Dataset Information

Detection of self-harm and suicidal ideation in emergency department triage notes.

ABSTRACT:

Objective

Accurate identification of self-harm presentations to Emergency Departments (ED) can lead to more timely mental health support, aid in understanding the burden of suicidal intent in a population, and support impact evaluation of public health initiatives related to suicide prevention. Given lack of manual self-harm reporting in ED, we aim to develop an automated system for the detection of self-harm presentations directly from ED triage notes.

Materials and methods

We frame this as supervised classification using natural language processing (NLP), utilizing a large data set of 477 627 free-text triage notes from ED presentations in 2012-2018 to The Royal Melbourne Hospital, Australia. The data were highly imbalanced, with only 1.4% of triage notes relating to self-harm. We explored various preprocessing techniques, including spelling correction, negation detection, bigram replacement, and clinical concept recognition, and several machine learning methods.

Results

Our results show that machine learning methods dramatically outperform keyword-based methods. We achieved the best results with a calibrated Gradient Boosting model, showing 90% Precision and 90% Recall (PR-AUC 0.87) on blind test data. Prospective validation of the model achieves similar results (88% Precision; 89% Recall).

Discussion

ED notes are noisy texts, and simple token-based models work best. Negation detection and concept recognition did not change the results while bigram replacement significantly impaired model performance.

Conclusion

This first NLP-based classifier for self-harm in ED notes has practical value for identifying patients who would benefit from mental health follow-up in ED, and for supporting surveillance of self-harm and suicide prevention efforts in the population.

SUBMITTER: Rozova V

PROVIDER: S-EPMC8800520 | biostudies-literature | 2022 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Detection of self-harm and suicidal ideation in emergency department triage notes.

Rozova Vlada V Witt Katrina K Robinson Jo J Li Yan Y Verspoor Karin K

Journal of the American Medical Informatics Association : JAMIA 20220101 3

<h4>Objective</h4>Accurate identification of self-harm presentations to Emergency Departments (ED) can lead to more timely mental health support, aid in understanding the burden of suicidal intent in a population, and support impact evaluation of public health initiatives related to suicide prevention. Given lack of manual self-harm reporting in ED, we aim to develop an automated system for the detection of self-harm presentations directly from ED triage notes.<h4>Materials and methods</h4>We fr ...[more]

PMID: 34897466

Similar Datasets

Project description:AimsSuicidal ideation constitutes a central element of most theories of suicide and is the defining facet separating suicide from other causes of death such as accidents. However, despite a high worldwide prevalence, most research has focused on suicidal behaviours, such as completed suicide and suicide attempts, while the greater proportion who experienced ideation, which frequently precedes suicidal behaviour, have received much less attention. This study aims to examine the characteristics of those presenting to EDs with suicidal ideation and quantify the associated risk of suicide and other causes of death.MethodsRetrospective cohort study was performed based on population-wide health administration data linked to data from the Northern Ireland Registry of Self-Harm and centrally held mortality records from April 2012 to December 2019. Mortality data, coded as suicide, all-external causes and all-cause mortality were analysed using Cox proportional hazards. Additional cause-specific analyses included accidental deaths, deaths from natural causes and drug and alcohol-related causes.ResultsThere were 1,662,118 individuals aged over 10 years, of whom 15,267 presented to the ED with ideation during the study period. Individuals with ideation had a 10-fold increased risk of death from suicide (hazard ratio [HRadj] = 10.84, 95% confidence interval [CI] 9.18, 12.80) and from all-external causes (HRadj = 10.65, 95% CI 9.66, 11.74) and a threefold risk of death from all-causes (HRadj = 3.01, 95% CI 2.84, 3.20). Further cause-specific analyses indicated that risk of accidental death (HRadj = 8.24, 95% CI 6.29, 10.81), drug-related (HRadj = 15.17, 95% CI 11.36, 20.26) and alcohol-related (HRadj = 10.57, 95% CI 9.07, 12.31) has also significantly increased. There were few socio-demographic and economic characteristics that would identify which of these patients are most at risk of suicide or other causes of death.ConclusionsIdentifying people with suicidal ideation is recognized to be both important but difficult in practice; this study shows that presentations to EDs with self-harm or suicide ideation represent an important potential intervention point for this hard-to-reach vulnerable population. However, and unlike individuals presenting with self-harm, clinical guidelines for the management and recommended best practice and care of these individuals are lacking. Whilst suicide prevention may be the primary focus of interventions aimed at those experiencing self-harm and suicide ideation, death from other preventable causes, especially substance misuse, should also be a cause of concern.

Project description:BackgroundThere is a lack of consensus about the effect of the COVID-19 pandemic on the mental health of children and adolescents. We aimed to compare rates of paediatric emergency department visits for attempted suicide, self-harm, and suicidal ideation during the pandemic with those before the pandemic.MethodsFor this systematic review and meta-analysis, we searched MEDLINE, Embase, and PsycINFO for studies published between Jan 1, 2020, and Dec 19, 2022. Studies published in English with data on paediatric (ie, those aged <19 years) emergency department visits before and during the COVID-19 pandemic were included. Case studies and qualitative analyses were excluded. Changes in attempted suicide, self-harm, suicidal ideation, and other mental-illness indicators (eg, anxiety, depression, and psychosis) were expressed as ratios of the rates of emergency department visits during the pandemic compared with those before the pandemic, and we analysed these with a random-effects meta-analysis. This study was registered with PROSPERO, CRD42022341897.Findings10 360 non-duplicate records were retrieved, which yielded 42 relevant studies (with 130 sample-estimates) representing 11·1 million emergency department visits for all indications of children and adolescents across 18 countries. The mean age of the samples of children and adolescents across studies was 11·7 years (SD 3·1, range 5·5-16·3), and there were on average 57·6% girls and 43·4% boys as a proportion of emergency department visits for any health reasons (ie, physical and mental). Only one study had data related to race or ethnicity. There was good evidence of an increase in emergency department visits for attempted suicide during the pandemic (rate ratio 1·22, 90% CI 1·08-1·37), modest evidence of an increase in emergency department visits for suicidal ideation (1·08, 0·93-1·25), and good evidence for only a slight change in self-harm (0·96, 0·89-1·04). Rates of emergency department visits for other mental-illness indications showed very good evidence of a decline (0·81, 0·74-0·89), and paediatric visits for all health indications showed strong evidence of a reduction (0·68, 0·62-0·75). When rates for attempted suicide and suicidal ideation were combined as a single measure, there was good evidence of an increase in emergency department visits among girls (1·39, 1·04-1·88) and only modest evidence of an increase among boys (1·06, 0·92-1·24). Self-harm among older children (mean age 16·3 years, range 13·0-16·3) showed good evidence of an increase (1·18, 1·00-1·39), but among younger children (mean age 9·0 years, range 5·5-12·0) there was modest evidence of a decrease (0·85, 0·70-1·05).InterpretationThe integration of mental health support within community health and the education system-including promotion, prevention, early intervention, and treatment-is urgently needed to increase the reach of mental health support that can mitigate child and adolescent mental distress. In future pandemics, increased resourcing in some emergency department settings would help to address their expected increase in visits for acute mental distress among children and adolescents.FundingNone.

Project description:IntroductionCanadian patients presenting to the emergency department (ED) typically undergo a triage process where they are assessed by a specially trained nurse and assigned a Canadian Triage and Acuity Scale (CTAS) score, indicating their level of acuity and urgency of assessment. We sought to assess the ability of patients to self-triage themselves through use of one of two of our proprietary self-triage tools, and how this would compare with the standard nurse-driven triage process.MethodsWe enrolled a convenience sample of ambulatory ED patients aged 17 years or older who presented with chief complaints of chest pain, abdominal pain, breathing problems, or musculoskeletal pain. Participants completed one, or both, of an algorithm generated self-triage (AGST) survey, or visual acuity scale (VAS) based self-triage tool which subsequently generated a CTAS score. Our primary outcome was to assess the accuracy of these tools to the CTAS score generated through the nurse-driven triage process.ResultsA total of 223 patients were included in our analysis. Of these, 32 (14.3%) presented with chest pain, 25 (11.2%) with shortness of breath, 75 (33.6%) with abdominal pain, and 91 (40.8%) with musculoskeletal pain. Of the total number of patients, 142 (47.2%) completed the AGST tool, 159 (52.8%) completed the VAS tool and 78 (25.9%) completed both tools. When compared to the nurse-driven triage standard, both the AGST and VAS tools had poor levels of agreement for each of the four presenting complaints.ConclusionsSelf-triage through use of an AGST or VAS tool is inaccurate compared to the established standard of nurse-driven triage. Although existing literature exists which suggests that self-triage tools developed for specific subsets of complaints may be feasible, our results would suggest that adopting the self-triage approach on a broader scale for all-comers to the ED does not appear to be a viable option to enhance the current triage process. Further study is required to show if self-triage can be used in the ED to optimize the triage process.

Dataset Information

Detection of self-harm and suicidal ideation in emergency department triage notes.

Objective

Materials and methods

Results

Discussion

Conclusion

Publications

Detection of self-harm and suicidal ideation in emergency department triage notes.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets