Gender and Age Bias in the Evaluation of Suicide Attempt Behavior in an Emergency Department

We aimed to investigate whether demographic characteristics such as age and gender of attempted suicide patients are associated with bias in the post-Emergency Department (ED) discharge program manager's evaluation of genuineness of the patients' suicide attempts. In the ED-Based Post-Suicide Attempt Case Management (ED-PSACM) program, the manager interviews patients with suicide attempts and makes subjective judgement on the patient's genuineness of the suicide attempt. After patients' discharge, the manager provides follow-up post-discharge care management services. Compared to ≥ 65 years old male patients as a reference group, 18-39 years old female patients showed significantly lower judgment for a genuine suicide attempt (OR = 0.34; 95% CI 0.12-0.81). Other groups did not show significant differences from the reference group. Our study result suggests the possibility of the effects of bias on young females on the judgment of the suicide attempt genuineness. Medical staff and interventions managers in the ED should be concerned to avoid knowledge-mediated bias, especially by gender and age.

Interventions to change clinicians' behavior related to suicide-prevention care in the emergency department: a scoping review

OBJECTIVE

The objective of this scoping review was to explore, characterize, and map the literature on interventions and intervention components implemented to change emergency department clinicians' behavior related to suicide prevention using the Behaviour Change Wheel as a guiding theoretical framework.

INTRODUCTION

An emergency department is a critical place for suicide prevention, yet patients are often discharged without proper suicide risk assessments and/or referrals. In response, we must support emergency department clinicians' behavior change to follow evidence-based suicide prevention strategies. However, reviews to date have yet to systematically and theoretically examine interventions' functional characteristics and how they can influence emergency department clinicians' behaviors related to suicide-prevention care.

INCLUSION CRITERIA

This review considered interventions that targeted emergency department clinicians' behavior change related to suicide prevention. Behavior change referred to observable practice changes as well as proxy measures of behavior change, including changes in knowledge and attitude.

METHODS

This review followed JBI methodology for scoping reviews. Searches included PubMed, PsycINFO, CINAHL, Embase, and gray literature, including targeted Google searches for relevant organizations/websites, ProQuest Dissertations and Theses Global, and Scopus conference papers (using a specific filter). This review did not apply any date limits, but our search was limited to the English language. Data extraction was undertaken using a charting table developed specifically for the review objective. Narrative descriptions of interventions were coded using the Behavior Change Wheel's intervention functions. Reported outcome measures were categorized. Findings are tabulated and synthesized narratively.

RESULTS

Forty-one studies were included from the database searches, representing a mixture of experimental (n = 2), quasi-experimental (n = 24), non-experimental (n = 12), qualitative (n = 1), and mixed methods (n = 2) approaches. An additional 29 citations were included from gray literature searches. One was a pilot mixed methods study, and the rest were interventions. In summary, this review included a total of 70 citations, describing 66 different interventions. Identified interventions comprised a wide range of Behaviour Change Wheel intervention functions to change clinicians' behavior: education (n = 48), training (n = 40), enablement (n = 36), persuasion (n = 21), environmental restructuring (n = 18), modeling (n = 7), and incentivisation (n = 2). Based on the Behaviour Change Wheel analysis, many interventions targeted more than one determinant of behavior change, often employing education and training to improve clinicians' knowledge and skills simultaneously. Among the 42 studies that reported outcome measures, effectiveness was measured at clinician (n = 38), patient (n = 4), and/or organization levels (n = 6). Few studies reported implementation outcomes, such as measures of reach (n = 4), adoption (n = 5), or fidelity (n = 1). There were no evaluation data reported on the interventions identified through Google searches.

CONCLUSIONS

Interventions included in this review were diverse and leveraged a range of mechanisms to change emergency department clinicians' behavior. However, most interventions relied solely on education and/or training to improve clinicians' knowledge and/or skills. Future research should consider diverse intervention functions to target both individual- and/or organization-level barriers for a given context. Secondly, the ultimate goal for changing emergency department clinicians' behavior is to improve patient health outcomes related to suicide-related thoughts and behaviors, but current research has most commonly evaluated clinicians' behavior in isolation of patient outcomes. Future studies should consider reporting patient-level outcomes alongside clinician-level outcomes.

Prognostic utility of lactate concentrations and kinetics to predict adverse events associated with acute drug overdose

OBJECTIVE

Clinical research on drug intoxication is necessary for appropriate action in emergency departments (EDs). However, currently, there are no evident biomarkers for predicting adverse events (AEs) in patients with drug intoxication. We aimed to evaluate the prognostic value of serum lactate concentrations and lactate kinetics for AEs such as cardiogenic or respiratory failure in patients admitted to the ED with acute drug overdose.

METHODS

We conducted a single-center retrospective study by reviewing the prospective suicide registry of patients visiting the ED. The primary outcome was composite AEs at any point during the ED visit or hospital stay.

RESULTS

A total of 566 patients with acute drug overdose were enrolled in this study. Of these, 62 patients had AEs, whereas 363 patients did not, yielding an AE rate of 14.6%. The median 0 h lactate concentrations in the AE and non-AE groups were 2.7 [2.1-5.1] mmol/L and 2.1 [1.4-2.9] mmol/L, respectively (p < 0.001). The median 6 h lactate concentrations in the AE and non-AE groups were 2.0 [1.5-3.9] mmol/L and 1.3 [0.9-2.2] mmol/L, respectively (p < 0.001). The area under the curve of lactate at 0 h for predicting AEs was 0.705 (95% CI: 0.659-0.748). The optimal lactate cutoff point was 4.2 mmol/L (37.1% sensitivity, 92.8% specificity). Multivariable analysis using a stepwise backward method showed that the 0 h lactate concentration was associated with AEs in acute drug intoxication after adjusting for confounders (adjusted OR of 0 h lactate, 1.47; 95% CI, 1.23-1.77). However, the 6 h lactate concentrations, lactate clearance, and delta lactate levels did not predict the outcomes.

CONCLUSION

Lactate concentrations and kinetics in patients admitted to the ED with an acute drug overdose exhibited limited prognostic utility in predicting AEs and should be interpreted with caution when considered for clinical decision-making.