Dataset Information

Machine learning discovery of longitudinal patterns of depression and suicidal ideation.

ABSTRACT:

Background and aim

Depression is often accompanied by thoughts of self-harm, which are a strong predictor of subsequent suicide attempt and suicide death. Few empirical data are available regarding the temporal correlation between depression symptoms and suicidal ideation. We investigated the anecdotal concern that suicidal ideation may increase during a period of depression improvement.

Data

Longitudinal Patient Health Questionnaire (PHQ)-9 is a questionnaire of 9 multiple-choice questions to assess the frequency of depressive symptoms within the previous two weeks. We analyzed a chronic depression treatment population's electronic health record (EHR) data, containing 610 patients' longitudinal PHQ-9 scores (62% age 45 and older; 68% female) within 40 weeks.

Methods

The irregular and sparse EHR data were transformed into continuous trajectories using Gaussian process regression. We first estimated the correlations between the symptoms (total score of the first 8 questions; PHQ-8) and suicide ideation (9th question score; Item 9) using the cross-correlation function. We then used an artificial neural network (ANN) to discover subtypes of depression patterns from the fitted depression trajectories. In addition, we conducted a separate analysis using the unfitted raw PHQ scores to examine PHQ-8's and Item 9's pattern changes.

Results

Results showed that the majority of patients' PHQ-8 and Item 9 scores displayed strong temporal correlations. We found five patterns in the PHQ-8 and the Item 9 trajectories. We also found 8% - 13% of the patients have experienced an increase in suicidal ideation during the improvement of their PHQ-8. Using a trajectory-based method for subtype pattern detection in depression progression, we provided a better understanding of temporal correlations between depression symptoms over time.

SUBMITTER: Gong J

PROVIDER: S-EPMC6754154 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning discovery of longitudinal patterns of depression and suicidal ideation.

Gong Jue J Simon Gregory E GE Liu Shan S

PloS one 20190920 9

<h4>Background and aim</h4>Depression is often accompanied by thoughts of self-harm, which are a strong predictor of subsequent suicide attempt and suicide death. Few empirical data are available regarding the temporal correlation between depression symptoms and suicidal ideation. We investigated the anecdotal concern that suicidal ideation may increase during a period of depression improvement.<h4>Data</h4>Longitudinal Patient Health Questionnaire (PHQ)-9 is a questionnaire of 9 multiple-choice ...[more]

PMID: 31539408

Similar Datasets

Project description:BackgroundRecent analyses have described metabolomic markers for depression and suicidal ideation in non-pregnant adults. We examined the metabolomic profile of antepartum depression and suicidal ideation during mid-pregnancy, a time of high susceptibility to mood disorders.MethodsWe collected fasting blood from 100 pregnant Peruvian women and profiled 307 plasma metabolites using liquid chromatography-mass spectrometry. We used the Patient Health Questionnaire 9 to define antepartum depression (score  ≥ 10) and suicidal ideation (having thoughts that you would be better off dead, or of hurting yourself). Logistic regression was used to calculate odds ratios (ORs).ResultsThree triacylglycerol metabolites (C48:5 triacylglycerol [OR = =1.89; 95% confidence interval (CI): 1.14-3.14], C50:6 triacylglycerol [OR = =1.88; 95%CI: 1.13-3.14], C46:4 triacylglycerol [OR = =1.89; 95%CI: 1.11-3.21]) were associated with higher odds of antepartum depression and 4 metabolites (betaine [OR = =0.56; 95%CI:0.33-0.95], citrulline [OR = =0.58; 95%CI: 0.34-0.98], C5 carnitine [OR = =0.59; 95%CI: 0.36-0.99], C5:1 carnitine [OR = =0.59; 95%CI: 0.35-1.00]) with lower odds of antepartum depression. Twenty-six metabolites, including 5-hydroxytryptophan (OR = =0.52; 95%CI: 0.30-0.92), phenylalanine (OR = =0.41; 95%CI: 0.19-0.91), and betaine (OR = =0.53; 95%CI: 0.28-0.99) were associated with lower odds of suicidal ideation.LimitationsOur cross-sectional study could not determine whether metabolites prospectively predict outcomes. No metabolites remained significant after multiple testing correction; these novel findings should be replicated in a larger sample.ConclusionsAntepartum suicidal ideation metabolomic markers are similar to markers of depression among non-pregnant adults, and distinct from markers of antepartum depression. Findings suggest that mood disorder in pregnancy shares metabolomic similarities to mood disorder at other times and may further understanding of these conditions' pathophysiology.

Dataset Information

Machine learning discovery of longitudinal patterns of depression and suicidal ideation.

Background and aim

Data

Methods

Results

Publications

Machine learning discovery of longitudinal patterns of depression and suicidal ideation.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure