There is a substantial proportion of patients who drop out of treatment before they receive minimally adequate care. They tend to have worse health outcomes than those who complete treatment. Our main goal is to describe the frequency and determinants of dropout from treatment for mental disorders in low-, middle-, and high-income countries.

Respondents from 13 low- or middle-income countries (N = 60 224) and 15 in high-income countries (N = 77 303) were screened for mental and substance use disorders. Cross-tabulations were used to examine the distribution of treatment and dropout rates for those who screened positive. The timing of dropout was examined using Kaplan–Meier curves. Predictors of dropout were examined with survival analysis using a logistic link function.

Dropout rates are high, both in high-income (30%) and low/middle-income (45%) countries. Dropout mostly occurs during the first two visits. It is higher in general medical rather than in specialist settings (nearly 60% v. 20% in lower income settings). It is also higher for mild and moderate than for severe presentations. The lack of financial protection for mental health services is associated with overall increased dropout from care.

Extending financial protection and coverage for mental disorders may reduce dropout. Efficiency can be improved by managing the milder clinical presentations at the entry point to the mental health system, providing adequate training, support and specialist supervision for non-specialists, and streamlining referral to psychiatrists for more severe cases.

Intermittent explosive disorder (IED) is characterised by impulsive anger attacks that vary greatly across individuals in severity and consequence. Understanding IED subtypes has been limited by lack of large, general population datasets including assessment of IED. Using the 17-country World Mental Health surveys dataset, this study examined whether behavioural subtypes of IED are associated with differing patterns of comorbidity, suicidality and functional impairment.

IED was assessed using the Composite International Diagnostic Interview in the World Mental Health surveys (n = 45 266). Five behavioural subtypes were created based on type of anger attack. Logistic regression assessed association of these subtypes with lifetime comorbidity, lifetime suicidality and 12-month functional impairment.

The lifetime prevalence of IED in all countries was 0.8% (s.e.: 0.0). The two subtypes involving anger attacks that harmed people (‘hurt people only’ and ‘destroy property and hurt people’), collectively comprising 73% of those with IED, were characterised by high rates of externalising comorbid disorders. The remaining three subtypes involving anger attacks that destroyed property only, destroyed property and threatened people, and threatened people only, were characterised by higher rates of internalising than externalising comorbid disorders. Suicidal behaviour did not vary across the five behavioural subtypes but was higher among those with (v. those without) comorbid disorders, and among those who perpetrated more violent assaults.

The most common IED behavioural subtypes in these general population samples are associated with high rates of externalising disorders. This contrasts with the findings from clinical studies of IED, which observe a preponderance of internalising disorder comorbidity. This disparity in findings across population and clinical studies, together with the marked heterogeneity that characterises the diagnostic entity of IED, suggests that it is a disorder that requires much greater research.

Epidemiological studies indicate that individuals with one type of mental disorder have an increased risk of subsequently developing other types of mental disorders. This study aimed to undertake a comprehensive analysis of pair-wise lifetime comorbidity across a range of common mental disorders based on a diverse range of population-based surveys.

The WHO World Mental Health (WMH) surveys assessed 145 990 adult respondents from 27 countries. Based on retrospectively-reported age-of-onset for 24 DSM-IV mental disorders, associations were examined between all 548 logically possible temporally-ordered disorder pairs. Overall and time-dependent hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using Cox proportional hazards models. Absolute risks were estimated using the product-limit method. Estimates were generated separately for men and women.

Each prior lifetime mental disorder was associated with an increased risk of subsequent first onset of each other disorder. The median HR was 12.1 (mean = 14.4; range 5.2–110.8, interquartile range = 6.0–19.4). The HRs were most prominent between closely-related mental disorder types and in the first 1–2 years after the onset of the prior disorder. Although HRs declined with time since prior disorder, significantly elevated risk of subsequent comorbidity persisted for at least 15 years. Appreciable absolute risks of secondary disorders were found over time for many pairs.

