Dataset Information

Identification of Suicidal Ideation in the Canadian Community Health Survey-Mental Health Component Using Deep Learning.

ABSTRACT: Introduction: Suicidal ideation (SI) is prevalent in the general population, and is a risk factor for suicide. Predicting which patients are likely to have SI remains challenging. Deep Learning (DL) may be a useful tool in this context, as it can be used to find patterns in complex, heterogeneous, and incomplete datasets. An automated screening system for SI could help prompt clinicians to be more attentive to patients at risk for suicide. Methods: Using the Canadian Community Health Survey-Mental Health Component, we trained a DL model based on 23,859 survey responses to classify patients with and without SI. Models were created to classify both lifetime SI and SI over the last 12 months. From 582 possible parameters we produced 96- and 21-feature versions of the models. Models were trained using an undersampling procedure that balanced the training set between SI and non-SI; validation was done on held-out data. Results: For lifetime SI, the 96 feature model had an Area under the receiver operating curve (AUC) of 0.79 and the 21 feature model had an AUC of 0.77. For SI in the last 12 months the 96 feature model had an AUC of 0.71 and the 21 feature model had an AUC of 0.68. In addition, sensitivity analyses demonstrated feature relationships in line with existing literature. Discussion: Although further study is required to ensure clinical relevance and sample generalizability, this study is an initial proof of concept for the use of DL to improve identification of SI. Sensitivity analyses can help improve the interpretability of DL models. This kind of model would help start conversations with patients which could lead to improved care and a reduction in suicidal behavior.

SUBMITTER: Desai S

PROVIDER: S-EPMC8264793 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:CANADIAN COMMUNITY HEALTH NURSES (CHNS) WORK IN DIVERSE URBAN, RURAL, AND REMOTE SETTINGS SUCH AS: public health units/departments, home health, community health facilities, family practices, and other community-based settings. Research into specific learning needs of practicing CHNs is sparsely reported. This paper examines Canadian CHNs learning needs in relation to the 2008 Canadian Community Health Nursing Standards of Practice (CCHN Standards). It answers: What are the learning needs of CHNs in Canada in relation to the CCHN Standards? What are differences in CHNs' learning needs by: province and territory in Canada, work setting (home health, public health and other community health settings) and years of nursing practice?Between late 2008 and early 2009 a national survey was conducted to identify learning needs of CHNs based on the CCHN Standards using a validated tool.Results indicated that CHNs had learning needs on 25 of 88 items (28.4%), suggesting CHNs have confidence in most CCHN Standards. Three items had the highest learning needs with mean scores > 0.60: two related to epidemiology (means 0.62 and 0.75); and one to informatics (application of information and communication technology) (mean = 0.73). Public health nurses had a greater need to know about "…evaluating population health promotion programs systematically" compared to home health nurses (mean 0.66 vs. 0.39, p <0.010). Nurses with under two years experience had a greater need to learn "… advocating for healthy public policy…" than their more experienced peers (p = 0.0029). Also, NPs had a greater need to learn about "…using community development principles when engaging the individual/community in a consultative process" compared to RNs (p = 0.05). Many nurses were unsure if they applied foundational theoretical frameworks (i.e., the Ottawa Charter of Health Promotion, the Jakarta Declaration, and the Population Health Promotion Model) in practice.CHN educators and practice leaders need to consider these results in determining where to strengthen content in graduate and undergraduate nursing programs, as well as professional development programs. For practicing CHNs educational content should be tailored based on learner's years of experience in the community and their employment sector.

Project description:ObjectivesTo address a gap in knowledge by simultaneously assessing a broad spectrum of individual socioeconomic and potential health determinants of suicidal ideation (SI) using validated measures in a large UK representative community sample.DesignIn this cross-sectional design, participants were recruited via random area probability sampling to participate in a comprehensive public health survey. The questionnaire examined demographic, health and socioeconomic factors. Logistic regression analysis was employed to identify predictors of SI.SettingCommunity setting from high (n=20) and low (n=8) deprivation neighbourhoods across the North West of England, UK.Participants4319 people were recruited between August 2015 and January 2016. There were 809 participants from low-deprivation neighbourhoods and 3510 from high-deprivation neighbourhoods. The sample comprised 1854 (43%) men and 2465 (57%) women.Primary outcome measuresSI was the dependent variable which was assessed using item 9 of the Patient Health Questionnaire-9 instrument.Results454 (11%) participants reported having SI within the last 2 weeks. Model 1 (excluding mental health variables) identified younger age, black and minority ethnic (BME) background, lower housing quality and current smoker status as key predictors of SI. Higher self-esteem, empathy and neighbourhood belonging, alcohol abstinence and having arthritis were protective against SI. Model 2 (including mental health variables) found depression and having cancer as key health predictors for SI, while identifying as lesbian, gay, bisexual, transgender or queer (LGBTQ) and BME were significant demographic predictors. Alcohol abstinence, having arthritis and higher empathy levels were protective against SI.ConclusionsThis study suggests that it could be useful to increase community support and sense of belonging using a public health approach for vulnerable groups (e.g. those with cancer) and peer support for people who identify as LGBTQ and/or BME. Also, interventions aimed at increasing empathic functioning may prove effective for reducing SI.

Project description:BACKGROUND:During the perinatal period, suicides are more likely to occur in those with depression and who are not receiving active treatment at the time of death. Suicide is a common outcome in people with suicide ideation. We developed an intervention program taking care of comprehensive perinatal maternal mental healthcare to prevent suicide ideation. We hypothesized that our intervention program could reduce postnatal suicide ideation and improve maternal mental health. METHODS:We performed a controlled trial to examine the usual postnatal care plus a maternal suicide prevention program (the intervention group) compared with usual postnatal care alone, which comprised home visits by public health nurses without mental health screening (the control group) in Nagano city, Japan. In total, 464 women were included; 230 were allocated to the control group and 234 to the intervention group. The intervention had three components: 1) all the women received postnatal mental health screening by public health nurses who completed home visits during the neonatal period, 2) the intervention was administered by a multidisciplinary clinical network, and 3) systematic follow-up sheets were used to better understand bio-psycho-social characteristics of both the mothers and their infants and develop responsive care plans. We measured the participants' mental health at 3-4?months postpartum (T1) and 7-8?months postpartum (T2) using the Japanese version of the Edinburgh Postnatal Depression Scale (EPDS). RESULTS:Suicidal ideation was significantly lower in the intervention group compared with the control group at T1 (p?=?0.014); however, this significant between-group difference did not continue to T2 (p?=?0.111). We measured the intervention effects on maternal mental health using the total score of the EPDS, which was significantly improved in the intervention group compared with the control group at T1. Here, the significant difference continued to T2 (p?=?0.049). CONCLUSIONS:Our results indicate that our program may reduce maternal suicidal ideation at 3-4?months postnatally and improve women's mental health during the postnatal periods of 3-4 to 7-8?months. Postnatal maternal mental healthcare, including services to reduce suicide ideation, should be included as an important component of general postnatal care. TRIAL REGISTRATION:Name of registry: A multidisciplinary intervention program for maternal mental health in perinatal periods. UMIN Clinical Trials Registry number: UMIN000033396 . Registration URL: https://upload.umin.ac.jp/cgibin/ctr/ctr_view_reg.cgi?recptno=R000038076 Registration date: July 15, 2018. Registration timing: retrospective.

Dataset Information

Identification of Suicidal Ideation in the Canadian Community Health Survey-Mental Health Component Using Deep Learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets