support their effectiveness in reducing suicide risk over time. Moreover, suicidal behaviors, in particular completed and attempted suicide, are relatively rare phenomena, and few studies have been adequately designed with sufficient power to determine the efficacy of interventions. It is often assumed that if treatments are effective for nonsuicidal persons with certain mental disorders, such treatments should be effective in reducing suicidal behavior among persons with the particular disorder as well. There is little evidence in support of this. Persons at high risk for suicide are often excluded from clinical trials: most treatment trials of individuals with mental disorders have excluded those with a history of suicide attempts, as well as those perceived to be at current or future risk for suicide. A common rationale offered for these exclusions was that the potential risks for the study participants exceeded benefits of the treatment offered, and that the capability of the investigators and the design of the treatments were not adequate for monitoring and treating suicidal crises. Although NIMH and the American Foundation for Suicide Prevention have recently funded a number of interventions focused on reducing suicidal behaviors per se, perceived liability risks to investigators and/or sponsors of research continue to limit research efforts in this area.
Consultation: In developing this statement, NIMH obtained input from NIH staff, Office for Human Research Protections (OHRP) staff, and individuals with expertise in clinical research with persons at high risk for suicidality, bioethics, and legal issues pertaining to liability risks with suicidal study participants. Additional input is currently being sought from representatives from the National Advisory Mental Health Council, professional and lay advocacy organizations, Institutional Review Board (IRB) and Data and Safety Monitoring Board (DSMB) members, and former research participants. This document is not intended to be exhaustive with regard to all ethical, legal or safety issues relevant to suicide prevention research, nor was it intended to serve as practice guidelines for any particular profession. Although NIH and OHRP staff have reviewed this discussion and have provided consultation, this document does not necessarily represent the views of NIH or the OHRP. (For further information about broader human research protection efforts, see the OHRP web page, http://www.hhs.gov/ohrp/qi ).
Relevant Policy Documents: This document builds upon all active Federal regulations and policies for support of human research. These regulations and policies include, but are not limited to, the following:
Code of Federal Regulations, Title 45, Part 46, Protection of Human Subjects
Data and Safety Monitoring for Phase I and Phase II Trials
Interim—Research Involving Individuals with Questionable Capacity to Consent: Points to Consider
I.A. Current Federal Initiatives
The critical need for effective treatments to reduce suicidality has been recently highlighted by the Surgeon General's Call to Action to Prevent Suicide (see http://www.surgeongeneral.gov/library/calltoaction/calltoaction.htm ) issued in July 1999 and the Surgeon General's Report on Mental Health issued later that year (http://profiles.nlm.nih.gov/ps/retrieve/ResourceMetadata/NNBBHS ).
Recent reviews of treatments for suicidality found surprisingly limited research on the topic and few effective treatments (Hawton et al., 1998, Linehan, 1997; Rudd, 2000). Several of the 20 or so studies designed to specifically reduce suicidal behavior appeared promising, with several demonstrating reductions in self-harming behavior. However, most of these studies did not have sufficient power to determine their effectiveness. The number of studies designed to treat disorders that have also examined changes in suicidality or rates of suicide deaths as part of broader outcome assessments is also very limited. This may be due to the fact that study participants at high risk for suicidality were excluded from these trials. Linehan (1997) has noted that among the more effective treatment studies for suicidality, those that included study participants with more severe suicidality showed a trend for stronger treatment effects. Some treatment studies (not specifically designed to reduce suicidal behavior) have suggested that certain medications may be useful in reducing suicidal behavior for persons with depression, bipolar illness, or schizophrenia (Baldessarini, et al., 1999; Beasley, et al., 1991; Letizia, et al., 1996; Meltzer & Okayli, 1995).
NIMH has responded to this limited information by placing a high priority on program development efforts to increase research on treatment for persons at risk for suicidality. Recognizing that inclusion of suicidal individuals presents unique challenges to clinical trials, NIMH has supported the development of this paper to consider safety and operational issues. NIMH hopes to see reducing untreated suicidality and hopelessness experienced by a significant proportion of persons with mental disorders become major foci for new research.
NIMH is also expanding its research portfolio to include more representative samples in clinical trials and to develop assessments of longer-term functional outcomes. This translates into broader inclusion criteria that allow for more comorbid conditions, and often more severe levels of illness (see Norquist et al., http://journals.apa.org/prevention/volume2/pre0020001a.html ). Such efforts are likely to increase the numbers of research participants who have been, or will become suicidal.
