Major depressive disorder subtypes to predict long-term course

A simulation-based network analysis of intervention targets for adolescent depressive and anxiety symptoms

Although previous research has well explored central and bridge symptoms of mental health problems, little examined whether these symptoms can serve as effective targets for intervention practices. Based on the Ising model, this study constructed a network structure of depressive and anxiety symptoms. The NodeIdentifyR algorithm (NIRA) was used to simulate interventions within this network, examining the effects of alleviating or aggravating specific symptoms on the network's sum scores. In this study, a total of 15,569 participants were recruited from China (50.87 % females, Mage = 13.44; SD = 0.97). The Ising model demonstrated that "sad mood" had the highest expected influence, and "irritability" had the highest bridge expected influence. Alleviating interventions suggested that decreasing the symptom value of "nervousness" resulted in the greatest projected reduction in network symptom activation, which may be a potential target symptom for treatment. Aggravating interventions indicated that elevating the symptom value of "sad mood" had the most projected increase in network activation, which may be a potential target for prevention. Additionally, network structure indices (e.g., central or bridge symptoms) need to be interpreted with more caution as intervention targets, since they may not be exactly the same. These findings enriched the comprehension of the depressive and anxiety network in Chinese adolescents, offering valuable insights for designing effective interventions.

Evolution of Depressive Symptoms Among Depression Subtypes of Clinical and Functional Variables in Primary Care in Chile

OBJECTIVE

To compare the evolution of depressive symptoms among depressive subtypes based on clinical and functional variables in a sample of primary care in Chile.

METHODS

A longitudinal study was conducted in the Maule Region of Chile, focusing on 8 primary care from February 2014 to September 2015. Clinical and functional variables, including Mini International Neuro-psychiatric Interview, Outcome Questionnaire interpersonal and social sub-scales, were applied in a latent class analysis. This analysis categorized 210 patients into 3 subtypes: complex depression (N = 100), recurrent depression (n = 96), and unique depression (n = 14). Complex depression, exhibited a higher probability of suicide attempts, interpersonal and social dysfunction, and association with adverse childhood experiences according the Brief Physical and Sexual Abuse Questionnaire. Patients were monitored over 1 year with the Hamilton scale. The Kruskal-Wallis, non-parametric test, followed by paired Mann-Whitney test evaluated difference in the severity of depressive symptoms between the groups. Additionally, data on mental health interventions were collected.

RESULTS

Out of the 210 patients, 89% were women, with a median age of 50 (range 37-58), and 40.1% with only primary education. Sociodemographic characteristics not differ between groups. Significant differences in depressive symptom severity between the groups were found (X2 90.06, P < .001, Kruskal-Wallis test, η2 = 0.084). Post hoc analyses indicated higher depressive symptoms in complex depression compared to recurrent (Z = -9.501, P < .001) and unique (Z = -2.877, P = .004) depression, with no significant difference between recurrent and unique depression (Z = -1.58, P = .113). There were no differences in the number of medical and psychological controls between the groups. The patients with complex depression required greater modifications in the pharmacological prescriptions than those belonging to the other groups.

CONCLUSION

These results provide additional evidence of a complex depression subtype in primary care in Chile associated with adverse childhood experiences, that had worse resolution of depressive symptoms. Contrary to expectations, patients belonging to this group did not receive further medical and psychological interventions, probably due to a lack of specific clinical recommendations.

Mapping network connection and direction between anxiety and depression symptoms across the early, middle, and late adolescents: Insights from a large Chinese sample

Depression and anxiety are two mental disorders prevailing among adolescents. However, issues regarding the trajectory of depression and anxiety are still controversial on both the disease and symptom dimensions. The novel method of network analysis was used to provide insight into the symptom dimension. 20,544 adolescents (female = 10,743, 52.3%) aged between 14 and 24 years (age mean ± sd = 16.9 ± 2.94) were divided into three subgroups according to age so that the course of depression and anxiety could be traced. Network analysis and the Bayesian network model were used in the current study. The results indicated that uncontrollable worry - excessive worry was the most significant edge for all adolescents, whereas concentration - motor had the highest edge weights for early adolescents, and anhedonia - energy was the most critical pairwise symptom for middle and late adolescents. Irritability can bridge anxiety and depression in the early and middle stages of adolescence, while suicide plays a bridging role in the early and late stages of adolescence. Restlessness and guilt can bridge anxiety and depression in middle- and late-stage adolescents, and feeling afraid plays a unique role in middle-stage adolescents. Except for sad mood, which can trigger middle adolescents' anxiety and depression, the other three subgroups were mainly triggered by nervousness. In addition, all results in our current study were shown to be stable and accurate. In treatment, targeting central and triggering symptoms at different stages of adolescence may be critical to alleviating the comorbidity of anxiety and depression.

Depression Subtypes in Systolic Heart Failure: A Secondary Analysis From a Randomized Controlled Trial

BACKGROUND

Heart failure (HF) is associated with an elevated risk of morbidity, mortality, hospitalization, and impaired quality of life. One potential contributor to these poor outcomes is depression. Yet the effectiveness of treatments for depression in patients with HF is mixed, perhaps due to the heterogeneity of depression.

METHODS

This secondary analysis applied latent class analysis (LCA) to data from a clinical trial to classify patients with systolic HF and comorbid depression into LCA subtypes based on depression symptom severity, and then examined whether these subtypes predicted treatment response and mental and physical health outcomes at 12 months follow-up.

RESULTS

In LCA of 629 participants (mean age 63.6 ± 12.9; 43% females), we identified 4 depression subtypes: mild (prevalence 53%), moderate (30%), moderately severe (12%), and severe (5%). The mild subtype was characterized primarily by somatic symptoms of depression (e.g., energy loss, sleep disturbance, poor appetite), while the remaining LCA subtypes additionally included nonsomatic symptoms of depression (e.g., depressed mood, anhedonia, worthlessness). At 12 months, LCA subtypes with more severe depressive symptoms reported significantly greater improvements in mental quality of life and depressive symptoms compared to the LCA mild subtype, but the incidence of cardiovascular- and noncardiovascular-related readmissions, and mortality was similar among all subtypes.

CONCLUSIONS

In patients with depression and systolic heart failure those with the LCA mild depression subtype may not meet full criteria for major depressive disorder, given the overlap between HF and somatic symptoms of depression. We recommend requiring depressed mood or anhedonia as a necessary symptom for major depressive disorder in patients with HF.

Predictive models for first-onset and persistence of depression and anxiety among university students

BACKGROUND

Depression and anxiety are both prevalent among university students. They frequently co-occur and share risk factors. Yet few studies have focused on identifying students at highest risk of first-onset and persistence of either of these conditions.

METHODS

Multicenter cohort study among Spanish first-year university students. At baseline, students were assessed for lifetime and 12-month Major Depressive Episode and/or Generalized Anxiety Disorder (MDE-GAD), other mental disorders, childhood-adolescent adversities, stressful life events, social support, socio-demographics, and psychological factors using web-based surveys; 12-month MDE-GAD was again assessed at 12-month follow-up.

RESULTS

A total of 1253 students participated in both surveys (59.2% of baseline respondents; mean age = 18.7 (SD = 1.3); 56.0% female). First-onset of MDE-GAD at follow-up was 13.3%. Also 46.7% of those with baseline MDE-GAD showed persistence at follow-up. Childhood/Adolescence emotional abuse or neglect (OR= 4.33), prior bipolar spectrum disorder (OR= 4.34), prior suicidal ideation (OR=4.85) and prior lifetime symptoms of MDE (ORs=2.33-3.63) and GAD (ORs=2.15-3.75) were strongest predictors of first-onset MDE-GAD. Prior suicidal ideation (OR=3.17) and prior lifetime GAD symptoms (ORs=2.38-4.02) were strongest predictors of MDE-GAD persistence. Multivariable predictions from baseline showed AUCs of 0.76 for first-onset and 0.81 for persistence. 74.9% of first-onset MDE-GAD cases occurred among 30% students with highest predicted risk at baseline.

LIMITATIONS

Self-report data were used; external validation of the multivariable prediction models is needed.

CONCLUSION

MDE-GAD among university students is frequent, suggesting the need to implement web-based screening at university entrance that identify those students with highest risk.

Predicting non-response to multimodal day clinic treatment in severely impaired depressed patients: a machine learning approach

A considerable number of depressed patients do not respond to treatment. Accurate prediction of non-response to routine clinical care may help in treatment planning and improve results. A longitudinal sample of N = 239 depressed patients was assessed at admission to multi-modal day clinic treatment, after six weeks, and at discharge. First, patient’s treatment response was modelled by identifying longitudinal trajectories using the Hamilton Depression Rating Scale (HDRS-17). Then, individual items of the HDRS-17 at admission as well as individual patient characteristics were entered as predictors of response/non-response trajectories into the binary classification model (eXtremeGradient Boosting; XGBoost). The model was evaluated on a hold-out set and explained in human-interpretable form by SHapley Additive explanation (SHAP) values. The prediction model yielded a multi-class AUC = 0.80 in the hold-out set. The predictive power for the binary classification yielded an AUC = 0.83 (sensitivity = .80, specificity = .77). Most relevant predictors for non-response were insomnia symptoms, younger age, anxiety symptoms, depressed mood, being unemployed, suicidal ideation and somatic symptoms of depressive disorder. Non-responders to routine treatment for depression can be identified and screened for potential next-generation treatments. Such predictors may help personalize treatment and improve treatment response.

