The impact of COVID-19 on a college freshman sample reveals genetic and nongenetic forms of susceptibility and resilience to stress.

Using a longitudinal approach, we sought to define the interplay between genetic and nongenetic factors in shaping vulnerability or resilience to COVID-19 pandemic stress, as indexed by the emergence of symptoms of depression and/or anxiety. University of Michigan freshmen were characterized at baseline using multiple psychological instruments. Subjects were genotyped, and a polygenic risk score for depression (MDD-PRS) was calculated. Daily physical activity and sleep were captured. Subjects were sampled at multiple time points throughout the freshman year on clinical rating scales, including GAD-7 and PHQ-9 for anxiety and depression, respectively. Two cohorts (2019 to 2021) were compared to a pre-COVID-19 cohort to assess the impact of the pandemic. Across cohorts, 26 to 40% of freshmen developed symptoms of anxiety or depression (N = 331). Depression symptoms significantly increased in the pandemic years and became more chronic, especially in females. Physical activity was reduced, and sleep was increased by the pandemic, and this correlated with the emergence of mood symptoms. While low MDD-PRS predicted lower risk for depression during a typical freshman year, this genetic advantage vanished during the pandemic. Indeed, females with lower genetic risk accounted for the majority of the pandemic-induced rise in depression. We developed a model that explained approximately half of the variance in follow-up depression scores based on psychological trait and state characteristics at baseline and contributed to resilience in genetically vulnerable subjects. We discuss the concept of multiple types of resilience, and the interplay between genetic, sex, and psychological factors in shaping the affective response to different types of stressors.

Unspoken Truths: Mental Health Among Academic Surgeons.

OBJECTIVE: To characterize the current state of mental health within the surgical workforce in the United States. BACKGROUND: Mental illness and suicide is a growing concern in the medical community; however, the current state is largely unknown. METHODS: Cross-sectional survey of the academic surgery community assessing mental health, medical error, and suicidal ideation. The odds of suicidal ideation adjusting for sex, prior mental health diagnosis, and validated scales screening for depression, anxiety, post-traumatic stress disorder (PTSD), and alcohol use disorder were assessed. RESULTS: Of 622 participating medical students, trainees, and surgeons (estimated response rate=11.4%-14.0%), 26.1% (141/539) reported a previous mental health diagnosis. In all, 15.9% (83/523) of respondents screened positive for current depression, 18.4% (98/533) for anxiety, 11.0% (56/510) for alcohol use disorder, and 17.3% (36/208) for PTSD. Medical error was associated with depression (30.7% vs. 13.3%, P <0.001), anxiety (31.6% vs. 16.2%, P =0.001), PTSD (12.8% vs. 5.6%, P =0.018), and hazardous alcohol consumption (18.7% vs. 9.7%, P =0.022). Overall, 13.2% (73/551) of respondents reported suicidal ideation in the past year and 9.6% (51/533) in the past 2 weeks. On adjusted analysis, a previous history of a mental health disorder (aOR: 1.97, 95% CI: 1.04-3.65, P =0.033) and screening positive for depression (aOR: 4.30, 95% CI: 2.21-8.29, P <0.001) or PTSD (aOR: 3.93, 95% CI: 1.61-9.44, P =0.002) were associated with increased odds of suicidal ideation over the past 12 months. CONCLUSIONS: Nearly 1 in 7 respondents reported suicidal ideation in the past year. Mental illness and suicidal ideation are significant problems among the surgical workforce in the United States.

Returning Individual Research Results from Digital Phenotyping in Psychiatry.

Psychiatry is rapidly adopting digital phenotyping and artificial intelligence/machine learning tools to study mental illness based on tracking participants' locations, online activity, phone and text message usage, heart rate, sleep, physical activity, and more. Existing ethical frameworks for return of individual research results (IRRs) are inadequate to guide researchers for when, if, and how to return this unprecedented number of potentially sensitive results about each participant's real-world behavior. To address this gap, we convened an interdisciplinary expert working group, supported by a National Institute of Mental Health grant. Building on established guidelines and the emerging norm of returning results in participant-centered research, we present a novel framework specific to the ethical, legal, and social implications of returning IRRs in digital phenotyping research. Our framework offers researchers, clinicians, and Institutional Review Boards (IRBs) urgently needed guidance, and the principles developed here in the context of psychiatry will be readily adaptable to other therapeutic areas.

Using machine learning with intensive longitudinal data to predict depression and suicidal ideation among medical interns over time.

BACKGROUND: Use of intensive longitudinal methods (e.g. ecological momentary assessment, passive sensing) and machine learning (ML) models to predict risk for depression and suicide has increased in recent years. However, these studies often vary considerably in length, ML methods used, and sources of data. The present study examined predictive accuracy for depression and suicidal ideation (SI) as a function of time, comparing different combinations of ML methods and data sources. METHODS: Participants were 2459 first-year training physicians (55.1% female; 52.5% White) who were provided with Fitbit wearable devices and assessed daily for mood. Linear [elastic net regression (ENR)] and non-linear (random forest) ML algorithms were used to predict depression and SI at the first-quarter follow-up assessment, using two sets of variables (daily mood features only, daily mood features + passive-sensing features). To assess accuracy over time, models were estimated iteratively for each of the first 92 days of internship, using data available up to that point in time. RESULTS: ENRs using only the daily mood features generally had the best accuracy for predicting mental health outcomes, and predictive accuracy within 1 standard error of the full 92 day models was attained by weeks 7-8. Depression at 92 days could be predicted accurately (area under the curve >0.70) after only 14 days of data collection. CONCLUSIONS: Simpler ML methods may outperform more complex methods until passive-sensing features become better specified. For intensive longitudinal studies, there may be limited predictive value in collecting data for more than 2 months.

Trends in Depressive Symptoms and Associated Factors During Residency, 2007 to 2019 : A Repeated Annual Cohort Study.

BACKGROUND: Efforts to address the high depression rates among training physicians have been implemented at various levels of the U.S. medical education system. The cumulative effect of these efforts is unknown. OBJECTIVE: To assess how the increase in depressive symptoms with residency has shifted over time and to identify parallel trends in factors that have previously been associated with resident physician depression. DESIGN: Repeated annual cohort study. SETTING: U.S. health care organizations. PARTICIPANTS: First-year resident physicians (interns) who started training between 2007 and 2019. MEASUREMENTS: Depressive symptoms (9-item Patient Health Questionnaire [PHQ-9]) assessed at baseline and quarterly throughout internship. RESULTS: Among 16 965 interns, baseline depressive symptoms increased from 2007 to 2019 (PHQ-9 score, 2.3 to 2.9; difference, 0.6 [95% CI, 0.3 to 0.8]). The prevalence of baseline predictors of greater increase in depressive symptoms with internship also increased across cohorts. Despite the higher prevalence of baseline risk factors, the average change in depressive symptoms with internship decreased 24.4% from 2007 to 2019 (change in PHQ-9 score, 4.1 to 3.0; difference, -1.0 [CI, -1.5 to -0.6]). This change across cohorts was greater among women (4.7 to 3.3; difference, -1.4 [CI, -1.9 to -0.9]) than men (3.5 to 2.9; difference, -0.6 [CI, -1.2 to -0.05]) and greater among nonsurgical interns (4.1 to 3.0; difference, -1.1 [CI, -1.6 to -0.6]) than surgical interns (4.0 to 3.2; difference, -0.8 [CI, -1.2 to -0.4]). In parallel to the decrease in depressive symptom change, there were increases in sleep hours, quality of faculty feedback, and use of mental health services and a decrease in work hours across cohorts. The decrease in work hours was greater for nonsurgical than surgical interns. Further, the increase in mental health treatment across cohorts was greater for women than men. LIMITATION: Data are observational and subject to biases due to nonrandom sampling, missing data, and unmeasured confounders, limiting causal conclusions. CONCLUSION: Although depression during physician training remains high, the average increase in depressive symptoms associated with internship decreased between 2007 and 2019. PRIMARY FUNDING SOURCE: National Institute of Mental Health.

Exploring the Relationship Between Privacy and Utility in Mobile Health: Algorithm Development and Validation via Simulations of Federated Learning, Differential Privacy, and External Attacks.

BACKGROUND: Although evidence supporting the feasibility of large-scale mobile health (mHealth) systems continues to grow, privacy protection remains an important implementation challenge. The potential scale of publicly available mHealth applications and the sensitive nature of the data involved will inevitably attract unwanted attention from adversarial actors seeking to compromise user privacy. Although privacy-preserving technologies such as federated learning (FL) and differential privacy (DP) offer strong theoretical guarantees, it is not clear how such technologies actually perform under real-world conditions. OBJECTIVE: Using data from the University of Michigan Intern Health Study (IHS), we assessed the privacy protection capabilities of FL and DP against the trade-offs in the associated model's accuracy and training time. Using a simulated external attack on a target mHealth system, we aimed to measure the effectiveness of such an attack under various levels of privacy protection on the target system and measure the costs to the target system's performance associated with the chosen levels of privacy protection. METHODS: A neural network classifier that attempts to predict IHS participant daily mood ecological momentary assessment score from sensor data served as our target system. An external attacker attempted to identify participants whose average mood ecological momentary assessment score is lower than the global average. The attack followed techniques in the literature, given the relevant assumptions about the abilities of the attacker. For measuring attack effectiveness, we collected attack success metrics (area under the curve [AUC], positive predictive value, and sensitivity), and for measuring privacy costs, we calculated the target model training time and measured the model utility metrics. Both sets of metrics are reported under varying degrees of privacy protection on the target. RESULTS: We found that FL alone does not provide adequate protection against the privacy attack proposed above, where the attacker's AUC in determining which participants exhibit lower than average mood is over 0.90 in the worst-case scenario. However, under the highest level of DP tested in this study, the attacker's AUC fell to approximately 0.59 with only a 10% point decrease in the target's R2 and a 43% increase in model training time. Attack positive predictive value and sensitivity followed similar trends. Finally, we showed that participants in the IHS most likely to require strong privacy protection are also most at risk from this particular privacy attack and subsequently stand to benefit the most from these privacy-preserving technologies. CONCLUSIONS: Our results demonstrated both the necessity of proactive privacy protection research and the feasibility of the current FL and DP methods implemented in a real mHealth scenario. Our simulation methods characterized the privacy-utility trade-off in our mHealth setup using highly interpretable metrics, providing a framework for future research into privacy-preserving technologies in data-driven health and medical applications.

Psychosocial Stress and Hypothalamic-Pituitary-Adrenal Axis Stress Reactivity: Variations by Race and Socioeconomic Status Among Adults at Risk of Diabetes.

OBJECTIVE: Although stress is posited to play a key role in health disparities, the extent to which commonly used self-report psychosocial stress measures are related to neurobiological stress processes, especially across diverse populations, is unresolved. This study examined how two measures of psychosocial stress, perceived stress and domain-specific stress, covary with the acute neurobiological stress response. METHODS: The Richmond Stress and Sugar Study includes a racially and socioeconomically diverse cohort of adults at risk for type 2 diabetes ( n = 125; mean age = 57 years, 48% Black, and 61% high neighborhood socioeconomic status [SES]). Hypothalamic-pituitary-adrenal axis reactivity was assessed by salivary cortisol response to the Trier Social Stress Test (TSST), a laboratory stressor. RESULTS: Higher perceived stress was associated with a lower cortisol response to the TSST (-7.5%; 95% confidence interval [CI] = -13.1% to -1.5%) but was not associated with cortisol recovery after the TSST (3%; 95% CI = -0.6% to 6.8%). In contrast, domain-specific stress was not associated with cortisol response (-2.1%; 95% CI = -20.7% to 20.9%) but was inversely associated with cortisol recovery (3.7%; 95% CI = 0.6% to 7.0%). SES modified these associations: both perceived stress and domain-specific stress were associated with TSST cortisol response only among participants from high-SES neighborhoods. There was minimal evidence of effect modification by race. CONCLUSIONS: Both self-report measures of psychosocial stress were associated with hypothalamic-pituitary-adrenal axis reactivity to an acute stressor. These associations varied by perceived versus domain-specific measurement scales and by neighborhood SES. Further efforts to refine stress measures and clarify biological linkages between social status and health are needed.

Deaths Attributed to Psychiatric Disorders in the United States, 2010-2018.

Objectives. To quantify deaths in the United States from 2010 through 2018 that were reported with an underlying cause of death as a psychiatric diagnosis, which do not indicate a clear mechanism of death, and that may be misclassified suicide and overdose deaths.Methods. We used national vital statistics data to identify rates and circumstances of deaths by specific underlying cause of death categories in the US population.Results. There were 115 442 deaths attributed to psychiatric diagnoses and 834 763 deaths attributed to suicide or overdose. The population rate of deaths attributed to psychiatric diagnoses increased from 3.26 to 4.96 per 100 000 US persons between 2010 and 2018.Conclusions. Psychiatric diagnoses may represent a fairly substantial number of misclassified overdose and suicide deaths. Improving mortality surveillance requires improving the accuracy of diagnoses reported on death certificates.

Assessing Real-Time Moderation for Developing Adaptive Mobile Health Interventions for Medical Interns: Micro-Randomized Trial.

BACKGROUND: Individuals in stressful work environments often experience mental health issues, such as depression. Reducing depression rates is difficult because of persistently stressful work environments and inadequate time or resources to access traditional mental health care services. Mobile health (mHealth) interventions provide an opportunity to deliver real-time interventions in the real world. In addition, the delivery times of interventions can be based on real-time data collected with a mobile device. To date, data and analyses informing the timing of delivery of mHealth interventions are generally lacking. OBJECTIVE: This study aimed to investigate when to provide mHealth interventions to individuals in stressful work environments to improve their behavior and mental health. The mHealth interventions targeted 3 categories of behavior: mood, activity, and sleep. The interventions aimed to improve 3 different outcomes: weekly mood (assessed through a daily survey), weekly step count, and weekly sleep time. We explored when these interventions were most effective, based on previous mood, step, and sleep scores. METHODS: We conducted a 6-month micro-randomized trial on 1565 medical interns. Medical internship, during the first year of physician residency training, is highly stressful, resulting in depression rates several folds higher than those of the general population. Every week, interns were randomly assigned to receive push notifications related to a particular category (mood, activity, sleep, or no notifications). Every day, we collected interns' daily mood valence, sleep, and step data. We assessed the causal effect moderation by the previous week's mood, steps, and sleep. Specifically, we examined changes in the effect of notifications containing mood, activity, and sleep messages based on the previous week's mood, step, and sleep scores. Moderation was assessed with a weighted and centered least-squares estimator. RESULTS: We found that the previous week's mood negatively moderated the effect of notifications on the current week's mood with an estimated moderation of -0.052 (P=.001). That is, notifications had a better impact on mood when the studied interns had a low mood in the previous week. Similarly, we found that the previous week's step count negatively moderated the effect of activity notifications on the current week's step count, with an estimated moderation of -0.039 (P=.01) and that the previous week's sleep negatively moderated the effect of sleep notifications on the current week's sleep with an estimated moderation of -0.075 (P<.001). For all three of these moderators, we estimated that the treatment effect was positive (beneficial) when the moderator was low, and negative (harmful) when the moderator was high. CONCLUSIONS: These findings suggest that an individual's current state meaningfully influences their receptivity to mHealth interventions for mental health. Timing interventions to match an individual's state may be critical to maximizing the efficacy of interventions. TRIAL REGISTRATION: ClinicalTrials.gov NCT03972293; http://clinicaltrials.gov/ct2/show/NCT03972293.

Altitude and risk of depression and anxiety: findings from the intern health study.

Multiple studies suggest that the risks of depression and suicide increase with increasing altitude of residence, but no studies have assessed whether changing altitude changes these risks. To address this gap, this study used data from the Intern Health Study, which follows students from the end of medical school through the first year of residency, recording depression via the 9-item Patient Health Questionnaire (PHQ-9), anxiety via the 7-item Generalized Anxiety Disorder Questionnaire (GAD-7), and multiple risk factors for these symptoms. Data from 3764 medical students representing 46 schools and 282 residencies were available. Odds ratios (OR) representing the effects of altitude on psychiatric symptoms were estimated using generalized linear models. After excluding participants with missing altitude data, 3731 medical students were analyzed. High altitude residence (> 900 m) was significantly associated with PHQ-9 total score (OR = 1.32, 95% CI = 1.001-1.75, p < 0.05), and PHQ-9 suicidal ideation (OR = 1.79, 95% CI = 1.08-0.02, p = 0.02). Moving from low to high altitude was significantly associated with PHQ-9 total score (OR = 1.47, 95% CI = 1.087-1.98, p = 0.01), GAD-7 total score (OR = 1.40, 95% CI = 1.0040-1.95, p < 0.05), and PHQ-9 suicidal ideation (OR = 1.10, 95% CI = 1.01-1.19, p = 0.02). The data suggest that moving from low to high altitude is associated with increasing symptoms of depression, anxiety, and suicidal ideation.

Effects of Sleep, Physical Activity, and Shift Work on Daily Mood: a Prospective Mobile Monitoring Study of Medical Interns.

BACKGROUND: Although short sleep, shift work, and physical inactivity are endemic to residency, a lack of objective, real-time information has limited our understanding of how these problems impact physician mental health. OBJECTIVE: To understand how the residency experience affects sleep, physical activity, and mood, and to understand the directional relationships among these variables. DESIGN: A prospective longitudinal study. SUBJECTS: Thirty-three first-year residents (interns) provided data from 2 months pre-internship through the first 6 months of internship. MAIN MEASURES: Objective real-time assessment of daily sleep and physical activity was assessed through accelerometry-based wearable devices. Mood scaled from 1 to 10 was recorded daily using SMS technology. Average compliance rates prior to internship for mood, sleep, and physical activity were 77.4, 80.2, and 93.7%, and were 78.8, 53.0, and 79.9% during internship. KEY RESULTS: After beginning residency, interns lost an average of 2 h and 48 min of sleep per week (t = - 3.04, p < .01). Mood and physical activity decreased by 7.5% (t = - 3.67, p < .01) and 11.5% (t = - 3.15, p < .01), respectively. A bidirectional relationship emerged between sleep and mood during internship wherein short sleep augured worse mood the next day (b = .12, p < .001), which, in turn, presaged shorter sleep the next night (b = .06, p = .03). Importantly, the effect of short sleep on mood was twice as large as mood's effect on sleep. Lastly, substantial shifts in sleep timing during internship (sleeping ≥ 3 h earlier or later than pre-internship patterns) led to shorter sleep (earlier: b = - .36, p < .01; later: b = - 1.75, p < .001) and poorer mood (earlier: b = - .41, p < .001; later: b = - .41, p < .001). CONCLUSIONS: Shift work, short sleep, and physical inactivity confer a challenging environment for physician mental health. Efforts to increase sleep opportunity through designing shift schedules to allow for adequate opportunity to resynchronize the circadian system and improving exercise compatibility of the work environment may improve mood in this depression-vulnerable population.

Prevalence of Burnout Among Physicians: A Systematic Review.

