Supervised Machine Learning: A Brief Primer

Machine learning is increasingly used in mental health research and has the potential to advance our understanding of how to characterize, predict, and treat mental disorders and associated adverse health outcomes (e.g., suicidal behavior). Machine learning offers new tools to overcome challenges for which traditional statistical methods are not well-suited. This paper provides an overview of machine learning with a specific focus on supervised learning (i.e., methods that are designed to predict or classify an outcome of interest). Several common supervised learning methods are described, along with applied examples from the published literature. We also provide an overview of supervised learning model building, validation, and performance evaluation. Finally, challenges in creating robust and generalizable machine learning algorithms are discussed.

Change in posttraumatic stress disorder symptoms: do clinicians and patients agree?

This study assessed the longitudinal association between clinician and patient ratings of posttraumatic stress disorder (PTSD) symptoms over the course of 2 different randomized clinical trials of veterans with chronic PTSD. One trial, the Department of Veterans Affairs Cooperative Study 420 (CSP 420; N = 360) compared trauma-focused and present-centered group therapies, and the 2nd trial compared cognitive processing theory and a waitlist control condition (N = 60). Linear mixed effects modeling revealed significant associations between clinician ratings (Clinician-Administered PTSD Scale; CAPS; D. D. Blake et al., 1990) and patient ratings (Posttraumatic Stress Disorder Checklist; PCL; F. W. Weathers, B. T. Litz, J. A. Herman, J. A. Huska, & T. M. Keane, 1993) in total and symptom clusters of PTSD. Contrary to hypothesis, the amount of change on the CAPS ranged from .75 to .82 standard deviations for every 1 standard deviation change on the PCL. The CAPS and PCL were more closely associated in the trauma-focused vs. present-centered treatment condition in CSP 420, and especially regarding hyperarousal symptoms. When comparing categorization of clinically significant change on the CAPS and PCL, the authors found no differences in the percentages of agreement between clinicians and patients in improvement and exacerbation. The value of multimodal assessment of PTSD treatment outcomes is discussed.

Posttraumatic stress disorder and completed suicide

Most research regarding posttraumatic stress disorder (PTSD) and suicide has focused on suicidal ideation or attempts; no known study of the association between PTSD and completed suicide in a population-based sample has been reported. This study examined the association between PTSD and completed suicide in a population-based sample. Data were obtained from the nationwide Danish health and administrative registries, which include data on all 5.4 million residents of Denmark. All suicides between January 1, 1994, and December 31, 2006, were included, and controls were selected from a sample of all Danish residents. Using this nested case-control design, the authors examined 9,612 suicide cases and 199,306 controls matched to cases on gender, date of birth, and time. Thirty-eight suicide cases (0.40%) and 95 controls (0.05%) were diagnosed with PTSD. The odds ratio associating PTSD with suicide was 9.8 (95% confidence interval: 6.7, 15). The association between PTSD and completed suicide remained after controlling for psychiatric and demographic confounders (odds ratio = 5.3, 95% confidence interval: 3.4, 8.1). Additionally, persons with PTSD and depression had a greater rate of suicide than expected based on their independent effects. In conclusion, a registry-based diagnosis of PTSD based on International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, is a risk factor for completed suicide.

Long-term outcomes of cognitive-behavioral treatments for posttraumatic stress disorder among female rape survivors

OBJECTIVE

We conducted a long-term follow-up (LTFU) assessment of participants from a randomized controlled trial comparing cognitive processing therapy (CPT) with prolonged exposure (PE) for posttraumatic stress disorder (PTSD). Competing hypotheses for positive outcomes (i.e., additional therapy, medication) were examined.

METHOD

Intention-to-treat (ITT) participants were assessed 5-10 years after participating in the study (M = 6.15, SD = 1.22). We attempted to locate the 171 original participants, women with PTSD who had experienced at least one rape. Of 144 participants located, 87.5% were reassessed (N = 126), which constituted 73.7% of the original ITT sample. Self-reported PTSD symptoms were the primary outcome. Clinician-rated PTSD symptoms, comorbid diagnoses, and self-reported depression were secondary outcomes.

RESULTS

Substantial decreases in symptoms due to treatment (as reported in Resick, Nishith, Weaver, Astin, & Feuer, 2002) were maintained throughout the LTFU period, as evidenced by little change over time from posttreatment through follow-up (effect sizes ranging from pr = .03 to .14). No significant differences emerged during the LTFU between the treatment conditions (Cohen's d = 0.06-0.29). The ITT examination of diagnostics indicated that 22.2% of CPT and 17.5% of PE participants met the diagnosis for PTSD according to the Clinician-Administered PTSD Scale (Blake et al., 1995) at the LTFU. Maintenance of improvements could not be attributed to further therapy or medications.

CONCLUSIONS

CPT and PE resulted in lasting changes in PTSD and related symptoms over an extended period of time for female rape victims with extensive histories of trauma.

Gender differences among veterans deployed in support of the wars in Afghanistan and Iraq

BACKGROUND

The changing scope of women's roles in combat operations has led to growing interest in women's deployment experiences and post-deployment adjustment.

OBJECTIVES

To quantify the gender-specific frequency of deployment stressors, including sexual and non-sexual harassment, lack of social support and combat exposure. To quantify gender-specific post-deployment mental health conditions and associations between deployment stressors and posttraumatic stress disorder (PTSD), to inform the care of Veterans returning from the current conflicts.

DESIGN

National mail survey of OEF/OIF Veterans randomly sampled within gender, with women oversampled.

SETTING

The community.

PARTICIPANTS

In total, 1,207 female and 1,137 male Veterans from a roster of all Operation Enduring Freedom/Operation Iraqi Freedom (OEF/OIF) Veterans. Response rate was 48.6 %.

MAIN MEASURES

Deployment stressors (including combat and harassment stress), PTSD, depression, anxiety and alcohol use, all measured via self-report.

KEY RESULTS

Women were more likely to report sexual harassment (OR = 8.7, 95% CI: 6.9, 11) but less likely to report combat (OR = 0.62, 95 % CI: 0.50, 0.76). Women and men were equally likely to report symptoms consistent with probable PTSD (OR = 0.87, 95 % CI: 0.70, 1.1) and symptomatic anxiety (OR = 1.1, 9 5% CI: 0.86, 1.3). Women were more likely to report probable depression (OR = 1.3, 95 % CI: 1.1, 1.6) and less likely to report problematic alcohol use (OR = 0.59, 9 5% CI: 0.47, 0.72). With a five-point change in harassment stress, adjusted odds ratios for PTSD were 1.36 (95 % CI: 1.23, 1.52) for women and 1.38 (95 % CI: 1.19, 1.61) for men. The analogous associations between combat stress and PTSD were 1.31 (95 % CI: 1.24, 1.39) and 1.31 (95 % CI: 1.26, 1.36), respectively.

CONCLUSIONS

Although there are important gender differences in deployment stressors-including women's increased risk of interpersonal stressors-and post-deployment adjustment, there are also significant similarities. The post-deployment adjustment of our nation's growing population of female Veterans seems comparable to that of our nation's male Veterans.

Cognitive-behavioral therapy for PTSD and depression symptoms reduces risk for future intimate partner violence among interpersonal trauma survivors

OBJECTIVE

Women who develop symptoms of posttraumatic stress disorder (PTSD) and depression subsequent to interpersonal trauma are at heightened risk for future intimate partner violence (IPV) victimization. Cognitive-behavioral therapy (CBT) is effective in reducing PTSD and depression symptoms, yet limited research has investigated the effectiveness of CBT in reducing risk for future IPV among interpersonal trauma survivors.

METHOD

This study examined the effect of CBT for PTSD and depressive symptoms on the risk of future IPV victimization in a sample of women survivors of interpersonal violence. The current sample included 150 women diagnosed with PTSD secondary to an array of interpersonal traumatic events; they were participating in a randomized clinical trial of different forms of cognitive processing therapy for the treatment of PTSD. Participants were assessed at 9 time points as part of the larger trial: pretreatment, 6 times during treatment, posttreatment, and 6-month follow-up.

RESULTS

As hypothesized, reductions in PTSD and in depressive symptoms during treatment were associated with a decreased likelihood of IPV victimization at a 6-month follow-up even after controlling for recent IPV (i.e., IPV from a current partner within the year prior to beginning the study) and prior interpersonal traumas.

CONCLUSIONS

These findings highlight the importance of identifying and treating PTSD and depressive symptoms among interpersonal trauma survivors as a method for reducing risk for future IPV.

