Interaction between mental disorders and social disconnectedness on mortality: a population-based cohort study.

BACKGROUND: Despite the recognised importance of mental disorders and social disconnectedness for mortality, few studies have examined their co-occurrence. AIMS: To examine the interaction between mental disorders and three distinct aspects of social disconnectedness on mortality, while taking into account sex, age and characteristics of the mental disorder. METHOD: This cohort study included participants from the Danish National Health Survey in 2013 and 2017 who were followed until 2021. Survey data on social disconnectedness (loneliness, social isolation and low social support) were linked with register data on hospital-diagnosed mental disorders and mortality. Poisson regression was applied to estimate independent and joint associations with mortality, interaction contrasts and attributable proportions. RESULTS: A total of 162 497 individuals were followed for 886 614 person-years, and 9047 individuals (5.6%) died during follow-up. Among men, interaction between mental disorders and loneliness, social isolation and low social support, respectively, accounted for 47% (95% CI: 21-74%), 24% (95% CI: -15 to 63%) and 61% (95% CI: 35-86%) of the excess mortality after adjustment for demographics, country of birth, somatic morbidity, educational level, income and wealth. In contrast, among women, no excess mortality could be attributed to interaction. No clear trends were identified according to age or characteristics of the mental disorder. CONCLUSIONS: Mortality among men, but not women, with a co-occurring mental disorder and social disconnectedness was substantially elevated compared with what was expected. Awareness of elevated mortality rates among socially disconnected men with mental disorders could be of importance to qualify and guide prevention efforts in psychiatric services.

Residential Proximity to Oil and Gas Development and Mental Health in a North American Preconception Cohort Study: 2013-2023.

Objectives. To evaluate associations between oil and gas development (OGD) and mental health using cross-sectional data from a preconception cohort study, Pregnancy Study Online. Methods. We analyzed baseline data from a prospective cohort of US and Canadian women aged 21 to 45 years who were attempting conception without fertility treatment (2013-2023). We developed residential proximity measures for active OGD during preconception, including distance from nearest site. At baseline, participants completed validated scales for perceived stress (10-item Perceived Stress Scale, PSS) and depressive symptoms (Major Depression Inventory, MDI) and reported psychotropic medication use. We used log-binomial regression and restricted cubic splines to estimate prevalence ratios (PRs) and 95% confidence intervals (CIs). Results. Among 5725 participants across 37 states and provinces, residence at 2 km versus 20 to 50 km of active OGD was associated with moderate to high perceived stress (PSS ≥ 20 vs < 20: PR = 1.08; 95% CI = 0.98, 1.18), moderate to severe depressive symptoms (MDI ≥ 20 vs < 20: PR = 1.27; 95% CI = 1.11, 1.45), and psychotropic medication use (PR = 1.11; 95% CI = 0.97, 1.28). Conclusions. Among North American pregnancy planners, closer proximity to OGD was associated with adverse preconception mental health symptomatology. (Am J Public Health. Published online ahead of print July 11, 2024:e1-e12. https://doi.org/10.2105/AJPH.2024.307730).

Exposure to Residential Greenness, Perceived Stress, and Depressive Symptoms in a North American Preconception Cohort.

BACKGROUND: /Aims: Studies suggest that greater exposure to natural vegetation (i.e., greenness) is associated with better mental health. However, there is limited research on greenness and mental health in the preconception period, a critical window of exposure in the life course. We investigated the associations of residential greenness with perceived stress and depressive symptoms using cross-sectional data from a cohort of pregnancy planners. METHODS: From 2013 to 2019, we enrolled female-identified participants aged 21-45 years who were trying to conceive without the use of fertility treatment into a North American preconception cohort study (Pregnancy Study Online [PRESTO]). On the baseline questionnaire, participants completed the 10-item Perceived Stress Scale (PSS) and the Major Depression Inventory (MDI). Using geocoded addresses, we estimated residential greenness exposure via satellite imagery (Normalized Difference Vegetation Index [NDVI]) in a 100m buffer. We estimated mean differences and 95% confidence intervals for the association of greenness with perceived stress and depression scores using linear regression models, adjusting for individual and neighborhood sociodemographic characteristics. We also evaluated the extent to which associations were modified by urbanicity and neighborhood socioeconomic status (SES). RESULTS: Among 9,718 participants, mean age was 29.9 years, 81.5% identified as non-Hispanic White, 25% had household incomes <$50,000, and mean neighborhood income was $61,932. In adjusted models, higher greenness was associated with lower stress and depression scores (mean difference per interquartile range in greenness: -0.20, 95% CI: -0.39, -0.01; and -0.19, 95% CI: -0.48, 0.10, respectively). The association was stronger among residents of lower SES neighborhoods in urban areas (PSS: -0.57, 95% CI: -1.00, -0.15; MDI: -0.72, 95% CI: -1.40, -0.04). CONCLUSIONS: Higher greenness exposure was associated with lower stress and depressive symptoms among pregnancy planners, particularly in lower-SES neighborhoods.

Moving From Traumatic Events to Traumatic Experiences in the Study of Traumatic Psychopathology.

Trauma is defined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) as an event that includes "actual or threatened death, serious injury, or sexual violence" (p. 271). The list of traumatic events included in the DSM-5 represents a long history of psychiatry and psychology's attempts to define trauma and differentiate these events from less severe stressors. In this commentary, we suggest that this strict distinction between traumatic events and stressful events is not useful for public health. The current DSM-5 list of traumatic events may work well for identifying people with the most severe experiences and highest conditional probability of distress who need clinical care. However, the public health field has different priorities. If we think about posttraumatic psychological distress on a population scale, it is not only helping those with the most severe experiences that is needed; rather, public health requires paying attention to all people experiencing distressing stress and trauma reactions. We propose that context is crucial to the development of a population-relevant definition of trauma and provide examples of situations in which stressors have resulted in posttraumatic psychological distress and in which traumatic event reactions have been mitigated by the context in which they occur. We discuss trauma context from an epidemiologic perspective and conclude with recommendations for the field. This article is part of a Special Collection on Mental Health.

Prediction of Suicide Attempts Among Persons with Depression: A Population-Based Case Cohort Study.

Studies have highlighted the potential importance of modeling interactions for suicide attempt prediction. This case-cohort study identified risk factors for suicide attempts among persons with depression in Denmark using statistical approaches that do (random forests) or do not model interactions (least absolute shrinkage and selection operator regression [LASSO]). Cases made a non-fatal suicide attempt (n = 6,032) between 1995 and 2015. The comparison subcohort was a 5% random sample of all persons in Denmark on January 1, 1995 (n = 11,963). We used random forests and LASSO for sex-stratified prediction of suicide attempts from demographic variables, psychiatric and somatic diagnoses, and treatments. Poisonings, psychiatric disorders, and medications were important predictors for both sexes. Area under the receiver operating characteristic curve (AUC) values were higher in LASSO models (0.85 [95% CI = 0.84, 0.86] in men; 0.89 [95% CI = 0.88, 0.90] in women) than random forests (0.76 [95% CI = 0.74, 0.78] in men; 0.79 [95% CI = 0.78, 0.81] in women). Automatic detection of interactions via random forests did not result in better model performance than LASSO models that did not model interactions. Due to the complex nature of psychiatric comorbidity and suicide, modeling interactions may not always be the optimal statistical approach to enhancing suicide attempt prediction in high-risk samples.

Stressful life events and incident depression among U.S. military personnel.

PURPOSE: Although stressful life events (i.e., stressors) and depression are often assumed to be linked, the relation between stressors and incident depression is rarely studied, particularly in the military. The National Guard is a part-time subset of the U.S. military for whom civilian life stressors may be particularly salient, due to the soldiers' dual roles and frequent transitions between military and civilian life. METHODS: We used a dynamic cohort study of National Guard members from 2010 to 2016 to investigate the relationship between recent stressful experiences (e.g., divorce) and incident depression, with an exploratory analysis of effect modification by income. RESULTS: Respondents endorsing at least one of nine past-year stressful events (a time-varying exposure, lagged by 1 year) had almost twice the adjusted rate of incident depression compared to those with no stressful events (HR = 1.8; 95% CI 1.4, 2.4). This association may be modified by income: among individuals making under $80,000 per year, those with past-year stressors had twice the rate of depression compared to those with no stressors, but among those making over $80,000, past-year stressors were associated with only 1.2 times the rate of depression. CONCLUSION: Stressful life events outside of deployment are important determinants of incident depression among National Guard servicemembers, but the effect of these events may be buffered by higher income.

Risk of psychopathology following traumatic events among immigrants and native-born persons in Denmark.

PURPOSE: Immigrants may have increased risk of mental disorders compared with native-born persons. We aimed to expand the limited research on immigrants' posttraumatic psychopathology related to traumatic experiences in their country of resettlement. METHODS: We obtained data from a cohort of Danish residents with ≥ 1 traumatic event recorded in health and administrative national registries during 1994-2016. We calculated risks of posttraumatic stress disorder (PTSD), depression, and substance use disorders (SUD) within 5 years post-trauma among native-born Danes and immigrants who had been in Denmark for ≥ 10 years at the time of their index trauma (including immigrants overall and immigrants from specific regions). Risks were compared via age- and sex-standardized risk ratios (SRR) with 95% confidence intervals (CI). We restricted to persons with no record of the disorder under investigation in the 10 years pre-trauma, and stratified by non-interpersonal (e.g., transport accidents) versus interpersonal trauma (e.g., assaults). RESULTS: Following non-interpersonal trauma, immigrants were more likely than native-born Danes to be diagnosed with PTSD (SRR = 5.2, 95% CI 4.6, 5.9), about as likely to be diagnosed with depression (SRR = 0.98, 95% CI 0.92, 1.1), and less likely to be diagnosed with SUD (SRR = 0.89, 95% CI 0.82, 0.95). Results were similar following interpersonal trauma, except the SRR for PTSD was reduced in magnitude (SRR = 3.0, 95% CI 1.7, 5.4). There were differences by region of birth. CONCLUSION: Immigrants to Denmark have higher risk of PTSD following traumatic experiences than do native-born Danes, possibly due to the combined influence of adverse pre-, peri-, and/or post-migration experiences.

Sampling Validation Data to Achieve a Planned Precision of the Bias-Adjusted Estimate of Effect.

Data collected from a validation substudy permit calculation of a bias-adjusted estimate of effect that is expected to equal the estimate that would have been observed had the gold standard measurement been available for the entire study population. In this paper, we develop and apply a framework for adaptive validation to determine when sufficient validation data have been collected to yield a bias-adjusted effect estimate with a prespecified level of precision. Prespecified levels of precision are decided a priori by the investigator, based on the precision of the conventional estimate and allowing for wider confidence intervals that would still be substantively meaningful. We further present an applied example of the use of this method to address exposure misclassification in a study of transmasculine/transfeminine youth and self-harm. Our method provides a novel approach to effective and efficient estimation of classification parameters as validation data accrue, with emphasis on the precision of the bias-adjusted estimate. This method can be applied within the context of any parent epidemiologic study design in which validation data will be collected and modified to meet alternative criteria given specific study or validation study objectives.

Comparative Effectiveness of Direct-Acting Antivirals for Posttraumatic Stress Disorder in Veterans Affairs Patients With Hepatitis C Virus Infection.

We recently conducted an exploratory study that indicated that several direct-acting antivirals (DAAs), highly effective medications for hepatitis C virus (HCV) infection, were also associated with improvement in posttraumatic stress disorder (PTSD) among a national cohort of US Department of Veterans Affairs (VA) patients treated between October 1, 1999, and September 30, 2019. Limiting the same cohort to patients with PTSD and HCV, we compared the associations of individual DAAs with PTSD symptom improvement using propensity score weighting. After identifying patients who had available baseline and endpoint PTSD symptom data as measured with the PTSD Checklist (PCL), we compared changes over the 8-12 weeks of DAA treatment. The DAAs most prescribed in conjunction with PCL measurement were glecaprevir/pibrentasvir (GLE/PIB; n = 54), sofosbuvir/velpatasvir (SOF/VEL; n = 54), and ledipasvir/sofosbuvir (LDV/SOF; n = 145). GLE/PIB was superior to LDV/SOF, with a mean difference in improvement of 7.3 points on the PCL (95% confidence interval (CI): 1.1, 13.6). The mean differences in improvement on the PCL were smaller between GLE/PIB and SOF/VEL (3.0, 95% CI: -6.3, 12.2) and between SOF/VEL and LDV/SOF (4.4, 95% CI: -2.4, 11.2). While almost all patients were cured of HCV (92.5%) regardless of the agent received, PTSD outcomes were superior for those receiving GLE/PIB compared with those receiving LDV/SOF, indicating that GLE/PIB may merit further investigation as a potential PTSD treatment.

Preexisting stress-related diagnoses and mortality: A Danish cancer cohort study.

BACKGROUND: This study evaluated the association between preexisting stress-related diagnoses and mortality in a Danish population-based cancer cohort. METHODS: This study included Danish patients with cancer diagnosed in 1995-2011 who had a stress-related diagnosis before their cancer diagnosis. Cancer patients without a prior stress-related diagnosis were matched 5:1 to the stress disorder cohort by cancer site, age group, calendar period, and sex. The 5-year cumulative incidence of cancer-specific and all-cause mortality was computed by stress-related diagnosis category. Hazard ratios and 95% confidence intervals (CIs) associating stress-related diagnoses with mortality were computed by follow-up time, stress-related diagnosis category, stage, comorbidity status, and cancer type. RESULTS: This study identified 4437 cancer patients with a preexisting stress-related diagnosis and 22,060 matched cancer cohort members. The 5-year cumulative risk of cancer-specific mortality was 33% (95% CI, 32%-35%) for those with a preexisting stress-related diagnosis and 29% (95% CI, 28%-29%) for those without a prior stress-related diagnosis. Cancer patients with a preexisting stress-related diagnosis had a 1.3 times higher cancer-specific mortality rate than the comparison cohort members (95% CI, 1.2-1.5). This increase persisted across categories of stress-related diagnosis. The association varied by stage and cancer type, with more pronounced associations found among those with a late stage at diagnosis and hematological malignancies. CONCLUSIONS: Cancer patients with preexisting stress-related diagnoses had increased rates of cancer-specific and all-cause mortality. The results suggest that psychiatric comorbidities may be an important consideration for cancer prognosis, and cancer treatment informed by a patient's history may improve outcomes.

The Recent Rise of Suicide Mortality in the United States.

Suicide is a major public health concern in the United States. Between 2000 and 2018, US suicide rates increased by 35%, contributing to the stagnation and subsequent decrease in US life expectancy. During 2019, suicide declined modestly, mostly owing to slight reductions in suicides among Whites. Suicide rates, however, continued to increase or remained stable among all other racial/ethnic groups, and little is known about recent suicide trends among other vulnerable groups. This article (a) summarizes US suicide mortality trends over the twentieth and early twenty-first centuries, (b) reviews potential group-level causes of increased suicide risk among subpopulations characterized by markers of vulnerability to suicide, and (c) advocates for combining recent advances in population-based suicide prevention with a socially conscious perspective that captures the social, economic, and political contexts in which suicide risk unfolds over the life course of vulnerable individuals.

The Joint Effects of Depression and Comorbid Psychiatric Disorders on Suicide Deaths: Competing Antagonism as an Explanation for Subadditivity.

BACKGROUND: Previous studies of the effect of interaction between psychiatric disorders on suicide have reported mixed results. We investigated the joint effect of depression and various comorbid psychiatric disorders on suicide. METHODS: We conducted a population-based case-cohort study with all suicide deaths occurring between 1 January 1995 and 31 December 2015 in Denmark (n = 14,103) and a comparison subcohort comprised of a 5% random sample of the source population at baseline (n = 265,183). We quantified the joint effect of pairwise combinations of depression and major psychiatric disorders (e.g., organic disorders, substance use disorders, schizophrenia, bipolar disorder, neurotic disorders, eating disorders, personality disorders, intellectual disabilities, developmental disorders, and behavioral disorders) on suicide using marginal structural models and calculated the relative excess risk due to interaction. We assessed for the presence of competing antagonism for negative relative excess risk due to interactions. RESULTS: All combinations of depression and comorbid psychiatric disorders were associated with increased suicide risk. For example, the rate of suicide among men with depression and neurotic disorders was 20 times (95% CI = 15, 26) the rate in men with neither disorder. Most disorder combinations were associated with subadditive suicide risk, and there was evidence of competing antagonism in most of these cases. CONCLUSIONS: Subadditivity may be explained by competing antagonism. When both depression and a comorbid psychiatric disorder are present, they may compete to cause the outcome such that having 2 disorders may be no worse than having a single disorder with respect to suicide risk.

Non-pharmacological Treatment for Chronic Pain in US Veterans Treated Within the Veterans Health Administration: Implications for Expansion in US Healthcare Systems.

BACKGROUND: Consensus guidelines recommend multimodal chronic pain treatment with increased use of non-pharmacological treatment modalities (NPM), including as first-line therapies. However, with many barriers to NPM uptake in US healthcare systems, NPM use may vary across medical care settings. Military veterans are disproportionately affected by chronic pain. Many veterans receive treatment through the Veterans Health Administration (VHA), an integrated healthcare system in which specific policies promote NPM use. OBJECTIVE: To examine whether veterans with chronic pain who utilize VHA healthcare were more likely to use NPM than veterans who do not utilize VHA healthcare. DESIGN: Cross-sectional nationally representative study. PARTICIPANTS: US military veterans (N = 2,836). MAIN MEASURES: In the 2019 National Health Interview Survey, veterans were assessed for VHA treatment, chronic pain (i.e., past 3-month daily or almost daily pain), symptoms of depression and anxiety, substance use, and NPM (i.e., physical therapy, chiropractic/spinal manipulation, massage, psychotherapy, educational class/workshop, peer support groups, or yoga/tai chi). KEY RESULTS: Chronic pain (45.2% vs. 26.8%) and NPM use (49.8% vs. 39.4%) were more prevalent among VHA patients than non-VHA veterans. After adjusting for sociodemographic characteristics, psychiatric symptoms, physical health indicators, and use of cigarettes or prescription opioids, VHA patients were more likely than non-VHA veterans to use any NPM (adjusted odds ratio [aOR] = 1.52, 95% CI: 1.07-2.16) and multimodal NPM (aOR = 1.80, 95% CI: 1.12-2.87) than no NPM. Among veterans with chronic pain, VHA patients were more likely to use chiropractic care (aOR = 1.90, 95% CI = 1.12-3.22), educational class/workshop (aOR = 3.02, 95% CI = 1.35-6.73), or psychotherapy (aOR = 4.28, 95% CI = 1.69-10.87). CONCLUSIONS: Among veterans with chronic pain, past-year VHA use was associated with greater likelihood of receiving NPM. These findings may suggest that the VHA is an important resource and possible facilitator of NPM. VHA policies may offer guidance for expanding use of NPM in other integrated US healthcare systems.

Association between post-traumatic stress disorder severity and death by suicide in US military veterans: retrospective cohort study.

BACKGROUND: There is mixed evidence regarding the direction of a potential association between post-traumatic stress disorder (PTSD) and suicide mortality. AIMS: This is the first population-based study to account for both PTSD diagnosis and PTSD symptom severity simultaneously in the examination of suicide mortality. METHOD: Retrospective study that included all US Department of Veterans Affairs (VA) patients with a PTSD diagnosis and at least one symptom severity assessment using the PTSD Checklist (PCL) between 1 October 1999 and 31 December 2018 (n = 754 197). We performed multivariable proportional hazards regression models using exposure groups defined by level of PTSD symptom severity to estimate suicide mortality rates. For patients with multiple PCL scores, we performed additional models using exposure groups defined by level of change in PTSD symptom severity. We assessed suicide mortality using the VA/Department of Defense Mortality Data Repository. RESULTS: Any level of PTSD symptoms above the minimum threshold for symptomatic remission (i.e. PCL score >18) was associated with double the suicide mortality rate at 1 month after assessment. This relationship decreased over time but patients with moderate to high symptoms continued to have elevated suicide rates. Worsening PTSD symptoms were associated with a 25% higher long-term suicide mortality rate. Among patients with improved PTSD symptoms, those with symptomatic remission had a substantial and sustained reduction in the suicide rate compared with those without symptomatic remission (HR = 0.56; 95% CI 0.37-0.88). CONCLUSIONS: Ameliorating PTSD can reduce risk of suicide mortality, but patients must achieve symptomatic remission to attain this benefit.