Identifying subtypes of depression in clinician-annotated text: a retrospective cohort study

Current criteria for depression are imprecise and do not accurately characterize its distinct clinical presentations. As a result, its diagnosis lacks clinical utility in both treatment and research settings. Data-driven efforts to refine criteria have typically focused on a limited set of symptoms that do not reflect the disorder's heterogeneity. By contrast, clinicians often write about patients in depth, creating descriptions that may better characterize depression. However, clinical text is not commonly used to this end. Here we show that clinically relevant depressive subtypes can be derived from unstructured electronic health records. Five subtypes were identified amongst 18,314 patients with depression treated at a large mental healthcare provider by using unsupervised machine learning: severe-typical, psychotic, mild-typical, agitated, and anergic-apathetic. Subtypes were used to place patients in groups for validation; groups were found to be associated with future outcomes and characteristics that were consistent with the subtypes. These associations suggest that these categorizations are actionable due to their validity with respect to disease prognosis. Moreover, they were derived with automated techniques that might theoretically be widely implemented, allowing for future analyses in more varied populations and settings. Additional research, especially with respect to treatment response, may prove useful in further evaluation.

How well do clinical and demographic characteristics predict Patient Health Questionnaire-9 scores among patients with treatment-resistant major depressive disorder in a real-world setting?

OBJECTIVES

To create and validate a model to predict depression symptom severity among patients with treatment-resistant depression (TRD) using commonly recorded variables within medical claims databases.

METHODS

Adults with TRD (here defined as > 2 antidepressant treatments in an episode, suggestive of nonresponse) and ≥ 1 Patient Health Questionnaire (PHQ)-9 record on or after the index TRD date were identified (2013-2018) in Decision Resource Group's Real World Data Repository, which links an electronic health record database to a medical claims database. A total of 116 clinical/demographic variables were utilized as predictors of the study outcome of depression symptom severity, which was measured by PHQ-9 total score category (score: 0-9 = none to mild, 10-14 = moderate, 15-27 = moderately severe to severe). A random forest approach was applied to develop and validate the predictive model.

RESULTS

Among 5,356 PHQ-9 scores in the study population, the mean (standard deviation) PHQ-9 score was 10.1 (7.2). The model yielded an accuracy of 62.7%. For each predicted depression symptom severity category, the mean observed scores (8.0, 12.2, and 16.2) fell within the appropriate range.

CONCLUSIONS

While there is room for improvement in its accuracy, the use of a machine learning tool that predicts depression symptom severity of patients with TRD can potentially have wide population-level applications. Healthcare systems and payers can build upon this groundwork and use the variables identified and the predictive modeling approach to create an algorithm specific to their population.

Investigation of early and lifetime clinical features and comorbidities for the risk of developing treatment-resistant depression in a 13-year nationwide cohort study

BACKGROUND

To investigate the risk of treatment-resistant depression (TRD) in patients with depression by examining their clinical features, early prescription patterns, and early and lifetime comorbidities.

METHODS

In total, 31,422 depressive inpatients were followed-up from diagnostic onset for more than 10-years. Patients were diagnosed with TRD if their antidepressant treatment regimen was altered ≥two times or if they were admitted after at least two different antidepressant treatments. Multiple Cox regression model were used to determine whether physical and psychiatric comorbidities, psychosis, and prescription patterns increased the risk of TRD by controlling for relevant demographic covariates. Survival analyses were performed for important TRD-associated clinical variables.

RESULTS

Females with depression (21.24%) were more likely to suffer from TRD than males (14.02%). Early anxiety disorders were more commonly observed in the TRD group than in the non-TRD group (81.48 vs. 58.96%, p < 0.0001). Lifetime anxiety disorders had the highest population attributable fraction (42.87%). Seventy percent of patients with multiple psychiatric comorbidities developed TRD during follow-up. Cox regression analysis further identified that functional gastrointestinal disorders significantly increased TRD risk (aHR = 1.19). Higher doses of antidepressants and benzodiazepines and Z drugs in the early course of major depressive disorder increased TRD risk (p < 0.0001).

CONCLUSION

Our findings indicate the need to monitor early comorbidities and polypharmacy patterns in patients with depression associated with elevated TRD risk.

Data mining algorithm predicts a range of adverse outcomes in major depression

BACKGROUND

Course of illness in major depression (MD) is highly varied, which might lead to both under- and overtreatment if clinicians adhere to a 'one-size-fits-all' approach. Novel opportunities in data mining could lead to prediction models that can assist clinicians in treatment decisions tailored to the individual patient. This study assesses the performance of a previously developed data mining algorithm to predict future episodes of MD based on clinical information in new data.

METHODS

We applied a prediction model utilizing baseline clinical characteristics in subjects who reported lifetime MD to two independent test samples (total n = 4226). We assessed the model's performance to predict future episodes of MD, anxiety disorders, and disability during follow-up (1-9 years after baseline). In addition, we compared its prediction performance with well-known risk factors for a severe course of illness.

RESULTS

Our model consistently predicted future episodes of MD in both test samples (AUC 0.68-0.73, modest prediction). Equally accurately, it predicted episodes of generalized anxiety disorder, panic disorder and disability (AUC 0.65-0.78). Our model predicted these outcomes more accurately than risk factors for a severe course of illness such as family history of MD and lifetime traumas.

LIMITATIONS

Prediction accuracy might be different for specific subgroups, such as hospitalized patients or patients with a different cultural background.

CONCLUSIONS

Our prediction model consistently predicted a range of adverse outcomes in MD across two independent test samples derived from studies in different subpopulations, countries, using different measurement procedures. This replication study holds promise for application in clinical practice.

Clustering patients by depression symptoms to predict venlafaxine ER antidepressant efficacy: Individual patient data analysis

OBJECTIVE

To identify clusters of patients with major depressive disorder (MDD) based on the baseline 17-item Hamilton Rating Scale for Depression (HAM-D₁₇) items and to evaluate the efficacy of venlafaxine extended release (VEN) vs placebo, and the potential effect of dose on efficacy, in each cluster.

METHODS

Cluster analysis was performed to identify clusters based on standardized HAM-D₁₇ item scores of individual patient data at baseline from 9 double-blind, placebo-controlled studies of VEN for MDD. Change from baseline in HAM-D₁₇ total score was analyzed using a mixed-effects model for repeated measures for each cluster; response and remission rates at week 8 were analyzed using logistic regression. Discontinuation rates were also evaluated in each cluster.

RESULTS

In 2599 patients, 3 patient clusters were identified, characterized as High modified Core (mCore) Symptoms/High Anxiety (cluster 1), High mCore Symptoms/Medium Anxiety (cluster 2), and Medium mCore Symptoms/Medium Anxiety (cluster 3). Significant effects of VEN vs placebo were observed on change from baseline in HAM-D₁₇ total score at week 8 for both clusters 1 and 2 (both P < 0.001), but not for cluster 3. In cluster 3, a significant treatment effect of VEN was observed at week 8 in the lower-dose subgroup but not in the higher-dose subgroup. All-cause discontinuation rates were significantly higher in placebo than VEN in each cluster.

CONCLUSIONS

Three unique clusters of patients were identified differing in baseline mCore symptoms and anxiety. Cluster membership may predict efficacy outcomes and contribute to dose effects in patients treated with VEN.

CLINICAL TRIALS REGISTRATION

NCT01441440; other studies included in this analysis were conducted before the requirement to register clinical studies took effect.

Stratifying major depressive disorder by polygenic risk for schizophrenia in relation to structural brain measures

BACKGROUND

Substantial clinical heterogeneity of major depressive disorder (MDD) suggests it may group together individuals with diverse aetiologies. Identifying distinct subtypes should lead to more effective diagnosis and treatment, while providing more useful targets for further research. Genetic and clinical overlap between MDD and schizophrenia (SCZ) suggests an MDD subtype may share underlying mechanisms with SCZ.

METHODS

The present study investigated whether a neurobiologically distinct subtype of MDD could be identified by SCZ polygenic risk score (PRS). We explored interactive effects between SCZ PRS and MDD case/control status on a range of cortical, subcortical and white matter metrics among 2370 male and 2574 female UK Biobank participants.

RESULTS

There was a significant SCZ PRS by MDD interaction for rostral anterior cingulate cortex (RACC) thickness (β = 0.191, q = 0.043). This was driven by a positive association between SCZ PRS and RACC thickness among MDD cases (β = 0.098, p = 0.026), compared to a negative association among controls (β = -0.087, p = 0.002). MDD cases with low SCZ PRS showed thinner RACC, although the opposite difference for high-SCZ-PRS cases was not significant. There were nominal interactions for other brain metrics, but none remained significant after correcting for multiple comparisons.

CONCLUSIONS

Our significant results indicate that MDD case-control differences in RACC thickness vary as a function of SCZ PRS. Although this was not the case for most other brain measures assessed, our specific findings still provide some further evidence that MDD in the presence of high genetic risk for SCZ is subtly neurobiologically distinct from MDD in general.