Importance: Burnout is a self-reported job-related syndrome increasingly recognized as a critical factor affecting physicians and their patients. An accurate estimate of burnout prevalence among physicians would have important health policy implications, but the overall prevalence is unknown. Objective: To characterize the methods used to assess burnout and provide an estimate of the prevalence of physician burnout. Data Sources and Study Selection: Systematic search of EMBASE, ERIC, MEDLINE/PubMed, psycARTICLES, and psycINFO for studies on the prevalence of burnout in practicing physicians (ie, excluding physicians in training) published before June 1, 2018. Data Extraction and Synthesis: Burnout prevalence and study characteristics were extracted independently by 3 investigators. Although meta-analytic pooling was planned, variation in study designs and burnout ascertainment methods, as well as statistical heterogeneity, made quantitative pooling inappropriate. Therefore, studies were summarized descriptively and assessed qualitatively. Main Outcomes and Measures: Point or period prevalence of burnout assessed by questionnaire. Results: Burnout prevalence data were extracted from 182 studies involving 109 628 individuals in 45 countries published between 1991 and 2018. In all, 85.7% (156/182) of studies used a version of the Maslach Burnout Inventory (MBI) to assess burnout. Studies variably reported prevalence estimates of overall burnout or burnout subcomponents: 67.0% (122/182) on overall burnout, 72.0% (131/182) on emotional exhaustion, 68.1% (124/182) on depersonalization, and 63.2% (115/182) on low personal accomplishment. Studies used at least 142 unique definitions for meeting overall burnout or burnout subscale criteria, indicating substantial disagreement in the literature on what constituted burnout. Studies variably defined burnout based on predefined cutoff scores or sample quantiles and used markedly different cutoff definitions. Among studies using instruments based on the MBI, there were at least 47 distinct definitions of overall burnout prevalence and 29, 26, and 26 definitions of emotional exhaustion, depersonalization, and low personal accomplishment prevalence, respectively. Overall burnout prevalence ranged from 0% to 80.5%. Emotional exhaustion, depersonalization, and low personal accomplishment prevalence ranged from 0% to 86.2%, 0% to 89.9%, and 0% to 87.1%, respectively. Because of inconsistencies in definitions of and assessment methods for burnout across studies, associations between burnout and sex, age, geography, time, specialty, and depressive symptoms could not be reliably determined. Conclusions and Relevance: In this systematic review, there was substantial variability in prevalence estimates of burnout among practicing physicians and marked variation in burnout definitions, assessment methods, and study quality. These findings preclude definitive conclusions about the prevalence of burnout and highlight the importance of developing a consensus definition of burnout and of standardizing measurement tools to assess the effects of chronic occupational stress on physicians.

Prevalence of Depression, Depressive Symptoms, and Suicidal Ideation Among Medical Students: A Systematic Review and Meta-Analysis.