Survey data from a range of sites confirms that comorbidity between mental disorders is common. Understanding the risks of temporally secondary disorders may help design practical programs for primary prevention of secondary disorders.

A substantial proportion of persons with mental disorders seek treatment from complementary and alternative medicine (CAM) professionals. However, data on how CAM contacts vary across countries, mental disorders and their severity, and health care settings is largely lacking. The aim was therefore to investigate the prevalence of contacts with CAM providers in a large cross-national sample of persons with 12-month mental disorders.

In the World Mental Health Surveys, the Composite International Diagnostic Interview was administered to determine the presence of past 12 month mental disorders in 138 801 participants aged 18–100 derived from representative general population samples. Participants were recruited between 2001 and 2012. Rates of self-reported CAM contacts for each of the 28 surveys across 25 countries and 12 mental disorder groups were calculated for all persons with past 12-month mental disorders. Mental disorders were grouped into mood disorders, anxiety disorders or behavioural disorders, and further divided by severity levels. Satisfaction with conventional care was also compared with CAM contact satisfaction.

An estimated 3.6% (standard error 0.2%) of persons with a past 12-month mental disorder reported a CAM contact, which was two times higher in high-income countries (4.6%; standard error 0.3%) than in low- and middle-income countries (2.3%; standard error 0.2%). CAM contacts were largely comparable for different disorder types, but particularly high in persons receiving conventional care (8.6–17.8%). CAM contacts increased with increasing mental disorder severity. Among persons receiving specialist mental health care, CAM contacts were reported by 14.0% for severe mood disorders, 16.2% for severe anxiety disorders and 22.5% for severe behavioural disorders. Satisfaction with care was comparable with respect to CAM contacts (78.3%) and conventional care (75.6%) in persons that received both.

CAM contacts are common in persons with severe mental disorders, in high-income countries, and in persons receiving conventional care. Our findings support the notion of CAM as largely complementary but are in contrast to suggestions that this concerns person with only mild, transient complaints. There was no indication that persons were less satisfied by CAM visits than by receiving conventional care. We encourage health care professionals in conventional settings to openly discuss the care patients are receiving, whether conventional or not, and their reasons for doing so.

Traumatic events are associated with increased risk of psychotic experiences, but it is unclear whether this association is explained by mental disorders prior to psychotic experience onset.

To investigate the associations between traumatic events and subsequent psychotic experience onset after adjusting for post-traumatic stress disorder and other mental disorders.

We assessed 29 traumatic event types and psychotic experiences from the World Mental Health surveys and examined the associations of traumatic events with subsequent psychotic experience onset with and without adjustments for mental disorders.

Respondents with any traumatic events had three times the odds of other respondents of subsequently developing psychotic experiences (OR=3.1, 95% CI 2.7–3.7), with variability in strength of association across traumatic event types. These associations persisted after adjustment for mental disorders.

Exposure to traumatic events predicts subsequent onset of psychotic experiences even after adjusting for comorbid mental disorders.

Research on post-traumatic stress disorder (PTSD) course finds a substantial proportion of cases remit within 6 months, a majority within 2 years, and a substantial minority persists for many years. Results are inconsistent about pre-trauma predictors.

The WHO World Mental Health surveys assessed lifetime DSM-IV PTSD presence-course after one randomly-selected trauma, allowing retrospective estimates of PTSD duration. Prior traumas, childhood adversities (CAs), and other lifetime DSM-IV mental disorders were examined as predictors using discrete-time person-month survival analysis among the 1575 respondents with lifetime PTSD.

20%, 27%, and 50% of cases recovered within 3, 6, and 24 months and 77% within 10 years (the longest duration allowing stable estimates). Time-related recall bias was found largely for recoveries after 24 months. Recovery was weakly related to most trauma types other than very low [odds-ratio (OR) 0.2–0.3] early-recovery (within 24 months) associated with purposefully injuring/torturing/killing and witnessing atrocities and very low later-recovery (25+ months) associated with being kidnapped. The significant ORs for prior traumas, CAs, and mental disorders were generally inconsistent between early- and later-recovery models. Cross-validated versions of final models nonetheless discriminated significantly between the 50% of respondents with highest and lowest predicted probabilities of both early-recovery (66–55% v. 43%) and later-recovery (75–68% v. 39%).