Approaches to Defining Common and Dissociable Neurobiological Deficits Associated With Psychopathology in Youth

Psychiatric disorders show high rates of comorbidity and nonspecificity of presenting clinical symptoms, while demonstrating substantial heterogeneity within diagnostic categories. Notably, many of these psychiatric disorders first manifest in youth. We review progress and next steps in efforts to parse heterogeneity in psychiatric symptoms in youths by identifying abnormalities within neural circuits. To address this fundamental challenge in psychiatry, a number of methods have been proposed. We provide an overview of these methods, broadly organized into dimensional versus categorical approaches and single-view versus multiview approaches. Dimensional approaches including factor analysis and canonical correlation analysis aim to capture dimensional associations between psychopathology and brain measures across a continuous spectrum from health to disease. In contrast, categorical approaches, such as clustering and community detection, aim to identify subtypes of individuals within a class of symptoms or brain features. We highlight several studies that apply these methods to samples of youths and discuss issues to consider when using these approaches. Finally, we end by highlighting avenues for future research.

Machine-learning models for depression and anxiety in individuals with immune-mediated inflammatory disease

OBJECTIVE

Individuals with immune-mediated inflammatory disease (IMID) have a higher prevalence of psychiatric disorders than the general population. We utilized machine-learning to identify patient-reported outcome measures (PROMs) that accurately predict major depressive disorder (MDD) and anxiety disorder in an IMID population.

METHODS

Participants with IMID were enrolled in a cohort study and completed a Structured Clinical Interview for DSM-IV-TR Axis I Disorders (SCID), and multiple PROMs. PROM items were ranked separately for MDD and anxiety disorder by the standardized mean difference between individuals with and without psychiatric disorders. Items were added sequentially to logistic regression (LR), neural network (NN), and random forest (RF) models. Discriminative performance was assessed with area under the receiver operator curve (AUC) and calibration was assessed with Brier scores. Ten-fold cross-validation was used.

RESULTS

Of 637 participants, 75% were female and average age was 51 years. AUC and Brier scores respectively ranged from 0.87-0.91 and 0.07 (i.e., no variation) for MDD models, and from 0.79-0.83 and 0.09-0.11 for anxiety disorder models. In LR and NN, few PROM items were required to obtain optimal discriminatory performance. RF did not perform as well as LR and NN when few PROM items were included.

CONCLUSIONS

Predictive model performance was respectable and revealed insight into PROM items that are predictive of MDD and anxiety disorder. Models that included only the items 'I felt depressed' and 'I felt like I needed help for my anxiety' performed similarly to models that included all items from multiple PROMs.

A methylation study of long-term depression risk

Recurrent and chronic major depressive disorder (MDD) accounts for a substantial part of the disease burden because this course is most prevalent and typically requires long-term treatment. We associated blood DNA methylation profiles from 581 MDD patients at baseline with MDD status 6 years later. A resampling approach showed a highly significant association between methylation profiles in blood at baseline and future disease status (P = 2.0 × 10^-16). Top MWAS results were enriched specific pathways, overlapped with genes found in GWAS of MDD disease status, autoimmune disease and inflammation, and co-localized with eQTLS and (genic enhancers of) of transcription sites in brain and blood. Many of these findings remained significant after correction for multiple testing. The major themes emerging were cellular responses to stress and signaling mechanisms linked to immune cell migration and inflammation. This suggests that an immune signature of treatment-resistant depression is already present at baseline. We also created a methylation risk score (MRS) to predict MDD status 6 years later. The AUC of our MRS was 0.724 and higher than risk scores created using a set of five putative MDD biomarkers, genome-wide SNP data, and 27 clinical, demographic and lifestyle variables. Although further studies are needed to examine the generalizability to different patient populations, these results suggest that methylation profiles in blood may present a promising avenue to support clinical decision making by providing empirical information about the likelihood MDD is chronic or will recur in the future.

The Predictive Validity of Machine Learning Models in the Classification and Treatment of Major Depressive Disorder: State of the Art and Future Directions

Major depressive disorder imposes a substantial disease burden worldwide, ranking as the third leading contributor to global disability. In spite of its ubiquity, classifying and treating depression has proven troublesome. One argument put forward to explain this predicament is the heterogeneity of patients diagnosed with the disorder. Recently, many areas of daily life have witnessed the surge of machine learning techniques, computational approaches to elucidate complex patterns in large datasets, which can be employed to make predictions and detect relevant clusters. Due to the multidimensionality at play in the pathogenesis of depression, it is suggested that machine learning could contribute to improving classification and treatment. In this paper, we investigated literature focusing on the use of machine learning models on datasets with clinical variables of patients diagnosed with depression to predict treatment outcomes or find more homogeneous subgroups. Identified studies based on best practices in the field are evaluated. We found 16 studies predicting outcomes (such as remission) and identifying clusters in patients with depression. The identified studies are mostly still in proof-of-concept phase, with small datasets, lack of external validation, and providing single performance metrics. Larger datasets, and models with similar variables present across these datasets, are needed to develop accurate and generalizable models. We hypothesize that harnessing natural language processing to obtain data 'hidden' in clinical texts might prove useful in improving prediction models. Besides, researchers will need to focus on the conditions to feasibly implement these models to support psychiatrists and patients in their decision-making in practice. Only then we can enter the realm of precision psychiatry.

Distinct Symptom-Specific Treatment Targets for Circuit-Based Neuromodulation

OBJECTIVE

Treatment of different depression symptoms may require different brain stimulation targets with different underlying brain circuits. The authors sought to identify such targets, which could improve the efficacy of therapeutic brain stimulation and facilitate personalized therapy.

METHODS

The authors retrospectively analyzed two independent cohorts of patients who received left prefrontal transcranial magnetic stimulation (TMS) for treatment of depression (discovery sample, N=30; active replication sample, N=81; sham replication sample, N=87). Each patient's TMS site was mapped to underlying brain circuits using functional connectivity MRI from a large connectome database (N=1,000). Circuits associated with improvement in each depression symptom were identified and then clustered based on similarity. The authors tested for reproducibility across data sets and whether symptom-specific targets derived from one data set could predict symptom improvement in the other independent cohort.

RESULTS

The authors identified two distinct circuit targets effective for two discrete clusters of depressive symptoms. Dysphoric symptoms, such as sadness and anhedonia, responded best to stimulation of one circuit, while anxiety and somatic symptoms responded best to stimulation of a different circuit. These circuit maps were reproducible, predicted symptom improvement in independent patient cohorts, and were specific to active compared with sham stimulation. The maps predicted symptom improvement in an exploratory analysis of stimulation sites from 14 clinical TMS trials.

CONCLUSIONS

Distinct clusters of depressive symptoms responded better to different TMS targets across independent retrospective data sets. These symptom-specific targets can be prospectively tested in a randomized clinical trial. This data-driven approach for identifying symptom-specific targets may prove useful for other disorders and facilitate personalized neuromodulation therapy.

Prospective prediction of suicide attempts in community adolescents and young adults, using regression methods and machine learning

BACKGROUND

The use of machine learning (ML) algorithms to study suicidality has recently been recommended. Our aim was to explore whether ML approaches have the potential to improve the prediction of suicide attempt (SA) risk. Using the epidemiological multiwave prospective-longitudinal Early Developmental Stages of Psychopathology (EDSP) data set, we compared four algorithms-logistic regression, lasso, ridge, and random forest-in predicting a future SA in a community sample of adolescents and young adults.

METHODS

The EDSP Study prospectively assessed, over the course of 10 years, adolescents and young adults aged 14-24 years at baseline. Of 3021 subjects, 2797 were eligible for prospective analyses because they participated in at least one of the three follow-up assessments. Sixteen baseline predictors, all selected a priori from the literature, were used to predict follow-up SAs. Model performance was assessed using repeated nested 10-fold cross-validation. As the main measure of predictive performance we used the area under the curve (AUC).

RESULTS

The mean AUCs of the four predictive models, logistic regression, lasso, ridge, and random forest, were 0.828, 0.826, 0.829, and 0.824, respectively.

CONCLUSIONS

Based on our comparison, each algorithm performed equally well in distinguishing between a future SA case and a non-SA case in community adolescents and young adults. When choosing an algorithm, different considerations, however, such as ease of implementation, might in some instances lead to one algorithm being prioritized over another. Further research and replication studies are required in this regard.

Development and internal validation of a depression severity prediction model for tinnitus patients based on questionnaire responses and socio-demographics

Tinnitus is a complex condition that is associated with major psychological and economic impairments – partly through various comorbidities such as depression. Understanding the interaction between tinnitus and depression may thus improve either symptom cluster’s prevention, diagnosis and treatment. In this study, we developed and validated a machine learning model to predict depression severity after outpatient therapy (T1) based on variables obtained before therapy (T0). 1,490 patients with chronic tinnitus (comorbid major depressive disorder: 52.2%) who completed a 7-day multimodal treatment encompassing tinnitus-specific components, cognitive behavioural therapy, physiotherapy and informational counselling were included. 185 variables were extracted from self-report questionnaires and socio-demographic data acquired at T0. We used 11 classification methods to train models that reliably separate between subclinical and clinical depression at T1 as measured by the general depression questionnaire. To ensure highly predictive and robust classifiers, we tuned algorithm hyperparameters in a 10-fold cross-validation scheme. To reduce model complexity and improve interpretability, we wrapped model training around an incremental feature selection mechanism that retained features that contributed to model prediction. We identified a LASSO model that included all 185 features to yield highest predictive performance (AUC = 0.87 ± 0.04). Through our feature selection wrapper, we identified a LASSO model with good trade-off between predictive performance and interpretability that used only 6 features (AUC = 0.85 ± 0.05). Thus, predictive machine learning models can lead to a better understanding of depression in tinnitus patients, and contribute to the selection of suitable therapeutic strategies and concise and valid questionnaire design for patients with chronic tinnitus with or without comorbid major depressive disorder.