Importance: Medical students are at high risk for depression and suicidal ideation. However, the prevalence estimates of these disorders vary between studies. Objective: To estimate the prevalence of depression, depressive symptoms, and suicidal ideation in medical students. Data Sources and Study Selection: Systematic search of EMBASE, ERIC, MEDLINE, psycARTICLES, and psycINFO without language restriction for studies on the prevalence of depression, depressive symptoms, or suicidal ideation in medical students published before September 17, 2016. Studies that were published in the peer-reviewed literature and used validated assessment methods were included. Data Extraction and Synthesis: Information on study characteristics; prevalence of depression or depressive symptoms and suicidal ideation; and whether students who screened positive for depression sought treatment was extracted independently by 3 investigators. Estimates were pooled using random-effects meta-analysis. Differences by study-level characteristics were estimated using stratified meta-analysis and meta-regression. Main Outcomes and Measures: Point or period prevalence of depression, depressive symptoms, or suicidal ideation as assessed by validated questionnaire or structured interview. Results: Depression or depressive symptom prevalence data were extracted from 167 cross-sectional studies (n = 116 628) and 16 longitudinal studies (n = 5728) from 43 countries. All but 1 study used self-report instruments. The overall pooled crude prevalence of depression or depressive symptoms was 27.2% (37 933/122 356 individuals; 95% CI, 24.7% to 29.9%, I2 = 98.9%). Summary prevalence estimates ranged across assessment modalities from 9.3% to 55.9%. Depressive symptom prevalence remained relatively constant over the period studied (baseline survey year range of 1982-2015; slope, 0.2% increase per year [95% CI, -0.2% to 0.7%]). In the 9 longitudinal studies that assessed depressive symptoms before and during medical school (n = 2432), the median absolute increase in symptoms was 13.5% (range, 0.6% to 35.3%). Prevalence estimates did not significantly differ between studies of only preclinical students and studies of only clinical students (23.7% [95% CI, 19.5% to 28.5%] vs 22.4% [95% CI, 17.6% to 28.2%]; P = .72). The percentage of medical students screening positive for depression who sought psychiatric treatment was 15.7% (110/954 individuals; 95% CI, 10.2% to 23.4%, I2 = 70.1%). Suicidal ideation prevalence data were extracted from 24 cross-sectional studies (n = 21 002) from 15 countries. All but 1 study used self-report instruments. The overall pooled crude prevalence of suicidal ideation was 11.1% (2043/21 002 individuals; 95% CI, 9.0% to 13.7%, I2 = 95.8%). Summary prevalence estimates ranged across assessment modalities from 7.4% to 24.2%. Conclusions and Relevance: In this systematic review, the summary estimate of the prevalence of depression or depressive symptoms among medical students was 27.2% and that of suicidal ideation was 11.1%. Further research is needed to identify strategies for preventing and treating these disorders in this population.

Web-Based Cognitive Behavioral Therapy Intervention for the Prevention of Suicidal Ideation in Medical Interns: A Randomized Clinical Trial.

IMPORTANCE: In the United States, approximately 1 physician dies by suicide every day. Training physicians are at particularly high risk, with suicidal ideation increasing more than 4-fold during the first 3 months of internship year. Despite this increase, to our knowledge, very few efforts have been made to prevent the escalation of suicidal thoughts among training physicians. OBJECTIVE: To assess the effectiveness of a web-based cognitive behavioral therapy (wCBT) program delivered prior to the start of internship year in the prevention of suicidal ideation in medical interns. DESIGN SETTING AND PARTICIPANTS: A randomized clinical trial conducted at 2 university hospitals with 199 interns from multiple specialties during academic years 2009-2010 or 2011-2012. The current study was conducted from May 2009 to June 2010 and May 2011 to June 2012, and data were analyzed using intent-to-treat principles, including last observation carried forward. INTERVENTIONS: Interns were randomly assigned to 2 study groups (wCBT and attention-control group [ACG]), and completed study activities lasting 30 minutes each week for 4 weeks prior to starting internship year. Participants assigned to wCBT completed online CBT modules and those assigned to ACG received emails with general information about depression, suicidal thinking, and local mental health professionals. MAIN OUTCOMES AND MEASURES: The Patient Health Questionnaire-9 was used to assess suicidal ideation (ie, "thoughts that you would be better off dead or hurting yourself in some way") prior to the start of intern year and at 3-month intervals throughout the year. RESULTS: A total of 62.2% of interns (199 of 320) agreed to take part in the study; 100 were assigned to the wCBT group and 99 to the ACG. During at least 1 point over the course of internship year, 12% of interns (12 of 100) assigned to wCBT endorsed suicidal ideation compared with 21.2% of interns (21 of 99) assigned to ACG. After adjusting for covariates identified a priori that have previously shown to increase the risk for suicidal ideation, interns assigned to wCBT were less likely to endorse suicidal ideation during internship year (relative risk, 0.40; 95% CI, 0.17-0.91; P = .03) compared with those assigned to ACG. CONCLUSIONS AND RELEVANCE: This study demonstrates that a free, easily accessible, brief wCBT program is associated with reduced likelihood of suicidal ideation among medical interns. Prevention programs with these characteristics could be easily disseminated to medical training programs across the country. TRIAL REGISTRATION: anzctr.org.au Identifier: ACTRN12610000628044.