Prediction of Sex-Specific Suicide Risk Using Machine Learning and Single-Payer Health Care Registry Data From Denmark

Importance

Suicide is a public health problem, with multiple causes that are poorly understood. The increased focus on combining health care data with machine-learning approaches in psychiatry may help advance the understanding of suicide risk.

Objective

To examine sex-specific risk profiles for death from suicide using machine-learning methods and data from the population of Denmark.

Design, Setting, and Participants

A case-cohort study nested within 8 national Danish health and social registries was conducted from January 1, 1995, through December 31, 2015. The source population was all persons born or residing in Denmark as of January 1, 1995. Data were analyzed from November 5, 2018, through May 13, 2019.

Exposures

Exposures included 1339 variables spanning domains of suicide risk factors.

Main Outcomes and Measures

Death from suicide from the Danish cause of death registry.

Results

A total of 14 103 individuals died by suicide between 1995 and 2015 (10 152 men [72.0%]; mean [SD] age, 43.5 [18.8] years and 3951 women [28.0%]; age, 47.6 [18.8] years). The comparison subcohort was a 5% random sample (n = 265 183) of living individuals in Denmark on January 1, 1995 (130 591 men [49.2%]; age, 37.4 [21.8] years and 134 592 women [50.8%]; age, 39.9 [23.4] years). With use of classification trees and random forests, sex-specific differences were noted in risk for suicide, with physical health more important to men's suicide risk than women's suicide risk. Psychiatric disorders and possibly associated medications were important to suicide risk, with specific results that may increase clarity in the literature. Generally, diagnoses and medications measured 48 months before suicide were more important indicators of suicide risk than when measured 6 months earlier. Individuals in the top 5% of predicted suicide risk appeared to account for 32.0% of all suicide cases in men and 53.4% of all cases in women.

Conclusions and Relevance

Despite decades of research on suicide risk factors, understanding of suicide remains poor. In this study, the first to date to develop risk profiles for suicide based on data from a full population, apparent consistency with what is known about suicide risk was noted, as well as potentially important, understudied risk factors with evidence of unique suicide risk profiles among specific subpopulations.

Gender differences in experiences of sexual harassment: data from a male-dominated environment

The goal of this investigation was to examine gender differences in experiences of sexual harassment during military service and the negative mental health symptoms associated with these experiences. Female (n = 2,319) and male (n = 1,627) former reservists were surveyed about sexual harassment during their military service and current mental health symptoms. As expected, women reported a higher frequency of sexual harassment. Further, women had increased odds of experiencing all subtypes of sexual harassment. Being female conferred the greatest risk for experiencing the most serious forms of harassment. For both men and women, sexual harassment was associated with more negative current mental health. However, at higher levels of harassment, associations with some negative mental health symptoms were stronger for men than women. Although preliminary, the results of this investigation suggest that although women are harassed more frequently than men, clinicians must increase their awareness of the potential for sexual harassment among men in order to provide the best possible care to all victims of harassment.

Prevalence and prognosis of stress disorders: a review of the epidemiologic literature

Given the ubiquity of traumatic events, it is not surprising that posttraumatic stress disorder (PTSD) - a common diagnosis following one of these experiences - is characterized as conferring a large burden for individuals and society. Although there is recognition of the importance of PTSD diagnoses throughout psychiatry, the literature on other diagnoses one may receive following a stressful or traumatic event is scant. This review summarizes the literature on stress disorders (classified according to the International Classification of Diseases, 10th Edition [ICD-10]), including acute stress reaction, PTSD, adjustment disorder and unspecified stress reactions. This review focuses on the literature related to common psychiatric and somatic consequences of these disorders. The prevalence and incidence of each disorder are described. A review of epidemiologic studies on comorbid mental health conditions, including depression, anxiety and substance abuse, is included, as well as a review of epidemiologic studies on somatic outcomes, including cancer, cardiovascular disease and gastrointestinal disorders. Finally, the current literature on all-cause mortality and suicide following stress disorder diagnoses is reviewed. Stress disorders are a critical public health issue with potentially deleterious outcomes that have a significant impact on those living with these disorders, the health care system and society. It is only through an awareness of the impact of stress disorders that appropriate resources can be allocated to prevention and treatment. Future research should expand the work done to date beyond the examination of PTSD, so that the field may obtain a more complete picture of the impact all stress disorders have on the many people living with these diagnoses.

Association Between Ambient Heat and Risk of Emergency Department Visits for Mental Health Among US Adults, 2010 to 2019

IMPORTANCE

The implications of extreme heat for physical health outcomes have been well documented. However, the association between elevated ambient temperature and specific mental health conditions remains poorly understood.

OBJECTIVE

To investigate the association between ambient heat and mental health-related emergency department (ED) visits in the contiguous US among adults overall and among potentially sensitive subgroups.

DESIGN, SETTING, AND PARTICIPANTS

This case-crossover study used medical claims data obtained from OptumLabs Data Warehouse (OLDW) to identify claims for ED visits with a primary or secondary discharge psychiatric diagnosis during warm-season months (May to September) from 2010 through 2019. Claims for adults aged 18 years or older with commercial or Medicare Advantage health insurance who were living in 2775 US counties were included in the analysis. Emergency department visits were excluded if the Clinical Classifications Software code indicated that the visits were for screening for mental health outcomes and impulse control disorders.

EXPOSURES

County-specific daily maximum ambient temperature on a continuous scale was estimated using the Parameter-Elevation Relationships on Independent Slopes model. Extreme heat was defined as the 95th percentile of the county-specific warm-season temperature distribution.

MAIN OUTCOMES AND MEASURES

The daily incidence rate of cause-specific mental health diagnoses and a composite end point of any mental health diagnosis were assessed by identifying ED visit claims using primary and secondary discharge diagnosis International Classification of Diseases, Ninth Revision and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes. Conditional logistic regression models were used to estimate the incidence rate ratio (IRR) and 95% CIs for the association between daily temperature and incidence rates of ED visits.

RESULTS

Data from 3 496 762 ED visits among 2 243 395 unique individuals were identified (56.8% [1 274 456] women; mean [SD] age, 51.0 [18.8] years); of these individuals, 14.3% were aged 18 to 26 years, 25.6% were aged 27 to 44 years, 33.3% were aged 45 to 64 years, and 26.8% were aged 65 years or older. Days of extreme heat were associated with an IRR of 1.08 (95% CI, 1.07-1.09) for ED visits for any mental health condition. Associations between extreme heat and ED visits were found for specific mental health conditions, including substance use disorders (IRR, 1.08; 95% CI, 1.07-1.10); anxiety, stress-related, and somatoform disorders (IRR, 1.07; 95% CI, 1.05-1.09); mood disorders (IRR, 1.07; 95% CI, 1.05-1.09); schizophrenia, schizotypal, and delusional disorders (IRR, 1.05; 95% CI, 1.03-1.07); self-harm (IRR, 1.06; 95% CI, 1.01-1.12); and childhood-onset behavioral disorders (IRR, 1.11; 95% CI, 1.05-1.18). In addition, associations were higher among men (IRR, 1.10; 95% CI, 1.08-1.12) and in the US Northeast (IRR, 1.10; 95% CI, 1.07-1.13), Midwest (IRR, 1.11; 95% CI, 1.09-1.13), and Northwest (IRR, 1.12; 95% CI, 1.03-1.21) regions.

CONCLUSIONS AND RELEVANCE

In this case-crossover study of a large population of US adults with health insurance, days of extreme heat were associated with higher rates of mental health-related ED visits. This finding may be informative for clinicians providing mental health services during periods of extreme heat to prepare for increases in health service needs when times of extreme heat are anticipated.

Traumatic brain injury, PTSD, and current suicidal ideation among Iraq and Afghanistan U.S. veterans

Suicide is a prevalent problem among veterans deployed to Iraq and Afghanistan. Traumatic brain injury (TBI) and psychiatric conditions, such as posttraumatic stress disorder (PTSD), are potentially important risk factors for suicide in this population, but the literature is limited by a dearth of research on female veterans and imprecise assessment of TBI and suicidal behavior. This study examined 824 male and 825 female U.S. veterans who were enrolled in the baseline assessment of the Veterans After-Discharge Longitudinal Registry (Project VALOR), an observational registry of veterans with and without PTSD who deployed in support of the wars in Iraq and Afghanistan and were enrolled in the Veterans Affairs healthcare system. Results indicated that current depressive symptoms, PTSD, and history of prior TBI were all significantly associated with current suicidal ideation (Cohen's d = 0.91, Cramers' Vs = .19 and .08, respectively). After adding a number of variables to the model, including psychiatric comorbidity, TBI history was associated with increased risk of current suicidal ideation among male veterans only (RR = 1.55). TBI is an important variable to consider in future research on suicide among veterans of the wars in Iraq and Afghanistan, particularly among male veterans.

Trauma, comorbidity, and mortality following diagnoses of severe stress and adjustment disorders: a nationwide cohort study

Longitudinal outcomes following stress or trauma diagnoses are receiving attention, yet population-based studies are few. The aims of the present cohort study were to examine the cumulative incidence of traumatic events and psychiatric diagnoses following diagnoses of severe stress and adjustment disorders categorized using International Classification of Diseases, Tenth Revision, codes and to examine associations of these diagnoses with all-cause mortality and suicide. Data came from a longitudinal cohort of all Danes who received a diagnosis of reaction to severe stress or adjustment disorders (International Classification of Diseases, Tenth Revision, code F43.x) between 1995 and 2011, and they were compared with data from a general-population cohort. Cumulative incidence curves were plotted to examine traumatic experiences and psychiatric diagnoses during the study period. A Cox proportional hazards regression model was used to examine the associations of the disorders with mortality and suicide. Participants with stress diagnoses had a higher incidence of traumatic events and psychiatric diagnoses than did the comparison group. Each disorder was associated with a higher rate of all-cause mortality than that seen in the comparison cohort, and strong associations with suicide were found after adjustment. This study provides a comprehensive assessment of the associations of stress disorders with a variety of outcomes, and we found that stress diagnoses may have long-lasting and potentially severe consequences.

Assessing Variability and Implementation Fidelity of Suicide Prevention Safety Planning in a Regional VA Healthcare System

BACKGROUND

In 2008, the Veterans Health Administration (VHA) implemented the use of safety planning for suicide prevention. A safety plan is a list of strategies, developed collaboratively with a provider, for a patient to use when suicide risk is elevated. Despite the use of safety plans in VHA, little is known about implementation fidelity, the extent to which safety plans are delivered as intended, or patient-level outcomes of safety planning.

AIMS

This study aimed to explore the implementation fidelity of safety planning in a regional VHA hospital and examine the associations between safety plan quality and completeness with patient outcomes.

METHOD

A comprehensive chart review was conducted for patients who were flagged as high risk for suicide (N = 200). Completeness and quality were coded, as well as information about patient and provider interactions regarding safety plan use.

RESULTS

Safety plans were mostly complete and of moderate quality, although variability existed, particularly in quality. Limited evidence of follow-up regarding safety planning was found in the medical charts. Higher quality was associated with fewer subsequent psychiatric hospitalizations.

CONCLUSION

Variability in implementation fidelity and infrequent follow-up suggest a need for additional training and support regarding the use of safety plans for suicide prevention.

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.

Associations between stress disorders and cardiovascular disease events in the Danish population

OBJECTIVES

Post-traumatic stress disorder (PTSD) is a well-documented risk factor for cardiovascular disease (CVD). However, it is unknown whether another common stress disorder-adjustment disorder--is also associated with an increased risk of CVD and whether gender modifies these associations. The aim of this study was to examine the overall and gender-stratified associations between PTSD and adjustment disorder and 4 CVD events.

DESIGN

Prospective cohort study utilising Danish national registry data.

SETTING

The general population of Denmark.

PARTICIPANTS

PTSD (n=4724) and adjustment disorder (n=64,855) cohorts compared with the general population of Denmark from 1995 to 2011.

PRIMARY OUTCOME MEASURES

CVD events including myocardial infarction (MI), stroke, ischaemic stroke and venous thromboembolism (VTE). Standardised incidence rates and 95% CIs were calculated.

RESULTS

Associations were found between PTSD and all 4 CVD events ranging from 1.5 (95% CI 1.1 to 1.9) for MI to 2.1 (95% CI 1.7 to 2.7) for VTE. Associations that were similar in magnitude were also found for adjustment disorder and all 4 CVD events: 1.5 (95% CI 1.4 to 1.6) for MI to 1.9 (95% CI 1.8 to 2.0) for VTE. No gender differences were noted.

CONCLUSIONS

By expanding beyond PTSD and examining a second stress disorder-adjustment disorder-this study provides evidence that stress-related psychopathology is associated with CVD events. Further, limited evidence of gender differences in associations for either of the stress disorders and CVD was found.

Treatment of posttraumatic stress disorder reduces suicidal ideation

BACKGROUND

Suicide is a significant public health problem. Although various studies have found evidence of posttraumatic stress disorder (PTSD) as a risk factor for suicidal behaviors, no study has examined whether or not PTSD treatment decreases suicidal thoughts. This study aims to fill this gap in the literature by examining changes in suicidal ideation over the course of a randomized clinical trial, which compared two widely used treatments for PTSD-cognitive processing therapy (CPT) and prolonged exposure (PE).

METHODS

Data from 163 trial participants over five time points (pre- and posttreatment, 3 and 9 months posttreatment, and 5-10 years posttreatment) were examined using multilevel growth curve analyses to determine if reductions in PTSD symptoms during treatment were associated with reductions in suicidal ideation. Major depression diagnosis and hopelessness were controlled.

RESULTS

Suicidal ideation decreased sharply during treatment with continued, but more subtle decreases, during the follow-up period. These decreases were associated with decreases in PTSD symptoms over the course of treatment. These associations were not accounted for by depression diagnoses at the start of the study or changes in hopelessness over the course of treatment.

CONCLUSIONS

Two widely used, effective treatments for PTSD reduce suicidal ideation. CPT exhibited a larger influence on suicidal ideation than PE, although the magnitude of the difference was small in size. Inclusion of PTSD screening and treatment could enhance suicide prevention efforts.

Gender differences in posttraumatic stress symptoms among OEF/OIF veterans: an item response theory analysis

Establishing whether men and women tend to express different symptoms of posttraumatic stress in reaction to trauma is important for both etiological research and the design of assessment instruments. Use of item response theory (IRT) can reveal how symptom reporting varies by gender and help determine if estimates of symptom severity for men and women are equally reliable. We analyzed responses to the PTSD Checklist (PCL) from 2,341 U.S. military veterans (51% female) who completed deployments in support of operations in Afghanistan and Iraq (Operation Enduring Freedom/Operation Iraqi Freedom [OEF/OIF]), and tested for differential item functioning by gender with an IRT-based approach. Among men and women with the same overall posttraumatic stress severity, women tended to report more frequent concentration difficulties and distress from reminders whereas men tended to report more frequent nightmares, emotional numbing, and hypervigilance. These item-level gender differences were small (on average d = 0.05), however, and had little impact on PCL measurement precision or expected total scores. For practical purposes, men's and women's severity estimates had similar reliability. This provides evidence that men and women veterans demonstrate largely similar profiles of posttraumatic stress symptoms following exposure to military-related stressors, and some theoretical perspectives suggest this may hold in other traumatized populations.

Association of Early-, Middle-, and Late-Life Depression With Incident Dementia in a Danish Cohort

Importance

Late-life depressive symptoms are associated with subsequent dementia diagnosis and may be an early symptom or response to preclinical disease. Evaluating associations with early- and middle-life depression will help clarify whether depression influences dementia risk.

Objective

To examine associations of early-, middle-, and late-life depression with incident dementia.

Design, Setting, and Participants

This was a nationwide, population-based, cohort study conducted from April 2020 to March 2023. Participants included Danish citizens from the general population with depression diagnoses who were matched by sex and birth year to individuals with no depression diagnosis. Participants were followed up from 1977 to 2018. Excluded from analyses were individuals followed for less than 1 year, those younger than 18 years, or those with baseline dementia.

Exposure

Depression was defined using diagnostic codes from the International Classification of Diseases (ICD) within the Danish National Patient Registry (DNPR) and Danish Psychiatric Central Research Register (DPCRR).

Main Outcomes and Measure

Incident dementia was defined using ICD diagnostic codes within the DPCRR and DNPR. Cox proportional hazards regression was used to examine associations between depression and dementia adjusting for education, income, cardiovascular disease, chronic obstructive pulmonary disease, diabetes, anxiety disorders, stress disorders, substance use disorders, and bipolar disorder. Analyses were stratified by age at depression diagnosis, years since index date, and sex.

Results

There were 246 499 individuals (median [IQR] age, 50.8 [34.7-70.7] years; 159 421 women [64.7%]) with diagnosed depression and 1 190 302 individuals (median [IQR] age, 50.4 [34.6-70.0] years; 768 876 women [64.6%]) without depression. Approximately two-thirds of those diagnosed with depression were diagnosed before the age of 60 years (684 974 [67.7%]). The hazard of dementia among those diagnosed with depression was 2.41 times that of the comparison cohort (95% CI, 2.35-2.47). This association persisted when the time elapsed from the index date was longer than 20 to 39 years (hazard ratio [HR], 1.79; 95% CI, 1.58-2.04) and among those diagnosed with depression in early, middle, or late life (18-44 years: HR, 3.08; 95% CI, 2.64-3.58; 45-59 years: HR, 2.95; 95% CI, 2.75-3.17; ≥60 years: HR, 2.31; 95% CI, 2.25-2.38). The overall HR was greater for men (HR, 2.98; 95% CI, 2.84-3.12) than for women (HR, 2.21; 95% CI, 2.15-2.27).

Conclusions and Relevance

Results suggest that the risk of dementia was more than doubled for both men and women with diagnosed depression. The persistent association between dementia and depression diagnosed in early and middle life suggests that depression may increase dementia risk.

A longitudinal study of risk factors for suicide attempts among Operation Enduring Freedom and Operation Iraqi Freedom veterans

BACKGROUND

Suicide rates among veterans have increased markedly since the onset of Operations Enduring Freedom (OEF) and Iraqi Freedom (OIF; LeardMann et al., 2013). Identification of factors with the greatest contribution to suicide risk among veterans is needed to inform risk assessment and to identify intervention targets.

METHODS

This study examined predictors of suicide attempts among participants in the Veterans After-Discharge Longitudinal Registry; a nationwide cohort of OEF/OIF veterans enrolled in Department of Veterans Affairs (VA) services. Veterans with and without probable posttraumatic stress disorder (PTSD) were sampled at a 3:1 ratio, and male and female veterans were sampled at a 1:1 ratio. Participants (N = 1,649) were assessed at two time points, roughly 2 years apart (M = 28.74 months, SD = 8.72).

RESULTS

Seventy-four participants (4.49%) attempted suicide during the follow-up period. The strongest predictors of suicide attempts among the full sample were suicidal intent, attempt history, suicide ideation, PTSD symptoms, alcohol use disorder (AUD) symptoms, and depression. Veterans with multiple risk factors were particularly vulnerable; of veterans with 0, ≥1, ≥2, ≥3, or ≥ 4 of these risk factors, 0%, 7.81%, 10.31%, 18.45%, and 20.51% made a suicide attempt, respectively.

CONCLUSIONS

This prospective study identified several strong predictors of suicide attempts among OEF/OIF veterans which may be important targets for suicide prevention efforts. Further, co-occurrence of multiple risk factors was associated with markedly greater risk for suicide attempts; veterans with multiple risk factors appear to be at the highest risk among OEF/OIF veterans enrolled in VA care.

Gender Differences in Machine Learning Models of Trauma and Suicidal Ideation in Veterans of the Iraq and Afghanistan Wars

Suicide rates among recent veterans have led to interest in risk identification. Evidence of gender-and trauma-specific predictors of suicidal ideation necessitates the use of advanced computational methods capable of elucidating these important and complex associations. In this study, we used machine learning to examine gender-specific associations between predeployment and military factors, traumatic deployment experiences, and psychopathology and suicidal ideation (SI) in a national sample of veterans deployed during the Iraq and Afghanistan conflicts (n = 2,244). Classification, regression tree analyses, and random forests were used to identify associations with SI and determine their classification accuracy. Findings converged on several associations for men that included depression, posttraumatic stress disorder (PTSD), and somatic complaints. Sexual harassment during deployment emerged as a key factor that interacted with PTSD and depression and demonstrated a stronger association with SI among women. Classification accuracy for SI presence or absence was good based on the receiver operating characteristic area under the curve, men = .91, women = .92. The risk for SI was classifiable with good accuracy, with associations that varied by gender. The use of machine learning analyses allowed for the discovery of rich, nuanced results that should be replicated in other samples and may eventually be a basis for the development of gender-specific actuarial tools to assess SI risk among veterans.