The desperate need for empirically based treatments for more representative study participant groups, including those who are suicidal, comes at a time when research involving persons with mental disorders is receiving additional scrutiny. The recent President's National Bioethics Advisory Commission (NBAC) report, Research Involving Person with Mental Disorders that May Affect Decisionmaking Capacity did not address the issue of suicidality explicitly but did raise many concerns about researcher's approaches to ensuring that individuals with mental disorders are adequately informed about risks and benefits of research studies, and alternative available treatments. An additional challenge facing clinical trial researchers seeking federal funding is the NIH guideline that requires inclusion of children and adolescents in clinical research unless there is good justification for not doing so (see http://www.nih.gov/grants/guide/notice-files/not98-024.html ).
Taken together, the research and safety challenges, along with new initiatives to test treatments with high risk study participants, require clinical trial researchers to carefully consider implications for consent, monitoring of study participant status, and risk management. In addition to investigators, IRB members, and DSMB members need to be aware of these issues. Testing treatments for suicidal study participants is high-risk research, and efforts are needed to protect the research participants, researchers, and the institutions willing to conduct this research. The purpose of this paper is to consider approaches to protect research participants who are suicidal in clinical trials, while at the same time advancing critically needed research on effective interventions for these individuals.
I.B. Consideration of Safety Efforts in Tandem with Research Evidence
As with all research studies, meeting ethical and legal requirements for informed consent and safety guidelines requires interpretation of available evidence regarding effective treatments and estimations of standards of care. If new research evidence finds effective treatments that reduce suicidality, the incorporation of that knowledge into risk management protocols can protect participants from known risks, and can also help inform general clinical care. That is, the utility of research findings and procedures used to minimize risk can have positive implications for improving clinical practice and its ethics. Persons at risk for suicidality face an elevated risk of death in their everyday lives. The importance of research or practice efforts to reduce such risks cannot be overstated.
Legal, ethical and safety approaches will be limited to current understanding of effective treatments for reducing suicidality. Data and Safety Monitoring Boards, through their oversight of interventions, can facilitate increased knowledge of treatments early in the research process. For example, early or interim analyses of study effects may indicate new information about treatment effectiveness for suicidality and safety, and result in study modification or termination of the study. Because we hope that the research evidence for decreasing suicidality will be increasing soon, we expect that the recommendations suggested here will require modification as more information on effective treatments becomes available. Risks and benefits of various treatment conditions will change as more information on effective treatments become available.
I.C. Factors Associated with Increased Risk for Suicidal Behavior
No study examining putative risk factors for suicide has been able to adequately predict who will complete suicide (e.g., Porkorny, 1983). For this reason, most research examining "risk factors" associated with suicide are based on correlational studies (Linehan, 1999; Moscicki, 1999). Psychological autopsy studies of adults who have died by suicide indicate that mental and/or substance abuse disorders, including personality disorders, were present in at least 90 percent of decedents. Among young and middle-aged adults, previous suicide attempts, impulsive or aggressive tendencies, and stressful life events involving losses in financial, work or social areas are common risk factors; while for older suicides, a late onset of depression is most common (Conwell & Brent, 1995). Common risk factors in youth include prior suicide attempt, symptoms of a mood disorder, substance abuse that was frequently comorbid with a mood disorder (Shaffer & Craft, 1999), parental depression (Brent, Bridge, Johnson, & Connolly, 1996), and stressful life events (Brent, et al. 1993). With regard to risk for suicide attempts, risk factors for suicide completion, as well as comorbid personality disorders, particularly cluster B disorders (borderline, antisocial, histrionic), have been suggested (Brent & Poling, 1997). Risk factors for attempted suicide among youth include the same risk factors for completed suicide in youth, as well as a history of physical and sexual abuse (Fergusson et al., 1996; Kaplan et al., 1997). Homosexuality has also been shown to be correlated with suicide attempts among youth (Fergusson, Horwood, & Beautrais, 1999).
These factors should form the basis for approaches to risk assessment in all appropriate clinical trials. Researchers need to carefully define and estimate the frequency of suicidality likely to be seen in their study participants enrolling in clinical trials. Assessment of risk for suicidality is an ongoing process, and some populations are more likely to have repeated suicidal ideation or behavior at frequent intervals. As described below, it is important to estimate, plan to treat, and inform study participants of these estimates and procedures.
II. Design Considerations
II.A. Treatment Comparison Conditions
Risks and benefits to consider for various treatment comparison conditions will vary as a function of the study purpose. For studies focused on reducing suicidality among research participants per se, the choice of comparator is critical. Many treatments aimed at reducing suicidality will be compared to the standard of care in the community, with the addition of increased monitoring of suicidality in the control or comparison groups. Increased monitoring may offer a level of enhanced care, and in that sense, community standards of treatment plus monitoring could be considered an enhanced treatment as well. Although enhanced treatment as usual as a comparison condition could be seen to possess undesirable risk, the lack of empirically supported interventions for suicidal persons may justify this design approach.
