Exploring Uniformity of Clinical Judgment: A Vignette Approach to Understanding Healthcare Professionals’ Suicide Risk Assessment Practices

Abstract

Objectives: Suicide risk assessment often requires health professionals to consider a complex interplay of multiple factors, with a significant reliance on judgment, which can be influenced by factors such as education and experience. Our study aimed at assessing the uniformity of decision making around suicide risk within healthcare professionals. Methods:We used a factorial survey approach to gather information on healthcare professionals' demographics, clinical experience, and their decision on 3 vignettes of patients with suicidal ideation. We used Kruskal-Wallis tests for determining if there were significant differences between groups for continuous variables and Spearman rank correlation for measuring the association between continuous variables. Content analysis was used for analyzing free-text comments.Results: Responses were gathered from 79 healthcare professionals (nurses, nurse practitioners, physicians) who worked in primary care, mental health, or emergency department settings. Median suicide risk rates across all respondents were 90%, 50%, and 53% for vignettes 1, 2, and 3, respectively. Confidence in healthcare professionals' decisions was significantly associated with the clinical designation and personal risk profile of the healthcare professional in certain vignettes, with nurses and those willing to take more risks having a higher confidence in their decisions for vignettes 1 and 3, respectively. Treatment decision was significantly associated with mental health experience (i.e., those with lengthier mental health experiences were less likely to choose "admit to psychiatry ward" for vignette 2), clinical designation (i.e., nurses were more likely to "admit to psychiatry ward" for vignette 1), and practice setting. It should be noted that these associations were not consistent across all 3 vignettes, and results for each association were only specific to 1 of the 3 vignettes.Discussion: Findings compare decision-making practices for suicide risk assessment across several types of healthcare professions over a range of practice settings, with the high-risk vignette showing the least variability. Insights from this study are relevant when building clinical decision support systems for suicide risk assessment. Designers should think about incorporating tailored messaging and alerts to health professionals' mental health experience and/or designation.Conclusions: Within our Canadian sample, there was considerable variability among healthcare professionals assessing the risk of suicide, with important implications for tailoring education and decision support.

Evaluating Suicide Risk Stratification in Outpatient Settings: A Vignette‐Based Approach

“…These differences may also be due to the use of the MIRECC model of suicide risk utilized in the current study, which notes protective factors and preparatory behaviours as key features that distinguish between high and intermediate acute risk.Despite the use of a specific model of risk stratification, the current study identified variability in the risk and disposition decisions of participants, consistent with outcomes in previous research(Pease et al 2017;Sequeira et al 2022). For example,Pease et al (2017) noted variability in assessments of imminent risk related to the presence of suicidal ideation and preparatory behaviours in the vignettes, whileSequeira et al (2022) identified variability in disposition determination related to provider experience. Due to sample size, impacts of clinician experience on variability could not be empirically tested in the current study.…”

“…Vignette studies examining the role of different risk factors on suicide risk assessment have found considerable variability in assessment outcomes between participants. While risk factors related to the presence of a plan and preparatory behaviours are among the most important factors for elevated risk, they also demonstrated the most variability in assessed risk levels among the vignettes (Pease et al 2017;Sequeira et al 2022). Weiland et al (2014) found that assessed risk level was the strongest predictor of disposition, and while they reported general consistency in disposition decisions, more experienced clinicians were more likely to recommend admittance in an emergency setting.…”

“…Vignettes are a frequently used method to elucidate clinical decision-making processes in mental health and suicide prevention settings (Evans et al 2015;Gale et al 2016;Pease et al 2017;Sequeira et al 2022) and have been shown to be useful tools in medical education (Peabody et al 2004). Vignette-based methods allow the construction of hypothetical situations where key variables of interest can be altered to reflect specific research questions, or training circumstances, in an efficient and ethical manner (Converse et al 2015;Evans et al 2015;Wu et al 2014).…”

“…Vignettes have been used in studies of clinical decision-making in suicide risk assessment to evaluate the influence of specific risk factors and warning signs (Pease et al 2017;Sequeira et al 2022) and to examine the relationship between risk levels, protective factors and disposition decisions (Weiland et al 2014). Vignette studies examining the role of different risk factors on suicide risk assessment have found considerable variability in assessment outcomes between participants.…”

“…Previous vignette-based studies of suicide risk assessment examined decision-making among providers conducting risk assessment in crisis settings, including doctors and nurses in emergency departments in Canada(Sequeira et al 2022) and Australia(Weiland et al 2014) and social workers, psychologists and nurses serving as Suicide Prevention Coordinators in VHA facilities(Pease et al 2017), while the current study examined decision-making among outpatient mental health clinicians operating in a low to moderate risk setting. These differences may also be due to the use of the MIRECC model of suicide risk utilized in the current study, which notes protective factors and preparatory behaviours as key features that distinguish between high and intermediate acute risk.Despite the use of a specific model of risk stratification, the current study identified variability in the risk and disposition decisions of participants, consistent with outcomes in previous research(Pease et al 2017;Sequeira et al 2022). For example,Pease et al (2017) noted variability in assessments of imminent risk related to the presence of suicidal ideation and preparatory behaviours in the vignettes, whileSequeira et al (2022) identified variability in disposition determination related to provider experience.…”