Background
Suicide and risk management protocols (SRMPs) in mental health research “
document procedures for identification of, and care for, suicidal participants within the context of a research study” ([
1], p.2). Importantly, SRMPs also aid in ensuring patient safety, provide vital information on how to assess suicidal ideation and manage other types of risk, and respond to adverse events [
2,
3]. Mental health researchers and their research staff may need to intervene in risk situations such as suicidal ideation, and disclosing of harm to self and/or others during research visits and/or interviews. Without broader guidelines on how to prepare for, and manage risk, within mental health research protocols, both researchers and Research Ethics Boards (REBs) may struggle with creating individualized SRMPs that meet the needs of study participants. While it is necessary to tailor these standardized procedures for suicide and risk management to local clinical and research contexts, study populations and study-specific needs, researchers would benefit from evidence-informed guidelines that specify the core components of a SRMP and guide researchers [
1].
Researchers communicate a study’s alignment with national and international safety and ethical standards through the development of SRMPs. These protocols should be consistent with the content of the International Conference on Harmonization who developed the Guidelines for Good Clinical Practice [
4]. Widely recognized and used in clinical trials, standardized protocols describe safe practices to: (a) develop consistency within research teams; (b) identify roles and responsibilities during specific events; and (c) promote the smooth functioning of the day-to-day activities of research projects [
5]. There is a multiplicity of terms in the literature that refer to SRMPs including standard operating procedures (SOPs), safety protocols, crisis protocols, and safety management procedures, among others. To address this inconsistency in terminology and be inclusive of suicide risk and other types of risk, we will use the term SRMP throughout this review.
It is important to acknowledge that the risk of suicide may require intervention as a part of the SRMP, and participants may require a clinical intervention [
2]. This unknown and controversial territory of risk for participants can lead to a complex and ambiguous REB review process that can precipitate confusion between REBs and researchers. To assist research teams in navigating the REB review process, there has been some literature addressing effective ways to communicate potential risks in mental health research using SRMPs [
3,
6]. However, according to Schatten et al. [
2], there is substantial variability in the identification and management of suicide risk amongst different studies. For individual researchers and their ethics committees, decision-making is typically based on the particular sample and research methodology.
There have been significant collaborative efforts to standardize REB processes across different countries around the world. In the United Kingdom, Research Ethics Services is the governing body that ensures standardization of ethical research practices across all established research committees. This includes establishment of a United Kingdom-wide set of SOPs and providing operational guidelines [
7]. In Australia, the ‘national statement’ has been developed by health and research organizations within the country that defines regulations for ethical practices and thus, assures standardization across research bodies [
8]. Similar to Australia, organizations such as the Food and Drug Administration in the United States have developed regulations as well as non-binding recommendations regarding ethical practices for institutions and Institutional Research Boards [
9]. In Canada, through open access, the Canadian Association of Research Ethics Boards and Network of Networks provide collaboratively developed, standardized, Canadian REB SOPs [
10]. These standardized procedures are specific to health sciences REBs and aim to encourage the maintenance of a single universal standard for REBs. Such efforts are in response to reported variations among REBs in assessing the same clinical research protocol [
11‐
13]. In some cases, there were no requirements outlined by a REB for researchers to implement a SRMP. Thus, once a study has received REB approval, it is at the discretion of research teams to develop relevant SRMPs for mitigating risk and prioritizing the safety of research participants. This activity is particularly important where there is a real or perceived risk, including studies where the population may be at higher risk of self-harm or suicidal ideation, or death from suicide [
14,
15]. Having standardized expectations for the content of SRMPs across REBs is in the best interest of researchers, REBs, and research participants [
16]. Yet, a preliminary review of the literature revealed a lack of studies that identified specific components and strategies to include in SRMPs for mental health research where there is a risk of suicide and harm to self and others.
To address this knowledge gap, a scoping review was conducted on SRMPs, used in mental health research documents that assess and manage risk and safety for research participants. The objective of this review was to identify (a) core components and (b) strategies of SRMPs to assess and manage the risk and safety of participants in mental health research. The findings will be used to investigate, review, and inform the development of recommended guidelines that clinicians, researchers, REBs, and academics can adopt when developing SRMPs for their mental health research studies where there is a risk of harm.