Risk benefit considerations will be different for studies in which suicidal participants will be enrolled, but where suicidality is not the focus of treatment. In these, treatment comparison conditions will have to involve approaches designed to address the illness, as well as reduce the suicidality. For these studies, there are likely a number of alternative, active treatments to consider as comparison conditions beyond treatment as usual. For studies aimed at reducing suicide risk per se, as well as studies focused on treating suicidality as part of an intervention focused on a mental or substance abuse disorder, investigators should consider whether increased monitoring of suicidality in all arms of the study may work to reduce the power to detect effects of the active treatment. Much of the early intervention research on reducing suicidality had inadequate statistical power to detect differences among treatment groups. Treatment trial designs must be scientifically sound in order to produce meaningful results, otherwise benefits based on improved knowledge about these issues cannot reasonably be expected.
II.B. Treatment Commensurate with Study participant Risk Status
Treatment trial protocols with participants who are or may become suicidal should include provisions for managing serious suicide attempts and facilitating greater intensity of care, such as day treatment or inpatient hospitalization. Increased intensity of care is likely to be a part of the treatment protocol in studies focused on reducing suicidality per se. For studies focused on treating mental or substance abuse disorders where suicidality is expected to be less frequent, a risk management protocol appropriate for the expected frequency of suicidality should be operationalized. In working with their Institutional Review Board (IRB) and Data and Safety Monitoring Board (DSMB), investigators should describe how likely serious attempts are to occur, how frequent inpatient hospitalization may be, and their protocols for these procedures. Since an emergency room visit or hospitalization does not guarantee that suicidal behavior will be avoided, investigators need to inform IRBs of how serious suicide attempts will be managed (e.g., individualized treatment in either an inpatient hospital or intensive outpatient setting).
II.C. Points to Consider in Planning an Intervention Trial with Suicidal Study Participants
1. Identify specific inclusion criteria and their measurement with regard to suicidality. Examples include: high levels of suicidal ideation as measured by a valid self-report scale; history of a near lethal suicide attempt as rated by an interview and lethality scale. Exclusion criteria, including those relevant to suicidality, should also be specified.
2. Specify the criteria for withdrawal from the treatment trial with regard to increased suicidality, increased related symptoms, lack of treatment response, and treatment side effects, and what alternative treatment or referral will be offered. The risks and benefits of withdrawing a study participant who attempts suicide from the study should be articulated.
3. Consider and establish criteria for hospitalization, where the hospitalization should take place, and procedures within the hospital that provide additional safety.
4. Describe procedures in the protocol for managing increases in suicidality, and how research staff is trained and available to provide such clinical management.
5. Have a procedure for emergency coverage that is clearly understood by the clinical research staff, study participants and families. Consider providing a written document describing this coverage.
6. As part of the consent process, consider having explicit discussion with relevant family members, guardians, or friends that includes the risks inherent when study participants are suicidal (risk of death, side effects of treatments); the procedures for handling increases in suicidality; the criteria for withdrawal from the study; the risks and benefits of the treatment and control conditions offered, and the limits of confidentiality. Investigators may want to consider having family members, guardians or friends as participants in the research study. Family members or friends' roles in the treatment should be clear and understood by both the study participant and family members.
7. Consider and identify the limits to confidentiality with respect to suicidality, as well as other circumstances. Communicate these limits to the study participants; in particular inform them that confidentiality will not be maintained if they are in imminent risk. Any additional limits to confidentiality for minors should be clear to them, and their parents or guardians.
8. Consider the impact of suicidality on the study participants' capacity to give informed consent. Develop additional procedures to ensure protection of study participants' rights, if needed.
9. Determine whether additional safeguards are needed to ensure the safety of the study participants. This includes procedures available to individual study participants, such as study participant advocates, or those relevant to the overall conduct of the study, such as a Data Safety Monitoring Board.
10. Consider situations in which a trial would be terminated prematurely. A Data Safety and Monitoring Board may independently review the trial's progress and relevant safety concerns, and address "stopping rules."
III. Monitoring and Risk Management Protocols.
III.A. Risk Management Protocols
All trials that explicitly recruit suicidal persons should establish a risk management protocol prior to initiating recruitment into the study. "Risk management protocol" describes the steps to be taken with life-threatening situations or with significant deterioration in clinical status (see Brown, Bruce & Pearson, in press, for an example). The risk management protocol identifies the signs, symptoms or conditions indicative of meaningful change in risk, establishes procedures for the documentation of this change and presents decision rules or algorithms for crisis intervention. This could result in the study participant's removal from the study. Criteria for withdrawal from the study should be explicit, including plans for additional or alternative treatment. This may or may not need to be reported as an adverse event, depending on the study (see NIH policy on reporting adverse events at http://grants.nih.gov/grants/guide/notice-files/not99-107.html , which is now relevant to single site clinical trials: http://grants.nih.gov/grants/guide/notice-files/NOT-OD-00-038.html ).
