Predicting functional impairment in euthymic patients with mood disorder: A 5-year follow-up

Major Depressive Disorder and Bipolar Disorder are psychiatric disorders associated with psychosocial impairment. Despite clinical improvement, functional complaints usually remain, mainly impairing occupational and cognitive performance. The aim of this study was to use machine learning techniques to predict functional impairment in patients with mood disorders. For that, analyzes were performed using a population-based cohort study. Participants diagnosed with a mood disorder at baseline and reassessed were considered (n = 282). Random forest (RF) with previous recursive feature selection and LASSO algorithms were applied to a training set with imputed data by bagged trees resulting in two main models. Following recursive feature selection, 25 variables were retained. The RF model had the best performance compared to LASSO. The most important variables in predicting functional impairment were sexual abuse, severity of depressive, anxiety, and somatic symptoms, physical neglect, emotional abuse, and physical abuse. The model demonstrated acceptable performance to predict functional impairment. However, our sample is composed of young participants and the model may not generalize to older individuals with mood disorders. More studies are needed in this direction. The presented calculator has clinical, sociodemographic, and environmental data, demonstrating that it is possible to use such information to predict functional performance.

Prognostic implications of NYHA class and NT-proBNP levels in mild heart failure: a PARADIGM-HF analysis

Background

Treatment recommendations for heart failure (HF) with reduced ejection fraction are primarily centered on New York Heart Association (NYHA) classification, such that apparently asymptomatic patients might not be eligible for disease-modifying therapies. NYHA classification, however, may be particularly limited to discriminate mild forms of HF.

Purpose

The present study aimed to determine the relationship between NYHA classification and an objective measure of HF severity (N-terminal pro–B-type natriuretic peptide [NT pro-BNP]), and their association with long-term prognosis in the PARADIGM-HF trial.

Methods

We compared PARADIGM-HF patients classified as NYHA class I, II, and III at randomization (NYHA class IV patients or with unavailable NYHA class were excluded [n=73]). We present kernel density estimation (KDE) plots–a non-parametric way to describe the underlying distribution of a variable–to compare NT-proBNP levels across NYHA classes. Logistic regression and the area under the receiver operating characteristic curve (AUC) were used to assess the ability to predict a patient's NYHA class using NT-proBNP levels. Time-to-event data were calculated with Kaplan–Meier estimates and NYHA class were further stratified by median baseline NT-proBNP (< or ≥1600 pg/ml). The primary outcome was cardiovascular death or first HF hospitalization.

Results

8326 patients were included in this analysis (median age, 64 years; women, 22%; and median left ventricular ejection fraction, 30%). Of 389 patients classified as NYHA class I at randomization, 228 (59%) changed functional class during the first year after randomization. For log-transformed NT-proBNP, KDE overlapped substantially across NYHA classes (Figure 1A). NT-proBNP levels were a poor predictor of NYHA classification: for NYHA class I vs. II, AUC (95% confidence interval [CI]) was 0.51 (0.48–0.54); for NHYA I vs. III, 0.57 (0.54–0.60); and for NYHA II vs. III, 0.56 (0.54–0.57). NYHA class III patients displayed a distinctively higher rate of cardiovascular deaths or first HF hospitalizations (Figure 1B). NYHA class I and II patients revealed lower event rates that were not significantly different (NYHA II vs. I, HR 1.24 [0.97–1.58]). Stratification by NT-proBNP levels identified subgroups with distinctive risk, such that NYHA I patients with high NT-proBNP levels (n=175) had a higher event rate than patients with low NT-proBNP with any NYHA class (Figure 1C).

Conclusion

NYHA class I and II patients overlapped substantially in objective HF measures and long-term prognosis. NYHA classification remains a powerful predictor of cardiovascular events but might be limited to differentiate mild forms of HF, as apparently asymptomatic patients based on physician-defined functional class might become symptomatic within a year and conceal subjects at substantial risk for adverse outcomes.

Funding Acknowledgement

Type of funding sources: None.

Reassessing the NYHA classification for heart failure: a comparison between classes I and II using cardiopulmonary exercise testing

Background

The New York Heart Association (NYHA) functional classification has evolved to become a major determinant of eligibility to medical interventions and clinical trials, but its reliability to discriminate patients with mild heart failure (HF) has been questioned by studies that demonstrated a major overlap in self-reported symptoms and laboratory markers between NYHA classes I and II.

Purpose

To assess the reliability of NYHA classification by comparing cardio-pulmonary exercise test (CPET) results and its overlap between HF patients classified as NYHA I and II.

Methods

We retrospectively analyzed data from HF patients who underwent CPET in 3 medical centers in Brazil. NYHA class was defined as recorded on CPET day or during the previous clinical visit. Inclusion criteria were diagnosis of HF, age ≥16, and NYHA class I or II. We analyzed overlap between kernel density estimations for the percent-predicted peak VO2, minute ventilation/carbon dioxide production (VE/VCO2) slope, and oxygen uptake efficiency slope (OUES) in patients in NYHA classes I and II. Categorical variables were compared using chi-square tests.

Results

We included 684 patients, of which 42% (284) were classified as NYHA I. Mean age was 56.1 years; 44% (303) were female and mean left ventricular ejection fraction was 36% (±14.2%). Regarding CPET measures, mean global percent-predicted peak VO2 was 56.6% (±26.1%), VE/VCO2 slope was 38.8 (±10.2), and OUES was 1.50 (±0.59). Kernel density overlap between NYHA classes I and II was considerable: 83% for percent-predicted peak VO2, 89% for VE/VCO2 slope, and 85% for OUES (Figure 1). There was no significant difference between NHYA I and II in CPET indicators of poor prognosis: percent-predicted peak VO2 <50% was present in 53% vs. 48% of patients classified as NYHA I and II, respectively (p=0.15); VE/VCO2 slope >36 in 54% vs. 57% (p=0.39); and OUES <1.4 in 46% vs. 48% (p=0.51).

Conclusions

HF patients classified as NYHA I and II overlap substantially in objective measures of functional capacity assessed by CPET. These findings suggest that NYHA classification is a poor discriminator of cardiopulmonary capacity among patients with mild HF, and raise questions about its use as a benchmark to guide HF therapy.

Funding Acknowledgement

Type of funding sources: None. Figure 1

Patterns of high-risk drinking among medical students: A web-based survey with machine learning

BACKGROUND

Prior studies have found increased rates of alcohol consumption among physicians and medical students. The present study aims to build machine learning (ML) models to identify patterns of high-risk drinking (HRD), including alcohol use disorder, within this population.

METHODS

We analyzed data collected through a web-based survey among Brazilian medical students. Variables included sociodemographic data, personal information, university status, and mental health. Stratification for HRD was carried out based on the AUDIT-C scores. Three ML algorithms were used to build classifiers to predict HRD among medical students: elastic net regularization, random forest, and artificial neural networks. Model interpretation techniques were adopted to assess the most influential predictors for models' decisions, which represent potential factors associated with HRD.

RESULTS

A total of 4840 medical students were included in the study. The prevalence of HRD was 53.03%. The three ML models built were able to distinguish individuals with HRD from low-risk drinking (LRD) with very similar performance. The average AUC scores in the cross-validation procedure were around 0.72, and this performance was replicated in the test set. The most important features for the ML models were the use of tobacco and cannabis, monthly family income, marital status, sexual orientation, and physical activities.

CONCLUSIONS

This study proposes that ML models may serve as tools for initial screening of students regarding their susceptibility for at-risk drinking or alcohol use disorder. In addition, we identified several key factors associated with HRD that could be further investigated and explored for preventive and assistance measures.

Loneliness, but not social distancing, is associated with the incidence of suicidal ideation during the COVID-19 outbreak: a longitudinal study

BACKGROUND

Although social distancing is necessary to decrease COVID-19 dissemination, it might also be associated with suicidal ideation. Therefore, we analyzed the impact of social distancing and loneliness in suicidal ideation.

METHODS

We performed two waves of a snowball sample, web-based survey in Brazil (W1: from May 6th to June 6th, 2020; W2: from June 6th to July 6th, 2020). We assessed whether risk factors related to social relationships (loneliness, living alone, not leaving home, and the number of days practicing social distancing) at W1 were associated with suicidal ideation at W1 and W2 using multiple regression models. Analyses were adjusted for sociodemographic, mental health, and lifestyle variables.

RESULTS

A total of 1,674 (18-75 years old; 86.5% females) were included in our longitudinal sample. Living alone (OR: 1.16; 95%CI = 1.03 - 1.30; p=0.015), number of days practicing social distancing (OR: 1.002; 95%CI = 1.000 - 1.004; p=0.027), and loneliness (OR: 1.49; 95%CI = 1.32 - 1.68; p<0.001) were associated with suicidal ideation in the cross-sectional analysis of W1. Only loneliness (OR= 2.12; 95%CI = 1.06 - 4.24; p = 0.033) remained significant as a risk factor to suicidal ideation in the longitudinal analysis between both waves.

LIMITATION

Snowball, convenience sample design limits outcome estimates. Assessments were not objectively performed.

CONCLUSION

Loneliness was consistently associated with the incidence of suicidal ideation, while other variables, such as living alone, not leaving home, and the number of days practicing social distancing, were not. Measures to overcome loneliness are therefore necessary to reduce suicidal ideation during pandemics.

Who attempts suicide among medical students?

OBJECTIVE

To identify factors associated with a history of suicide attempt in medical students.

METHODS

A Web-based survey was sent out to a sample of medical students. A multi-predictor Poisson regression was performed to identify factors associated with a history of suicide attempt. In addition, an elastic net regularization was used to build a risk calculator to identify students at risk for attempted suicide.

RESULTS

A total of 4,840 participants were included in the study. Prevalence of suicide attempts in the sample was 8.94%. Risk factors associated with past suicide attempt in the multi-predictor Poisson regression were as follows: female gender (P < 0.001); homosexuality (P < 0.001); low income (P = 0.026); bullying by university peers (P = 0.006); childhood (P = 0.001) or adult (P = 0.001) trauma; family history of suicide (P = 0.005); suicidal ideation within the last month (P < 0.001); daily tobacco use (P = 0.037); and being at severe risk for alcohol abuse (P = 0.023). Our elastic net model performed well with an AUC of 0.83.

CONCLUSIONS

This study identifies a number of key factors associated with a history of suicide attempts among medical students. Future longitudinal studies should assess the causal relationship between these factors and suicide attempts. Additionally, these results demonstrate that current available data on suicide attempts among medical students can be used to develop an accurate risk algorithm.

Revisiting the Werther Effect in the 21st Century: Bullying and Suicidality Among Adolescents Who Watched 13 Reasons Why

Unlike most leading causes of death in the United States, suicide rates have not declined during the past 50 years.¹ Among young people the situation is even more dramatic, because suicide rates are rising,² and suicide is now the second cause of death in 15- to 29-year-olds globally.³ It has been suggested that descriptions of suicide in the media might affect behavior and that the young might be more vulnerable to this effect.⁴.

Down-up sequential separation of the placenta

OBJECTIVE

To characterize the patterns of placental separation during the third stage of labor.

METHODS

Continuous real-time ultrasound was performed during the third stage of labor in 101 normal deliveries. The sequence of placental separation was recorded for determining whether the process was multiphasic, the site from which separation commenced and the mode of its progression.

RESULTS

Separation in 97 cases was multiphasic. Monophasic separation in which all parts of the placenta appeared to separate simultaneously occurred in two cases only. Pathological prolongation of the third stage precluded determination of separation in two cases. Ninety-two cases had a uterine wall placenta (anterior or posterior); the separation commenced at one pole and progressed sequentially towards the opposite side in 89 of them. The process started at the lower pole (down-up separation) in 83/92 cases (90.2%) and began from the upper pole (up-down separation) in only 6/92 cases (6.5%). Nine cases had a fundal placenta; of these the separation was also multiphasic but began sequentially from either the anterior or posterior pole, or simultaneously from both, in 8 (88.9%) cases so that the fundal part was separated last (bipolar separation).

CONCLUSIONS

Placental separation is usually an orderly multiphasic phenomenon that begins mostly from the lower pole of the placenta and propagates sequentially upwards. Fundal placentae, however, separate first at their poles with the fundal part being separated last. Recognition of the sequence of events and understanding of the mechanism of placental separation may aid in detecting cases prone to third-stage complications and in managing pathological ones.