Discussion
This scoping review identified 22 peer-reviewed articles and 14 grey literature documents that aimed to identify core components and strategies included in SRMPs to assess and manage risk and safety for research participants. Acknowledging a need for and creating accepted guidelines for SRMPs to address suicide and risk for mental health studies should not be mistaken with the false notion that there is an inherent risk in conducting mental health research. Beliefs about exaggerated risk are maintained through assumptions that participating in mental health research, particularly suicide-related research, increases the risk of suicidal ideation and behaviour [
57‐
62]. A meta-analysis by Blades et al. [
48] demonstrated that asking participants about suicide provided small, yet significant, benefits to participants. Misconceptions on the degree of risk in mental health research has negative implications on research proposal submissions to REBs, where participant safety is the primary concern.
There are many collaborative efforts to standardize research processes around the world. However, specific components and strategies to include in SRMPs to mitigate suicide and risk in mental health research are lacking. This review revealed that SRMPs provide vital information for research teams on how to assess and manage risk and respond to research participants in crisis. Our review identified five core components that researchers, research agencies, and REBs commonly identified as essential in the development and administration of SRMPs. These components included: (1) training; (2) educational resources for research staff; (3) educational resources for research participants; (4) risk assessment and management; and (5) clinical and research oversight. All five core components should be considered when conducting mental health research to ensure the safety of the research participant and rigor of the research. An array of strategies related to each component were identified that could be selected or tailored across a variety of research contexts. However, there are few documents that collectively recommended implementing all five core components and relevant strategies [
3,
16]. This is surprising given the importance of ensuring participant safety and highlights the need for REBs within organizations to standardize SRPMs across research projects. More research is required to verify the core components and strategies, including how to tailor them across different stakeholder groups and contexts.
Training was identified as an essential component of SRMPs for all research teams conducting mental health research on participants at risk for suicide [
3]. Training is generally seen to have a vital part in achieving organizational goals in healthcare [
63] and is a catalyst to staff efficiency, career development, and job satisfaction [
64]. In mental health research, training needs to include specific direction on risk assessment and how to handle a variety of different situations such as participant’s report of suicide ideation and/or action [
29]. Research team leads (e.g. PI, research associate) need to ensure all research staff are trained in how to respond to suicide and assess risk. Campbell et al. [
29] stress that training activities need to be extensive and varied. They strongly recommend that sufficient funds are allocated for development of suicide and risk management learning resources.
While some training strategies were frequently cited in the reviewed documents (e.g. follow-up and debriefing with research staff [
2,
28,
30]), others were only mentioned in a single document (e.g. team building [
29]). This lack of consistency in identifying specific training strategies could suggest that SRMPs are at different stages of development based on the type of study being conducted, source of funding, and organizational training requirements. SRMPs may also be utilized differently depending on the expertise and availability of the research staff. Furthermore, while some training strategies were described in depth (e.g. three-day workshop [
29]), others lacked clarity and precise details (e.g. establishing relevant experience, delineating roles, and ensuring qualifications for the research team [
38]). Inconsistencies in training details provided throughout the reviewed documents highlights the need to determine the aspects of training that are of greatest benefit across diverse research settings in order to develop standardized training modules for assessing and responding to suicide risk that can be implemented similarly to training in other research procedures.
Strategies addressing the essential qualifications for research team members to assess risk of suicide, was primarily targeted towards the PI or team lead (e.g. research associate). Specific research and/or clinical qualifications (e.g. professional licensure) and experience for team member roles (e.g. research assistant, research coordinator, trainees) was either absent or vague [
38]. The sufficiency of required knowledge for various team members was also not specified. Better descriptions of qualifications and knowledge requirements for all team member roles would assist with tailoring the research training for individual projects. Ongoing leadership training also needs to occur to ensure that an individual with the required qualifications and knowledge is appropriately designated to identify training needs, provide training for the research team members and build in time for follow-up of the research activities. Davidson et al. [
65] suggest that successful leaders in health research organizations need to be skilled and competent to enhance collaboration, communication, team effectiveness, data quality, and scientific productivity.
Educational resources are another key component recommended and used in the reviewed documents, to contribute to a safe environment for both individuals (e.g. research staff and research participants) and organizations (e.g. REBs) [
14]. The use of evidence-based protocols and educational resources is consistent with effective knowledge translation strategies that support the successful implementation of research evidence in practice [
66]. Hom et al. [
3] draw attention to the fact that even though individuals and organizations may be wary about research involving participants at risk for suicide, evidence-based protocols to assess and manage this risk can be effectively adapted for different research contexts. The evidence supporting educational strategies and resources needs to be updated regularly by a responsible person to conduct the updates. This individual will need a degree of information literacy to decipher and interpret the immense amount of knowledge that is generated through healthcare research [
67].