While empirically validated instruments can aide in the judgment of risk status, the decision to shift into a risk management protocol is best left to the clinical judgment offered by adequately trained and experienced clinicians. NIMH encourages the development of improved operationalization of this approach, and its inclusion for review by their IRB, scientific review, and by the DSMB. Risk management could include use of multiple thresholds in screening and monitoring. If these measures are administered by persons with little clinical experience, supervision should be established to assure timely and appropriate action. Beyond screening, the next level of risk evaluation should include the systematic recording of clinicians' concerns. Review of existing research can assist researchers in deciding the detection/specification area of assessment, as well as more in-depth assessment of suicidality or such protective factors as reasons for living. Investigators may also wish to consider other approaches to additional safeguards. NIH has proposed options for additional safeguards for research involving increasing risk and/or participants with impairment, such as the use of an independent monitor (see http://grants.nih.gov/grants/policy/questionablecapacity.htm ).
With regard to a risk management protocol for persons who become suicidal, Linehan (1999) has proposed detailed steps for assessing and treating suicidal behavior. Actions recommended include the identification of the study participant's access to means of suicide. In an acute situation, the study participant should not be left alone until risk is further evaluated. When appropriate, family members should be informed of the urgency of limiting access to any means of suicide (firearm, medications). A third action is to maintain or increase contact and treatment and support intensity when suicide risk is imminent and high.
It is often recommended that clinicians develop a "no-suicide contract" to minimize risk for suicidality. Although there is no empirical evidence that such contracts reduce risk for suicidal behavior (or liability risk for therapists), the components of a contract are consistent with recommended treatments for suicidal patients and include: a) providing the opportunity for both research participant and therapist to commit to actions that decrease suicidality, and not being ambivalent about this goal; b) defining the thoughts and behaviors that precede suicidal behaviors, as well as defining the suicidal thoughts and behaviors themselves, which may help the research participant and his or her support network to better monitor downward trends; c) identify possible steps to take to reduce these thoughts and behaviors; d) informing the research participant and his/her support network how to access crisis care, including the treating professional. Some groups using cognitive-behavioral approaches in treatment have recommended that, along with the steps described above, therapists consider with the patients likely scenarios of when suicide risk will increase again, and rehearse strategies and develop alternative reactions and behaviors more appropriate and effective than becoming suicidal (Beck & Brown, 1999; Brent & Poling, 1997).
III.B. Increased Monitoring and Supervision
The frequency of clinical monitoring of suicidality is determined by the level of risk of research participants. Some people are persistently suicidal. Others may become suicidal only sporadically. Provision for planned, routine monitoring is a necessary step to ensure early identification of distress and appropriate crisis intervention.
The quality of data gathered on suicidal states needs to be monitored and evaluated in a timely manner by appropriately trained and supervised personnel. Enhanced monitoring can also address other safety aspects of treatment trials with persons at high risk for suicidal behavior, including treatment side effects, lack of treatment response, or an increase in related symptoms. NIMH has recently requested that intervention studies focused on reducing suicidality per se establish a data monitoring system, specifically a Data and Safety Monitoring Board (DSMB) that reports to the Institute and the investigator's IRB as a way of increasing safety through monitoring. In the case of multi-site trials, the DSMB can serve as the central point for reporting adverse events. OHRP notes that for multicenter trials for which there is a DSMB, "IRBs conducting continuing review of research may rely on a current statement from the DSMB indicating that it has reviewed study-wide adverse events, interim findings, and any recent literature that may be relevant to the research, in lieu of requiring that this information be submitted directly to the IRB. Of course, the IRB must still receive and review reports of local, on-site unanticipated problems involving risks to subjects or others and any other information needed to ensure that its continuing review is substantive and meaningful." (for the full text of this guidance, see .
Depending on the level of risk of lethality likely to be seen among research participants in a study, the investigator may want to work with their IRB and DSMB to determine the frequency and level of severity of suicidality for which reports of events will be provided. For example, if the rate of self-harming behavior is anticipated to be high, and treatment protocols are designed to manage such high rates, a periodic report may be sufficient to keep the IRB and NIMH informed. It is critical that the clinical research staff have sufficient expertise and experience with high-risk study participants in order to minimize false positive reporting of adverse effects (e.g., reporting suicide ideation without imminent plans or risk). Incorporating measures with sufficient validity and specificity in protocols, follow-up procedures to further determine risk, along with the clinical judgment of experienced and competent staff, can be used to minimize false positive reports of adverse events. It is the responsibility of the investigators to know the adverse events reporting requirements for their state and/or institution, as well as follow-up procedures. For example, some institutions request that clinical researchers not directly involved in the trial be consulted with regard to retaining or withdrawing a study participant from the study when an adverse event occurs. NIH policy regarding the reporting of adverse events can be found at the following web site: http://grants.nih.gov/grants/guide/notice-files/not99-107.html Specific NIMH policy on adverse events monitoring in clinical trials is being finalized, and will be consistent with the NIH policy cited above.