Distinct groups of firearm owners with differential risk for suicide in the United States: A latent class analysis.

Suicide is a leading cause of mortality and firearm suicide accounts for the majority of fatalities. Firearm owners are a diverse population and firearm-specific suicide prevention programs should be tailored to distinct at-risk firearm-owning groups. This study set out to identify groups of firearm owners with differential suicide risk having unique characteristics that could be used to customize suicide prevention efforts. We conducted a nationally-representative survey of 2646 firearm owners to assess individual suicide risk, suicide risk factors, and demographic characteristics. A Latent Class Analysis identified unique segments of firearm owners at increased risk of suicide with similar underlying suicide risk factors and demographic characteristics. We found almost one in ten (9.6%) of firearm owners were at increased risk of suicide with 25% reporting suicide ideation, 6.6% reporting suicide planning, and 1.8% reporting previous suicide attempts. We identified three unique groups of firearm owners with higher than average suicide risk. Relative to other groups of firearm owners, one at-risk group were more affluent with a history of adverse experiences and mental health challenges, a second group had more male veterans with high levels of alcohol consumption, and third group had more non-heterosexual women who experienced trauma. We conclude that there are three unique groups of firearm owners with higher than average suicide risk with very different characteristics. In addition to broad suicide prevention efforts, customized firearm suicide prevention programs should be developed individually for these different firearm-owning populations, taking into consideration the unique suicide risk factors and demographics of each group.

Using Danish national registry data to understand psychopathology following potentially traumatic experiences.

Research on posttraumatic psychopathology has focused primarily on posttraumatic stress disorder (PTSD); other posttraumatic psychiatric diagnoses are less well documented. The present study aimed to (a) develop a methodology to derive a cohort of individuals who experienced potentially traumatic events (PTEs) from registry-based data and (b) examine the risk of psychopathology within 5 years of experiencing a PTE. Using data from Danish national registries, we created a cohort of individuals with no age restrictions (range: 0-108 years) who experienced at least one of eight possible PTEs between 1994 and 2016 (N = 1,406,637). We calculated the 5-year incidence of nine categories of ICD-10 psychiatric disorders among this cohort and examined standardized morbidity ratios (SMRs) comparing the incidence of psychopathology in this group to the incidence in a nontraumatic stressor cohort (i.e., nonsuicide death of a relative; n = 423,270). Stress disorders (2.5%), substance use disorders (4.1%), and depressive disorders (3.0%) were the most common diagnoses following PTEs. Overall, the SMRs for the associations between any PTE and psychopathology varied from 1.9, 95% CI [1.9, 2.0], for stress disorders to 5.2, 95% CI [5.1. 5.3], for personality disorders. All PTEs except pregnancy-related trauma were associated with all forms of psychopathology. Associations were consistent regardless of whether a stress disorder was present. Traumatic experiences have a broad impact on psychiatric health. The present findings demonstrate one approach to capturing trauma exposure in medical record registry data. Increased traumatic experience characterization across studies will help improve the field's understanding of posttraumatic psychopathology.

Changes in Alcohol Consumption following Direct-Acting Antiviral Treatment for Hepatitis C in VA Patients with Comorbid Alcohol Use Disorder and PTSD.

OBJECTIVE: To investigate whether direct-acting antivirals (DAA) for hepatitis C viral infection (HCV): glecaprevir/pibrentasvir (GLE/PIB), ledipasvir/sofosbuvir (LDV/SOF), and sofosbuvir/velpatasvir (SOF/VEL) are associated with reduced alcohol consumption among veterans with alcohol use disorder (AUD) and co-occurring post-traumatic stress disorder (PTSD). METHODS: We measured change in Alcohol Use Disorder Identification Test-Consumption Module (AUDIT-C) scores in a retrospective cohort of veterans with PTSD and AUD receiving DAAs for HCV. RESULTS: One thousand two hundred and eleven patients were included (GLE/PIB n = 174, LDV/SOF n = 808, SOF/VEL n = 229). Adjusted frequencies of clinically meaningful improvement were 30.5% for GLE/PIB, 45.5% for LDV/SOF, and 40.5% for SOF/VEL. The frequency was lower for GLE/PIB than for LDV/SOF (OR = 0.59; 95% CI [0.40, 0.87]) or SOF/VEL (OR = 0.66; 95% CI [0.42, 1.04]). CONCLUSIONS: DAA treatment for HCV was associated with a substantial reduction in alcohol use in patients with AUD and co-occurring PTSD. Further exploration of the role of DAAs in AUD treatment is warranted.