Inflammatory bowel disease and completed suicide in Danish adults

BACKGROUND

The risk of suicide is increased among people with inflammatory bowel disease (IBD); however, most of the evidence regarding this association comes from studies of all-cause mortality, with small numbers of people who have committed suicide, or from clinical case reports. The current study examined the association between the two forms of IBD: Crohn's disease (CD) and ulcerative colitis (UC), and completed suicide in a population-based sample.

METHODS

Data for this nested case-control study were obtained from the Danish national healthcare and social registries. We included 27,053 cases who completed suicide between January 1, 1981, and December 31, 2006. Controls were selected from the general population and matched to cases on gender, date of birth, and calendar time (n = 551,060).

RESULTS

There was an increased rate of suicide among participants with both CD (odds ratio [OR] = 1.6, 95% confidence interval [CI]: 1.1, 2.3) and UC (OR = 1.9, 95% CI: 1.4, 2.4), when adjusting for the matched factors and identified confounders.

CONCLUSIONS

The current study suggests that physicians treating IBD patients should be vigilant about expressions of suicidal ideation or signs and symptoms of self-harm.

Acute stress reaction and completed suicide

BACKGROUND

Acute stress reaction is a diagnosis given immediately following the experience of an exceptional mental or physical stressor. To the best of our knowledge, no study has examined the association between acute stress reaction diagnosis and suicide. The current study examined this association in a population-based sample. In addition, we examined comorbid psychiatric diagnoses as modifiers of this association.

METHODS

Data for the current study were obtained from the nationwide Danish health and administrative registries, which include data for all 5.4 million residents of Denmark. All suicides between 1 January 1994 and 31 December 2006 were included and controls were selected from a sample of all Danish residents. Using this nested case-control design, we examined 9612 suicide cases and 199 306 controls matched to cases with respect to gender, date of birth and time.

RESULTS

In total, 95 cases (0.99%) and 165 controls (0.08%) had a diagnosis of acute stress reaction. Those diagnosed with acute stress reaction had 10 times the rate of completed suicide compared with those without this diagnosis, adjusting for the control to case matching, depression and marital status (95% confidence interval 7.7-14). Additionally, persons with acute stress reaction and depression, or acute stress reaction and substance abuse, had a greater rate of suicide than expected based on their independent effects.

CONCLUSIONS

Acute stress reaction is a risk factor for completed suicide.

Use of penicillin and other antibiotics and risk of multiple sclerosis: a population-based case-control study

A 2006 study from the United Kingdom found that penicillin use may decrease the risk of multiple sclerosis (MS). To confirm this finding, the authors conducted a nationwide case-control study in Denmark, using the Danish Multiple Sclerosis Registry to identify 3,259 patients with MS onset from 1996 to 2008, and selected 10 population controls per case (n = 32,590), matched on sex and age. Through the National Prescription Database, prescriptions for antibiotics redeemed from 1995 to 2008 and before the date of first MS symptom/index date were identified. Conditional logistic regression analysis was used to compute odds ratios associating antibiotic use with MS occurrence. In total, 1,922 patients (59%) redeemed penicillin prescriptions before the index date and 2,292 (70%) redeemed any type of antibiotic prescription. Penicillin use was associated with an increased risk of MS (odds ratio = 1.21, 95% confidence interval: 1.10, 1.27). Use of any type of antibiotic was similarly associated with an increased risk of MS (odds ratio = 1.41, 95% confidence interval: 1.29, 1.53). The odds ratios for different types of antibiotics ranged between 1.08 and 1.83. Thus, this study found that penicillin use and use of other antibiotics were similarly associated with increased risk of MS, suggesting that the underlying infections may be causally associated with MS.

Sexual harassment in the Marines, posttraumatic stress symptoms, and perceived health: evidence for sex differences

Sex differences and pretrauma functioning have been understudied in examinations of posttraumatic stress symptoms (PSS) and health. This study examined relationships between sexual harassment and assault in the military (MST), PSS, and perceived physical health when accounting for pre-MST PSS, pre-MST health, and current depression. Relationships were examined separately in 226 female and 91 male Marines endorsing recent MST (past 6 months). MST predicted increased PSS for women and especially men. For men, higher levels of MST were associated with worse perceived physical health, whereas for women, lower levels of MST were associated with worse perceived health. For men with MST, there was some evidence for the association being partially mediated by PSS, but no mediation was found in women.