III.C. Research Clinician Competencies
Initial review groups assess whether investigators and their research teams have the clinical training, capacity and expertise to perform the scientific work of the study. In research involving suicidal persons, this should include a description of the relevant qualifications of personnel who will be working with study participants who are likely to become suicidal. Prior to initiating the treatment protocol, potential treaters of persons at risk for suicidal behavior need to be conversant in both the approaches and steps to take when a research participant reports having suicidal thoughts and/or is planning to engage in suicidal behavior. Research teams should be adequately trained to minimize their ambivalence, fear, or confusion about gathering information about, and acting on treatment plans to reduce or prevent suicidality.
Investigators should have plans in place for evaluating, treating, and/or referral of individuals who are in a suicidal crisis or other emergency. This includes consideration of how to manage or refer potential research participants if imminent risk for suicidal behavior is detected during study recruitment, prior to consent. Voluntary or involuntary inpatient hospitalization requires knowledge of insurance plans, state laws, professional access to facilities and procedures. All professionals involved in the treatment study should be well versed under what circumstances such steps are to be taken, and understand the legal ramifications of these actions for individual participants, such as breaking confidentiality and involuntary commitment, and the potential involvement of other agencies and entities (e.g., protective services). Research participants, and if appropriate, family members, need to be informed of these potential actions and consequences.
In the event of a suicide by a research participant, investigators should follow their institution or treatment facility policies on critical incident or sentinel event review, staff debriefing and procedures for informing family members, as well as providing clinical referrals to family members. Investigators are also encouraged to have in place plans for helping staff review and debrief around the events and interactions pertaining to the study participant to address distress among research team members.
III.D. Legal Risks to Investigators and Institutions Conducting Research with Persons at High Risk for Suicidal Behavior
Practitioners and insurance companies have suggested a number of ways to conduct treatment with suicidal patients to reduce liability claims (e.g., Gutheil,1999; American Professional Agency, 2000). Many of these recommendations include adequate suicide risk assessment, monitoring and documentation, and informing family members of the status of the suicidal patient. However, there is limited case law to guide clinical researchers with regard to their liability in studies involving suicidal study participants. In a review of case law pertaining to clinicians treating suicidal patients, an acceptable standard of care requires an initial and periodic evaluation of suicide potential for all patients seen in clinical practice (Bongar, 1991; Packman & Harris, 1998). If the diagnosis, treatment and surveillance of a patient were seen as adequate, the practice of care is usually considered of sound judgment and the clinician is typically not found liable. Practitioners are responsible for assessing risk for suicide, and implementing a treatment plan to reduce or eliminate the risk. Assessment for elevated risk is not the same as prediction. The courts typically recognize that the prediction of suicide is fraught with uncertainty, and that if providers were considered completely responsible for patients' suicidal behaviors, no health care provider would risk liability exposure to treat such patients, denying suicidal persons necessary treatment. The majority of decisions in malpractice related to suicide deaths are cases concerning inpatients, as hospitals and institutions appear to be held to a higher standard of care since they are assumed to have greater control over the patient environment. Courts have held practitioners or hospitals culpable in the following situations: when the a practitioner failed to investigate previous psychiatric history and current mental state; when a treatment plan has been overlooked or a practitioner has neglected evidence of suicidal tendencies; when a patient has been inadequately supervised; when a patient is released from a hospital while acutely suicidal. Liability is less likely to be found when a patient denies suicidal intent.
Clinical researchers conducting intervention trials with suicidal study participants should consult their institution and/or facilities where the research will be conducted, to determine possible recommendations for type and amount of professional liability insurance to be maintained for the research team members.
IV. Checklist of Informed Consent Issues
For actively suicidal participants in treatment trials, it is particularly important to convey foreseeable risks and reasonably expected benefits, alternatives to study participation such as individualized treatment available outside the study, as well as the limits of confidentiality. Because there has been so little systematic investigation of effective treatments for reducing suicidal behavior, the description of risks and benefits of research on this topic are relatively straightforward, compared to studies where a number of effective treatments exist. Most clinical trials that test treatments aimed at reducing suicidal behavior will reflect the research evidence and inform participants that there are no treatments proven to effectively reduce suicidal behavior per se. However, investigators should consider informing potential participants what might be standard practice for treating their condition. For example, if patients with depression and suicidality are usually treated with an SSRI and/or psychotherapy, then potential participants should be informed about this, even though such a combination has not been "proven" to be successful in reducing suicidal behavior. Best estimates of success rates for depression treatment should be conveyed, along with information about suicide risk associated with untreated depression.
Increased monitoring will also influence the likelihood that other parties may be informed of study protocols and actions, as in the case of reporting adverse events to an IRB or DSMB. Such reporting requirements should be noted on the consent document. Informing research participants and their families that this level of oversight is taking place may also be reassuring to them and improve recruitment and retention. Asking permission of participants to contact and inform a third party of their research participation, and solicit input from them, is an important way of increasing monitoring. Consent documents should include information for participants and designated third parties on how to contact a professional for assistance in emergencies and/or how to contact someone on the research team at any time (i.e., 24 hour availability). Investigators are encouraged to develop an informed consent process that is thorough, yet understandable.
The following checklist suggests specific issues investigators may want to consider addressing in consent for intervention studies involving study participants at high risk for suicidality. Each of the categories listed below are usually contained in consent forms for research. We have indicated under each category issues that may be specific to research with suicidal persons. Format and content for consent forms will vary across institutions.
INFORMATION ON THE RESEARCH STUDY
This section should succinctly describe the main purpose of the study, and why it is necessary to include people at risk for suicidality. The hypotheses of the study should be clearly stated, including whether certain interventions are expected to reduce suicidality.
2. RESEARCH PARTICIPANTS
This section should describe the characteristics of the research participants, including age range, possible clinical characteristics (e.g., a mental disorder with increased risk for suicidality; a recent suicide attempt), or the service setting that characterizes why they are being sought (e.g. consecutive patients seen in an emergency room for a suicide attempt). Investigators may consider describing particular inclusion and exclusion criteria in this section. For example, if potential research participants are likely to be imminently suicidal at enrollment, investigators may wish to indicate that a number of factors will be used as part of an initial evaluation to determine whether enrollment in the study is appropriate at that time.
3. FREE CHOICE TO JOIN OR TO TERMINATE PARTICIPATION
This issue can be complicated when potential research participants are psychiatric inpatients on locked or open units as might be the case if a protocol involves enrolling persons who are actively suicidal or patients who had recently attempted suicide. Although inpatient units and the psychiatric emergency room are likely places to find such persons, it is important to be able to justify that persons in the unit are an appropriate sample, and not just a convenient sample. While being on a locked unit does not necessarily preclude a voluntary choice to refuse or agree to participate in research, extra care should be taken to ensure that recruitment is not tainted by coercion or undue inducement. IRBs will want to assure that potential research participants understand that their current status and their receipt of appropriate treatment does not depend on research participation, and that there is no loss of benefits for nonparticipation. Similarly, if research participants have been involuntarily committed to treatment due to their imminent suicidal or homicidal status, voluntary choice with regard to research participation, and access to treatment must be clarified. Many IRBs will not allow involuntary patients to participate. Others may invoke subpart C of the Code of Federal Regulations, Title 45, Part 46, Protection of Human Subjects, for prisoners, to provide additional protections. IRBs are also likely to inquire whether patients in this category (e.g., inpatient suicidal committed patients) are likely to benefit from the knowledge to be gained.
4. STUDY PROTOCOL AND EXPLANATION OF PROCEDURES
The rationale for increased monitoring for suicidal behavior should be described.
5. ALTERNATIVE TREATMENTS
For studies focused on reducing suicidality per se, most alternative treatments will be limited with regard to their known effectiveness for decreasing suicidality. Investigators may want to indicate that the best way of treating suicidality is not known, although treatments that are routinely used may be effective.
For studies focused on treating mental or substance abuse disorders and associated suicidality, alternative treatments that have shown effectiveness in the treatment of the disorder that are relevant to the study population should be described as such.
If one of the treatments being studied is determined effective in reducing suicidality, that treatment could be offered to the research participant at the end of the study, with no cost for a reasonable duration, if they were assigned to the treatment shown to be ineffective.
6. FOLLOW-UP CARE
For study participants perceived to be at increased risk for suicidality at the end of the study, appropriate referrals should be provided to the study participant, and family members if appropriate. Study participants should be informed that in the case of imminent suicidality, research staff may need to limit confidentiality in order to obtain appropriate care for the individual.
7. RISKS, DISCOMFORTS AND INCONVENIENCES OF THE RESEARCH, AND MEASURES TO BE TAKEN TO MINIMIZE THEM
Statements about steps that will be taken to monitor suicide risks, as well as the possible use of risk management protocols, should be described. Risks associated with the treatments provided should be articulated (e.g., side effects of medications). The consequences of ineffective treatment, such as continued risk for suicidality, continuing depression, and negative effects on social, educational, or work outcomes should be described.
8. WITHDRAWAL FROM THE STUDY
The risks and benefits of being withdrawn from the study, whether initiated by the investigator, or initiated by the participant, should be considered and discussed with the IRB. Investigators should describe what will happen if the participant wishes to withdraw from the study. Participants should be informed of steps that will be taken to determine if they are at high risk for suicidal behavior, and what treatment or referral will be provided if needed. Potential limits to confidentiality at the time of the study participant's withdrawal should be explained.
Consents should note that the investigators may need to withdraw the participant from the study, and the criteria for such. These may include withdrawal from the intervention trial due to increased suicidality, increased related symptoms, lack of treatment response, or treatment side effects. Alternative treatments or referrals to be offered should be specified.
9. MEASURES TO PRESERVE CONFIDENTIALITY OF THE INFORMATION COLLECTED, PRIVACY OF THE SUBJECT, AND LIMITS TO CONFIDENTIALITY
Identify the limits to confidentiality with respect to suicidality, as well as other circumstances, such as referral to appropriate care when the research participant wishes to withdraw from the study, or if the investigators withdraw a participant from the study. Investigators and research participants may want to jointly develop procedures for contacting third parties under various circumstances to maintain confidentiality as well as safety (e.g., contact a family member before contacting law enforcement). Such limitations of confidentiality should be clear to both the participant and research team. Investigators may also wish to emphasize that confidentiality will not be maintained if they are in imminent risk. Any additional limits to confidentiality for minors should be clear to them, and their parents or guardians.
10. EXPECTED DIRECT BENEFITS TO THE RESEARCH SUBJECTS
The anticipated direct benefits of treatment and monitoring in reducing suicidality can be described in this section. The possibility that the study participant will not benefit from the treatments studied should also be stated.
11. EXPECTED INDIRECT BENEFITS TO OTHERS
The type of information from the study expected to improve understanding of effective interventions for suicidal persons is stated here. The likelihood that the improved understanding will guide the clinical practice of mental health professionals and inform further research on interventions to reduce suicidality can also be described.
12. MANAGEMENT OF ANY PHYSICAL INJURY
Persons who are research participants, and attempt suicide, may incur physical injury. The institution's plans for provision of acute treatment related to suicidal behavior injury, injury related to the study treatment, and injury unrelated to suicide intent or study treatments, should be clarified. Availability (or lack thereof) of long term treatment or other compensation should also be noted in the consent document.
13. PAYMENTS TO THE SUBJECT FOR PARTICIPATING IN THE STUDY
Payments should not be considered a benefit to be balanced against research risks. It is important for investigators to avoid undue inducement to be in a study with an inactive treatment where the participant may be asked to delay the opportunity to receive individualized treatment.
14. COSTS TO THE SUBJECT OR SUBJECTS' HEALTH INSURANCE CARRIER RESULTING FROM PARTICIPATION IN THE STUDY
Investigators should be aware of research participants' insurance carrier policies with regard to coverage for treatments of suicidal behavior. Participants should know whether their insurance will be billed and how this may affect future coverage.
15. HOW TO LEARN MORE ABOUT THE STUDY OR RAISE CONCERNS, & WHO TO ASK (no special issues)
16. DOCUMENTATION OF CONSENT
If family members or other designated third parties are not directly involved in the intervention study, investigators may want to consider a policy of informing them of the research participant's consent, and provide them with the same information provided to the participant regarding who to contact, and how, in the face of an emergency or crisis situation.
American Professional Agency, Inc. (2000). Insight: Safeguarding Psychologists Against Liability Risks. Amityville, NY.
Baldessarini RJ, Tondo L, & Hennen J. (1999). Effects of lithium treatment and its discontinuation on suicidal behavior in bipolar manic-depressive disorders. Journal of Clinical Psychiatry, 60 (suppl 2), 77-84.
Beasley CM Jr., Dornseif BE, Bosomworth JC, Sayler ME, Rampey AH Jr., Heiligenstein JH, Thompson VL, Murphy DJ, & Masica DN. (1991). Fluoxetine and suicide: a meta-analysis of controlled trials of treatment for depression. British Medical Journal, 21, 685-92.
Beck AT, & Brown G. (1999). Manual for treating suicide attempters. Unpublished manuscript.
Bongar B. (1991). The suicidal patient: Clinical and legal standards of care. Washington, D.C.: American Psychological Association.
Brent DA, Bridge J, Johnson BA, & Connolly J. (1996). Suicidal behavior runs in families: a controlled family study of adolescent suicide victims. Archives of General Psychiatry, 53, 1145-1152.
Brent DA, Perper JA, Moritz G, Baugher M, Roth C, Balach L, & Schweers J. (1993). Stressful life events, psychopathology, and adolescent suicide: a case control study. Suicide & Life Threatening Behavior, 23, 179-87.
Brent DA & Poling K. (1997). Cognitive Therapy Treatment Manual for Depressed and Suicidal Youth. University of Pittsburgh, Services for Teens at Risk (STAR-Center), Pittsburgh, PA.
Brown GK, Bruce ML, & Pearson JL. (in press). High-risk management guidelines for elderly suicidal patients in primary care settings. International Journal of Geriatric Psychiatry.
Conwell Y. & Brent D. (1995). Suicide and aging I: Patterns of psychiatric diagnosis. International Psychogeriatrics, 7, 149-181.
Fergusson DM, Horwood LJ, & Beautrais AL. (1999). Is sexual orientation related to mental health problems and suicidality in young people? Archives of General Psychiatry, 56, 876-880.
Fergusson DM, Horwood LJ, & Lynskey MT. (1996). Childhood sexual abuse and psychiatric disorder in young adulthood, II: psychiatric outcomes of childhood sexual abuse. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 1365-1374.
Guthiel TG. (1999). Liability issues and liability prevention in suicide. In D.G. Jacobs (Ed.). The Harvard Medical School Guide to Suicidal Assessment and Interventions (pp. 561-578). San Francisco, Jossey-Bass.
Hawton K, Arensman E, Townsend E, Bremner S, Feldman E, Goldney R, Gunnell D, Hazell P, van Heeringen K, House A, Owens D, Sakinofsky I, & Traskman-Bendz L. (1998). Deliberate self-harm: systematic review of efficacy of psychosocial and pharmacological treatments in preventing repetition. British Medical Journal, 317, 441-447.
Hoyert DL, Kochanek KD, Murphy SL. (1999). Deaths: Final data for 1997. National Vital Statistics Report 47(19). Hyattsville, MD: National Center for Health Statistics. DHHS Publication No. (PHS) 99-1120.
Human Subject Protection Regulations, Code of Federal Regulations, Title 45, Part 46, Protection of Human Subjects. http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.html
Kaplan SJ, Pelcovitz D, Salzinger S, Mandel F, & Weiner M. (1997). Adolescent physical abuse and suicide attempts. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 799-808.
Letizia C, Kapik B, & Flanders WD. (1996). Suicidal risk during controlled clinical investigations of fluvoxamine. Journal of Clinical Psychiatry, 57, 415-421.
Linehan MM. (1997). Behavioral treatments of suicidal behaviors: Definitional obfuscation and treatment outcomes. Annals of the New York Academy of Sciences, 836, 302-328.
Linehan MM. (1999). Standard protocol for assessing and treating suicidal behaviors for patients in treatment. In D.G. Jacobs (Ed.). The Harvard Medical School Guide to Suicidal Assessment and Interventions (pp. 146-187). San Francisco, Jossey-Bass.
Meltzer HY, & Okalyli G. (1995). Reduction of suicidality during clozapine treatment of neuroleptic-resistant schizophrenia: Impact on risk-benefit assessment. American Journal of Psychiatry, 152, 183-190.
Moscicki EK. (1999). Epidemiology of suicide. In D. Jacobs (ed.), The Harvard Medical School guide to suicide assessment and intervention (pp. 40-71). San Francisco: Jossey-Bass.
Packman WL & Harris EA. (1998). Legal issues and risk management in suicidal patients. In B. Bongar, A.L. Berman, R.W. Maris, M.M. Silverman, E.A. Harris, & W.L. Packman (Eds.), Risk Management with Suicidal Patients (pp.150-186). NY: Guilford Press.
Porkorny AD. (1983). Prediction of suicide in psychiatric patients. Archives of General Psychiatry, 40, 249-257.
Rudd MD. (2000). Integrating science into the practice of clinical suicidology: A review of the psychotherapy literature and research agenda for the future. In R.W. Maris, S.S. Cannetto, J.L. McIntosh, M.M. Silverman (Eds.), Review of Suicidology, 2000 (pp. 47-83). NY: Guilford Press.
Schaffer D, & Craft L. (1999). Methods of adolescent suicide prevention. Journal of Clinical Psychiatry, 60 (suppl 2), 70-74.
World Health Organization (1977). International classification of diseases. Manual of the international statistical classification of disease, injuries, and causes of death. 9th revision, Vol. 1. Geneva, Switzerland.