Validation of the Patient Health Questionnaire-9 (PHQ-9) for Detecting Depression Among Pregnant Women in Lima, Peru.

Depression during pregnancy is linked to adverse perinatal and offspring outcomes. The Patient Health Questionnaire-9 (PHQ-9) has been validated for identifying depression in pregnant women in limited cultural contexts. Construct validity and reliability have been assessed in Lima, Peru, but criterion validity has not. This study aimed to comprehensively evaluate the PHQ-9 among pregnant Peruvian women in the Pregnancy Outcomes, Maternal and Infant Study (PrOMIS). Using Composite International Diagnostic Interview (CIDI) criteria for past-12-month major depressive disorder as the reference standard, sensitivity, specificity, and predictive value of the PHQ-9 for detecting depression were assessed at various cutpoints of the PHQ-9. Confirmatory factor analysis (CFA) was used to evaluate one- and two-factor structures for the PHQ-9. Cronbach's alpha was computed for the entire PHQ-9 scale and for subscales supported by CFA. A cutpoint of ≥8 maximized combined sensitivity (61%) and specificity (62%). At this cutpoint, positive predictive value was low (15%) and negative predictive values was high (93%). Reliability for the full scale was high (α=0.80). Both one- and two-factor solutions were appropriate for this population, but a two-factor solution containing an affective/mood factor (α=0.67) and a somatic factor (α=0.75) was optimal (CFI=0.93, RMSEA=0.075). Among pregnant women in Lima, screening with the PHQ-9 can identify those in need of mental health care, but may identify a large number of false positive cases.

Addressing Measurement Error in Random Forests Using Quantitative Bias Analysis.

Although variables are often measured with error, the impact of measurement error on machine-learning predictions is seldom quantified. The purpose of this study was to assess the impact of measurement error on the performance of random-forest models and variable importance. First, we assessed the impact of misclassification (i.e., measurement error of categorical variables) of predictors on random-forest model performance (e.g., accuracy, sensitivity) and variable importance (mean decrease in accuracy) using data from the National Comorbidity Survey Replication (2001-2003). Second, we created simulated data sets in which we knew the true model performance and variable importance measures and could verify that quantitative bias analysis was recovering the truth in misclassified versions of the data sets. Our findings showed that measurement error in the data used to construct random forests can distort model performance and variable importance measures and that bias analysis can recover the correct results. This study highlights the utility of applying quantitative bias analysis in machine learning to quantify the impact of measurement error on study results.

Predicting Sex-Specific Nonfatal Suicide Attempt Risk Using Machine Learning and Data From Danish National Registries.

Suicide attempts are a leading cause of injury globally. Accurate prediction of suicide attempts might offer opportunities for prevention. This case-cohort study used machine learning to examine sex-specific risk profiles for suicide attempts in Danish nationwide registry data. Cases were all persons who made a nonfatal suicide attempt between 1995 and 2015 (n = 22,974); the subcohort was a 5% random sample of the population at risk on January 1, 1995 (n = 265,183). We developed sex-stratified classification trees and random forests using 1,458 predictors, including demographic factors, family histories, psychiatric and physical health diagnoses, surgery, and prescribed medications. We found that substance use disorders/treatment, prescribed psychiatric medications, previous poisoning diagnoses, and stress disorders were important factors for predicting suicide attempts among men and women. Individuals in the top 5% of predicted risk accounted for 44.7% of all suicide attempts among men and 43.2% of all attempts among women. Our findings illuminate novel risk factors and interactions that are most predictive of nonfatal suicide attempts, while consistency between our findings and previous work in this area adds to the call to move machine learning suicide research toward the examination of high-risk subpopulations.