Educational resources have been developed for research staff to support their knowledge of, and use of, study procedures that outline to manage and respond to participants in crisis [
1,
29]. Strategies such as flow charts and algorithms (Strategy 2.2) promote a standardized approach to managing the research processes and aid in decision making. Vannoy et al. [
1] state that although flow charts for managing suicide risk and management are rare, they are particularly useful in outlining decision-making around “specific actions to be taken, when, and by whom” (p. 5). Automatic assignment of risk mentioned in Strategy 2.3 [
30] also supports standardized decision making. However, this strategy could be prone to malfunction (and require technical intervention) and may not consider all pertinent information related to clinical context and judgment.
Educational resources for research participants are also important to ensure they are well informed about the study and its potential risks, and that they have knowledge about how to access immediate support and available mental health services, if required. Written educational strategies (Strategy 3.1) have similar limitations for participants as for research staff. They need to be developed, specifically targeting the participant stakeholder group, written in plain language, and address key messages of importance for this group [
16,
68]. Strategies that address the need for immediate or available mental health services [
43,
52] should be well researched, current, and accurate. Providing this information to participants may be considered a co-intervention that may impact study results. Achieving a balance between providing mental health information to keep the participant’s safe and maintaining the study rigor may be challenging. The content of the information given to research participants is important for researchers to track and disclose. Identification of a social support person is also appealing, however studies do not clearly provide information on social support or “support person” or how participants may benefit from reaching out for this support [
3,
42]. Operationalization of social support as part of SRMPs is also not well described in the reviewed documents and the logistics for implementation are lacking.
This review highlights the need for research staff to be trained on how to identify suicide risk level and to respond accordingly. Strategies that could assist in achieving this would include developing a structured and explicit SRMP (Strategy 4.1) within the study proposal and utilization of validated risk assessment measures (Strategy 4.2). These strategies, however, depend on having knowledgeable and capable researchers and clinicians as part of the research team to develop SRMPs and to conduct training with research staff. This training should take into consideration different levels of risk and diverse levels of research staff knowledge and qualifications [
69]. Across documents, levels of risk were mostly matched with required interventions [
30,
42]. Strategies tailored to risk severity can be advantageous to ensure participants receive adequate treatment while also allowing research staff to make decisions with confidence.
Among the components and strategies identified in this review, there was a lack of input from research participants and/or stakeholders described in the development of SRMPs. This is an important consideration for future SRMPs in order to be participant-oriented and align with their needs. Manafo et al. [
70] highlighted significant momentum for engaging patients in all phases of the research process. Working collaborations between patients and researchers has been shown to improve patient and health care outcomes [
71]. With increased focus on patient engagement in clinical research [
70], it is imperative to include participant perspectives in the development of research documents such as SRMPs. Enhancing the value of SRMPs will largely depend on careful evaluation by multiple stakeholders.
While this review found two articles reporting best practices for online SRMPS [
30,
47], there remains a lack of literature on how to adequately prepare researchers and REBs for conducting mental health research involving e-health applications [
72,
73]. As mental health research is rapidly evolving, the utilization of technology and online clinical research studies has become more common as care delivery approaches [
74]. Specifically, research involving e-health interventions, while promising as sustainable and scalable [
74], raises questions on how to ensure participant safety when in-person and face-to-face interactions do not occur [
72]. With online administration of e-health applications in mental health research, there is a need to have appropriate safeguards in place to clearly lay out the best course of action during unanticipated and crisis situations. Defining specific criteria required in SRMPs used in e-health research studies, including online administered and digital interventions, could eliminate uncertainties and relieve the burden currently placed on both REBs and mental health researchers.
Limitations
While this review has several strengths, including the use of an iterative search strategy and a range of databases, there are limitations to note. First, the use of varying terminology to describe SRMPs in the literature (e.g. SOP, action plan, safety protocol, high risk management procedures, suicide risk assessment) may have contributed to the limited number of documents identified for this review. Inconsistent terminology can also impede the development of standardized SRMPs where there needs to be clarity on processes such as risk assessment and determination of adverse effects. Second, the language of included documents was limited to English; therefore, relevant non-English documents may have been missed. Third, the study participants of included documents was limited to age 16 years and up. Important considerations from younger populations (e.g. under 16 years of age) may have contributed more information about potential strategies to support the implementation of SRPMs for children and adolescents at risk.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.