We found PTSD recovery trajectories similar to those in previous studies. The weak associations of pre-trauma factors with recovery, also consistent with previous studies, presumably are due to stronger influences of post-trauma factors.

Sexual assault is a global concern with post-traumatic stress disorder (PTSD), one of the common sequelae. Early intervention can help prevent PTSD, making identification of those at high risk for the disorder a priority. Lack of representative sampling of both sexual assault survivors and sexual assaults in prior studies might have reduced the ability to develop accurate prediction models for early identification of high-risk sexual assault survivors.

Data come from 12 face-to-face, cross-sectional surveys of community-dwelling adults conducted in 11 countries. Analysis was based on the data from the 411 women from these surveys for whom sexual assault was the randomly selected lifetime traumatic event (TE). Seven classes of predictors were assessed: socio-demographics, characteristics of the assault, the respondent's retrospective perception that she could have prevented the assault, other prior lifetime TEs, exposure to childhood family adversities and prior mental disorders.

Prevalence of Diagnostic and Statistical Manual of Mental Disorders IV (DSM-IV) PTSD associated with randomly selected sexual assaults was 20.2%. PTSD was more common for repeated than single-occurrence victimization and positively associated with prior TEs and childhood adversities. Respondent's perception that she could have prevented the assault interacted with history of mental disorder such that it reduced odds of PTSD, but only among women without prior disorders (odds ratio 0.2, 95% confidence interval 0.1–0.9). The final model estimated that 40.3% of women with PTSD would be found among the 10% with the highest predicted risk.

Whether counterfactual preventability cognitions are adaptive may depend on mental health history. Predictive modelling may be useful in targeting high-risk women for preventive interventions.

Traumatic events are common globally; however, comprehensive population-based cross-national data on the epidemiology of posttraumatic stress disorder (PTSD), the paradigmatic trauma-related mental disorder, are lacking.

Data were analyzed from 26 population surveys in the World Health Organization World Mental Health Surveys. A total of 71 083 respondents ages 18+ participated. The Composite International Diagnostic Interview assessed exposure to traumatic events as well as 30-day, 12-month, and lifetime PTSD. Respondents were also assessed for treatment in the 12 months preceding the survey. Age of onset distributions were examined by country income level. Associations of PTSD were examined with country income, world region, and respondent demographics.

The cross-national lifetime prevalence of PTSD was 3.9% in the total sample and 5.6% among the trauma exposed. Half of respondents with PTSD reported persistent symptoms. Treatment seeking in high-income countries (53.5%) was roughly double that in low-lower middle income (22.8%) and upper-middle income (28.7%) countries. Social disadvantage, including younger age, female sex, being unmarried, being less educated, having lower household income, and being unemployed, was associated with increased risk of lifetime PTSD among the trauma exposed.

PTSD is prevalent cross-nationally, with half of all global cases being persistent. Only half of those with severe PTSD report receiving any treatment and only a minority receive specialty mental health care. Striking disparities in PTSD treatment exist by country income level. Increasing access to effective treatment, especially in low- and middle-income countries, remains critical for reducing the population burden of PTSD.

Although there is robust evidence linking childhood adversities (CAs) and an increased risk for psychotic experiences (PEs), little is known about whether these associations vary across the life-course and whether mental disorders that emerge prior to PEs explain these associations.

We assessed CAs, PEs and DSM-IV mental disorders in 23 998 adults in the WHO World Mental Health Surveys. Discrete-time survival analysis was used to investigate the associations between CAs and PEs, and the influence of mental disorders on these associations using multivariate logistic models.

Exposure to CAs was common, and those who experienced any CAs had increased odds of later PEs [odds ratio (OR) 2.3, 95% confidence interval (CI) 1.9–2.6]. CAs reflecting maladaptive family functioning (MFF), including abuse, neglect, and parent maladjustment, exhibited the strongest associations with PE onset in all life-course stages. Sexual abuse exhibited a strong association with PE onset during childhood (OR 8.5, 95% CI 3.6–20.2), whereas Other CA types were associated with PE onset in adolescence. Associations of other CAs with PEs disappeared in adolescence after adjustment for prior-onset mental disorders. The population attributable risk proportion (PARP) for PEs associated with all CAs was 31% (24% for MFF).

Exposure to CAs is associated with PE onset throughout the life-course, although sexual abuse is most strongly associated with childhood-onset PEs. The presence of mental disorders prior to the onset of PEs does not fully explain these associations. The large PARPs suggest that preventing CAs could lead to a meaningful reduction in PEs in the population.

Research on post-traumatic stress disorder (PTSD) following natural and human-made disasters has been undertaken for more than three decades. Although PTSD prevalence estimates vary widely, most are in the 20–40% range in disaster-focused studies but considerably lower (3–5%) in the few general population epidemiological surveys that evaluated disaster-related PTSD as part of a broader clinical assessment. The World Mental Health (WMH) Surveys provide an opportunity to examine disaster-related PTSD in representative general population surveys across a much wider range of sites than in previous studies.

Although disaster-related PTSD was evaluated in 18 WMH surveys, only six in high-income countries had enough respondents for a risk factor analysis. Predictors considered were socio-demographics, disaster characteristics, and pre-disaster vulnerability factors (childhood family adversities, prior traumatic experiences, and prior mental disorders).

Disaster-related PTSD prevalence was 0.0–3.8% among adult (ages 18+) WMH respondents and was significantly related to high education, serious injury or death of someone close, forced displacement from home, and pre-existing vulnerabilities (prior childhood family adversities, other traumas, and mental disorders). Of PTSD cases 44.5% were among the 5% of respondents classified by the model as having highest PTSD risk.

Disaster-related PTSD is uncommon in high-income WMH countries. Risk factors are consistent with prior research: severity of exposure, history of prior stress exposure, and pre-existing mental disorders. The high concentration of PTSD among respondents with high predicted risk in our model supports the focus of screening assessments that identify disaster survivors most in need of preventive interventions.

This is the first cross-national study of intermittent explosive disorder (IED).

A total of 17 face-to-face cross-sectional household surveys of adults were conducted in 16 countries (n = 88 063) as part of the World Mental Health Surveys initiative. The World Health Organization Composite International Diagnostic Interview (CIDI 3.0) assessed DSM-IV IED, using a conservative definition.

Lifetime prevalence of IED ranged across countries from 0.1 to 2.7% with a weighted average of 0.8%; 0.4 and 0.3% met criteria for 12-month and 30-day prevalence, respectively. Sociodemographic correlates of lifetime risk of IED were being male, young, unemployed, divorced or separated, and having less education. The median age of onset of IED was 17 years with an interquartile range across countries of 13–23 years. The vast majority (81.7%) of those with lifetime IED met criteria for at least one other lifetime disorder; co-morbidity was highest with alcohol abuse and depression. Of those with 12-month IED, 39% reported severe impairment in at least one domain, most commonly social or relationship functioning. Prior traumatic experiences involving physical (non-combat) or sexual violence were associated with increased risk of IED onset.

Conservatively defined, IED is a low prevalence disorder but this belies the true societal costs of IED in terms of the effects of explosive anger attacks on families and relationships. IED is more common among males, the young, the socially disadvantaged and among those with prior exposure to violence, especially in childhood.

The Fukushima Daiichi and Daini Nuclear Power Plant workers experienced multiple stressors as both victims and onsite workers after the 2011 Great East Japan Earthquake and subsequent nuclear accidents. Previous studies found that disaster-related exposures, including discrimination/slurs, were associated with their mental health. Their long-term impact has yet to be investigated.

A total of 968 plant workers (Daiichi, n = 571; Daini, n = 397) completed self-written questionnaires 2–3 months (time 1) and 14–15 months (time 2) after the disaster (response rate 55.0%). Sociodemographics, disaster-related experiences, and peritraumatic distress were assessed at time 1. At time 1 and time 2, general psychological distress (GPD) and post-traumatic stress response (PTSR) were measured, respectively, using the K6 scale and Impact of Event Scale Revised. We examined multivariate covariates of time 2 GPD and PTSR, adjusting for autocorrelations in the hierarchical multiple regression analyses.

Higher GPD at time 2 was predicted by higher GPD at time 1 (β = 0.491, p < 0.001) and discrimination/slurs experiences at time 1 (β = 0.065, p = 0.025, adjusted R 2 = 0.24). Higher PTSR at time 2 was predicted with higher PTSR at time 1 (β = 0.548, p < 0.001), higher age (β = 0.085, p = 0.005), and discrimination/slurs experiences at time 1 (β = 0.079, p = 0.003, adjusted R 2 = 0.36).

Higher GPD at time 2 was predicted by higher GPD and discrimination/slurs experience at time 1. Higher PTSR at time 2 was predicted by higher PTSR, higher age, and discrimination/slurs experience at time 1.

According to diathesis–stress models, personality traits, such as negative emotionality (NE) and positive emotionality (PE), may moderate the effects of stressors on the development of depression. However, relatively little empirical research has directly examined whether NE and PE act as diatheses in the presence of stressful life events, and no research has examined whether they moderate the effect of disaster exposure on depressive symptoms. Hurricane Sandy, the second costliest hurricane in US history, offers a unique opportunity to address these gaps.

A total of 318 women completed measures of NE and PE 5 years prior to Hurricane Sandy. They were also assessed for lifetime depressive disorders on two occasions, the latter occurring an average of 1 year before the hurricane. Approximately 8 weeks after the disaster (mean = 8.40, s.d. = 1.48 weeks), participants completed a hurricane stress exposure questionnaire and a measure of current depressive symptoms.

Adjusting for lifetime history of depressive disorders, higher levels of stress from Hurricane Sandy predicted elevated levels of depressive symptoms, but only in participants with high levels of NE or low levels of PE.

These findings support the role of personality in the development of depression and suggest that personality traits can be useful in identifying those most vulnerable to major stressors, including natural disasters.

Post-traumatic symptomatology is one of the signature effects of the pernicious exposures endured by responders to the World Trade Center (WTC) disaster of 11 September 2001 (9/11), but the long-term extent of diagnosed Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) post-traumatic stress disorder (PTSD) and its impact on quality of life are unknown. This study examines the extent of DSM-IV PTSD 11–13 years after the disaster in WTC responders, its symptom profiles and trajectories, and associations of active, remitted and partial PTSD with exposures, physical health and psychosocial well-being.

Master's-level psychologists administered sections of the Structured Clinical Interview for DSM-IV and the Range of Impaired Functioning Tool to 3231 responders monitored at the Stony Brook University World Trade Center Health Program. The PTSD Checklist (PCL) and current medical symptoms were obtained at each visit.

In all, 9.7% had current, 7.9% remitted, and 5.9% partial WTC-PTSD. Among those with active PTSD, avoidance and hyperarousal symptoms were most commonly, and flashbacks least commonly, reported. Trajectories of symptom severity across monitoring visits showed a modestly increasing slope for active and decelerating slope for remitted PTSD. WTC exposures, especially death and human remains, were strongly associated with PTSD. After adjusting for exposure and critical risk factors, including hazardous drinking and co-morbid depression, PTSD was strongly associated with health and well-being, especially dissatisfaction with life.

This is the first study to demonstrate the extent and correlates of long-term DSM-IV PTSD among responders. Although most proved resilient, there remains a sizable subgroup in need of continued treatment in the second decade after 9/11.

Considerable research has documented that exposure to traumatic events has negative effects on physical and mental health. Much less research has examined the predictors of traumatic event exposure. Increased understanding of risk factors for exposure to traumatic events could be of considerable value in targeting preventive interventions and anticipating service needs.

General population surveys in 24 countries with a combined sample of 68 894 adult respondents across six continents assessed exposure to 29 traumatic event types. Differences in prevalence were examined with cross-tabulations. Exploratory factor analysis was conducted to determine whether traumatic event types clustered into interpretable factors. Survival analysis was carried out to examine associations of sociodemographic characteristics and prior traumatic events with subsequent exposure.

Over 70% of respondents reported a traumatic event; 30.5% were exposed to four or more. Five types – witnessing death or serious injury, the unexpected death of a loved one, being mugged, being in a life-threatening automobile accident, and experiencing a life-threatening illness or injury – accounted for over half of all exposures. Exposure varied by country, sociodemographics and history of prior traumatic events. Being married was the most consistent protective factor. Exposure to interpersonal violence had the strongest associations with subsequent traumatic events.

Given the near ubiquity of exposure, limited resources may best be dedicated to those that are more likely to be further exposed such as victims of interpersonal violence. Identifying mechanisms that account for the associations of prior interpersonal violence with subsequent trauma is critical to develop interventions to prevent revictimization.

Civilian suicide rates vary by occupation in ways related to occupational stress exposure. Comparable military research finds suicide rates elevated in combat arms occupations. However, no research has evaluated variation in this pattern by deployment history, the indicator of occupation stress widely considered responsible for the recent rise in the military suicide rate.

The joint associations of Army occupation and deployment history in predicting suicides were analysed in an administrative dataset for the 729 337 male enlisted Regular Army soldiers in the US Army between 2004 and 2009.

There were 496 suicides over the study period (22.4/100 000 person-years). Only two occupational categories, both in combat arms, had significantly elevated suicide rates: infantrymen (37.2/100 000 person-years) and combat engineers (38.2/100 000 person-years). However, the suicide rates in these two categories were significantly lower when currently deployed (30.6/100 000 person-years) than never deployed or previously deployed (41.2–39.1/100 000 person-years), whereas the suicide rate of other soldiers was significantly higher when currently deployed and previously deployed (20.2–22.4/100 000 person-years) than never deployed (14.5/100 000 person-years), resulting in the adjusted suicide rate of infantrymen and combat engineers being most elevated when never deployed [odds ratio (OR) 2.9, 95% confidence interval (CI) 2.1–4.1], less so when previously deployed (OR 1.6, 95% CI 1.1–2.1), and not at all when currently deployed (OR 1.2, 95% CI 0.8–1.8). Adjustment for a differential ‘healthy warrior effect’ cannot explain this variation in the relative suicide rates of never-deployed infantrymen and combat engineers by deployment status.

Efforts are needed to elucidate the causal mechanisms underlying this interaction to guide preventive interventions for soldiers at high suicide risk.

Previous research suggests that many people receiving mental health treatment do not meet criteria for a mental disorder but are rather ‘the worried well’.

To examine the association of past-year mental health treatment with DSM-IV disorders.

The World Health Organization's World Mental Health (WMH) Surveys interviewed community samples of adults in 23 countries (n = 62 305) about DSM-IV disorders and treatment in the past 12 months for problems with emotions, alcohol or drugs.

Roughly half (52%) of people who received treatment met criteria for a past-year DSM-IV disorder, an additional 18% for a lifetime disorder and an additional 13% for other indicators of need (multiple subthreshold disorders, recent stressors or suicidal behaviours). Dose–response associations were found between number of indicators of need and treatment.

The vast majority of treatment in the WMH countries goes to patients with mental disorders or other problems expected to benefit from treatment.

The Army Study to Assess Risk and Resilience in Servicemembers (Army STARRS) has found that the proportional elevation in the US Army enlisted soldier suicide rate during deployment (compared with the never-deployed or previously deployed) is significantly higher among women than men, raising the possibility of gender differences in the adverse psychological effects of deployment.

Person-month survival models based on a consolidated administrative database for active duty enlisted Regular Army soldiers in 2004–2009 (n = 975 057) were used to characterize the gender × deployment interaction predicting suicide. Four explanatory hypotheses were explored involving the proportion of females in each soldier's occupation, the proportion of same-gender soldiers in each soldier's unit, whether the soldier reported sexual assault victimization in the previous 12 months, and the soldier's pre-deployment history of treated mental/behavioral disorders.

The suicide rate of currently deployed women (14.0/100 000 person-years) was 3.1–3.5 times the rates of other (i.e. never-deployed/previously deployed) women. The suicide rate of currently deployed men (22.6/100 000 person-years) was 0.9–1.2 times the rates of other men. The adjusted (for time trends, sociodemographics, and Army career variables) female:male odds ratio comparing the suicide rates of currently deployed v. other women v. men was 2.8 (95% confidence interval 1.1–6.8), became 2.4 after excluding soldiers with Direct Combat Arms occupations, and remained elevated (in the range 1.9–2.8) after adjusting for the hypothesized explanatory variables.

These results are valuable in excluding otherwise plausible hypotheses for the elevated suicide rate of deployed women and point to the importance of expanding future research on the psychological challenges of deployment for women.

Although variation in the long-term course of major depressive disorder (MDD) is not strongly predicted by existing symptom subtype distinctions, recent research suggests that prediction can be improved by using machine learning methods. However, it is not known whether these distinctions can be refined by added information about co-morbid conditions. The current report presents results on this question.

Data came from 8261 respondents with lifetime DSM-IV MDD in the World Health Organization (WHO) World Mental Health (WMH) Surveys. Outcomes included four retrospectively reported measures of persistence/severity of course (years in episode; years in chronic episodes; hospitalization for MDD; disability due to MDD). Machine learning methods (regression tree analysis; lasso, ridge and elastic net penalized regression) followed by k-means cluster analysis were used to augment previously detected subtypes with information about prior co-morbidity to predict these outcomes.

Predicted values were strongly correlated across outcomes. Cluster analysis of predicted values found three clusters with consistently high, intermediate or low values. The high-risk cluster (32.4% of cases) accounted for 56.6–72.9% of high persistence, high chronicity, hospitalization and disability. This high-risk cluster had both higher sensitivity and likelihood ratio positive (LR+; relative proportions of cases in the high-risk cluster versus other clusters having the adverse outcomes) than in a parallel analysis that excluded measures of co-morbidity as predictors.

Although the results using the retrospective data reported here suggest that useful MDD subtyping distinctions can be made with machine learning and clustering across multiple indicators of illness persistence/severity, replication with prospective data is needed to confirm this preliminary conclusion.

The US Army suicide rate has increased sharply in recent years. Identifying significant predictors of Army suicides in Army and Department of Defense (DoD) administrative records might help focus prevention efforts and guide intervention content. Previous studies of administrative data, although documenting significant predictors, were based on limited samples and models. A career history perspective is used here to develop more textured models.

The analysis was carried out as part of the Historical Administrative Data Study (HADS) of the Army Study to Assess Risk and Resilience in Servicemembers (Army STARRS). De-identified data were combined across numerous Army and DoD administrative data systems for all Regular Army soldiers on active duty in 2004–2009. Multivariate associations of sociodemographics and Army career variables with suicide were examined in subgroups defined by time in service, rank and deployment history.

Several novel results were found that could have intervention implications. The most notable of these were significantly elevated suicide rates (69.6–80.0 suicides per 100 000 person-years compared with 18.5 suicides per 100 000 person-years in the total Army) among enlisted soldiers deployed either during their first year of service or with less than expected (based on time in service) junior enlisted rank; a substantially greater rise in suicide among women than men during deployment; and a protective effect of marriage against suicide only during deployment.

A career history approach produces several actionable insights missed in less textured analyses of administrative data predictors. Expansion of analyses to a richer set of predictors might help refine understanding of intervention implications.