Toward a New Model of Understanding, Preventing, and Treating Adolescent Depression Focusing on Exhaustion and Stress

OBJECTIVE

Adolescent depression is a heterogeneous disorder, with a wide variety of symptoms and inconsistent treatment response, and is not completely understood. A dysregulated stress system is a consistent finding, however, and exhaustion is a consistent trait in adolescent patients. The aim of this paper is to critically assess current hypotheses in adolescent depression research and reframe causes and treatment approaches.

METHODS

A mixed-method approach involved a review based on publications from PubMed, Embase and PsycInfo, and two exemplary adolescent cases.

RESULTS

Both cases show a spiral of stress and exhaustion, but with a different profile of symptoms and coping mechanisms. Reframing both cases from the perspective of coping behavior, searching for the sources of experienced stress and exhaustion, showed coping similarities. This proved essential in the successful personalized treatment and recovery process. In combination with recent evidence, both cases support the functional reframing of depression as the outcome of a stress- and exhaustion-related spiralling mechanism.

CONCLUSIONS

We propose to open up a symptom-based, mood-centered view to a model in which adolescent depression is framed as a consecutive failure of stress coping mechanisms and chronic exhaustion. Addressing exhaustion and coping primarily as a treatment strategy in adolescents and young adults might work in synergy with existing treatments and improve overall outcomes. This perspective warrants further investigation.

Identifying depression in the National Health and Nutrition Examination Survey data using a deep learning algorithm

BACKGROUND

As depression is the leading cause of disability worldwide, large-scale surveys have been conducted to establish the occurrence and risk factors of depression. However, accurately estimating epidemiological factors leading up to depression has remained challenging. Deep-learning algorithms can be applied to assess the factors leading up to prevalence and clinical manifestations of depression.

METHODS

Customized deep-neural-network and machine-learning classifiers were assessed using survey data from 19,725 participants from the NHANES database (from 1999 through 2014) and 4949 from the South Korea NHANES (K-NHANES) database in 2014.

RESULTS

A deep-learning algorithm showed area under the receiver operating characteristic curve (AUCs) of 0.91 and 0.89 for detecting depression in NHANES and K-NHANES, respectively. The deep-learning algorithm trained with serial datasets (NHANES, from 1999 to 2012), predicted the prevalence of depression in the following two years of data (NHANES, 2013 and 2014) with an AUC of 0.92. Machine learning classifiers trained with NHANES could further predict depression in K-NHANES. There, logistic regression had the highest performance (AUC, 0.77) followed by deep learning algorithm (AUC, 0.74).

CONCLUSIONS

Deep neural-networks managed to identify depression well from other health and demographic factors in both the NHANES and K-NHANES datasets. The deep-learning algorithm was also able to predict depression relatively well on new data set-cross temporally and cross nationally. Further research can delineate the clinical implications of machine learning and deep learning in detecting disease prevalence and progress as well as other risk factors for depression and other mental illnesses.

Progress in psychoradiology, the clinical application of psychiatric neuroimaging

Psychoradiology is an emerging field that applies radiological imaging technologies to psychiatric conditions. In the past three decades, brain imaging techniques have rapidly advanced understanding of illness and treatment effects in psychiatry. Based on these advances, radiologists have become increasingly interested in applying these advances for differential diagnosis and individualized patient care selection for common psychiatric illnesses. This shift from research to clinical practice represents the beginning evolution of psychoradiology. In this review, we provide a summary of recent progress relevant to this field based on their clinical functions, namely the (1) classification and subtyping; (2) prediction and monitoring of treatment outcomes; and (3) treatment selection. In addition, we provide guidelines for the practice of psychoradiology in clinical settings and suggestions for future research to validate broader clinical applications. Given the high prevalence of psychiatric disorders and the importance of increased participation of radiologists in this field, a guide regarding advances in this field and a description of relevant clinical work flow patterns help radiologists contribute to this fast-evolving field.

Results of the European Group for the Study of Resistant Depression (GSRD) - basis for further research and clinical practice

Objectives: The overview outlines two decades of research from the European Group for the Study of Resistant Depression (GSRD) that fundamentally impacted evidence-based algorithms for diagnostics and psychopharmacotherapy of treatment-resistant depression (TRD). Methods: The GSRD staging model characterising response, non-response and resistance to antidepressant (AD) treatment was applied to 2762 patients in eight European countries. Results: In case of non-response, dose escalation and switching between different AD classes did not show superiority over continuation of original AD treatment. Predictors for TRD were symptom severity, duration of the current major depressive episode (MDE), suicidality, psychotic and melancholic features, comorbid anxiety and personality disorders, add-on treatment, non-response to the first AD, adverse effects, high occupational level, recurrent disease course, previous hospitalisations, positive family history of MDD, early age of onset and novel associations of single nucleoid polymorphisms (SNPs) within the PPP3CC, ST8SIA2, CHL1, GAP43 and ITGB3 genes and gene pathways associated with neuroplasticity, intracellular signalling and chromatin silencing. A prediction model reaching accuracy of above 0.7 highlighted symptom severity, suicidality, comorbid anxiety and lifetime MDEs as the most informative predictors for TRD. Applying machine-learning algorithms, a signature of three SNPs of the BDNF, PPP3CC and HTR2A genes and lacking melancholia predicted treatment response. Conclusions: The GSRD findings offer a unique and balanced perspective on TRD representing foundation for further research elaborating on specific clinical and genetic hypotheses and treatment strategies within appropriate study-designs, especially interaction-based models and randomized controlled trials.

Clinical factors predicting treatment resistant depression: affirmative results from the European multicenter study

OBJECTIVES

Clinical variables were investigated in the 'treatment resistant depression (TRD)- III' sample to replicate earlier findings by the European research consortium 'Group for the Study of Resistant Depression' (GSRD) and enable cross-sample prediction of treatment outcome in TRD.

EXPERIMENTAL PROCEDURES

TRD was defined by a Montgomery and Åsberg Depression Rating Scale (MADRS) score ≥22 after at least two antidepressive trials. Response was defined by a decline in MADRS score by ≥50% and below a threshold of 22. Logistic regression was applied to replicate predictors for TRD among 16 clinical variables in 916 patients. Elastic net regression was applied for prediction of treatment outcome.

RESULTS

Symptom severity (odds ratio (OR) = 3.31), psychotic symptoms (OR = 2.52), suicidal risk (OR = 1.74), generalized anxiety disorder (OR = 1.68), inpatient status (OR = 1.65), higher number of antidepressants administered previously (OR = 1.23), and lifetime depressive episodes (OR = 1.15) as well as longer duration of the current episode (OR = 1.022) increased the risk of TRD. Prediction of TRD reached an accuracy of 0.86 in the independent validation set, TRD-I.

CONCLUSION

Symptom severity, suicidal risk, higher number of lifetime depressive episodes, and comorbid anxiety disorder were replicated as the most prominent risk factors for TRD. Significant predictors in TRD-III enabled robust prediction of treatment outcome in TRD-I.

Finding factors that predict treatment-resistant depression: Results of a cohort study

BACKGROUND

Treatment for depressive disorders often requires subsequent interventions. Patients who do not respond to antidepressants have treatment-resistant depression (TRD). Predicting who will develop TRD may help healthcare providers make more effective treatment decisions. We sought to identify factors that predict TRD in a real-world setting using claims databases.

METHODS

A retrospective cohort study was conducted in a US claims database of adult subjects with newly diagnosed and treated depression with no mania, dementia, and psychosis. The index date was the date of antidepressant dispensing. The outcome was TRD, defined as having at least three distinct antidepressants or one antidepressant and one antipsychotic within 1 year after the index date. Predictors were age, gender, medical conditions, medications, and procedures 1 year before the index date.

RESULTS

Of 230,801 included patients, 10.4% developed TRD within 1 year. TRD patients at baseline were younger; 10.87% were between 18 and 19 years old versus 7.64% in the no-TRD group, risk ratio (RR) = 1.42 (95% confidence interval [CI] 1.37-1.48). TRD patients were more likely to have an anxiety disorder at baseline than non-TRD patients, RR = 1.38 (95% CI 1.35-1.14). At 3.68, fatigue had the highest RR (95% CI 3.18-4.25). TRD patients had substance use disorders, psychiatric conditions, insomnia, and pain more often at baseline than non-TRD patients.

CONCLUSION

Ten percent of subjects newly diagnosed and treated for depression developed TRD within a year. They were younger and suffered more frequently from fatigue, substance use disorders, anxiety, psychiatric conditions, insomnia, and pain than non-TRD patients.

Transdiagnostic Symptom Clusters and Associations With Brain, Behavior, and Daily Function in Mood, Anxiety, and Trauma Disorders

IMPORTANCE

The symptoms that define mood, anxiety, and trauma disorders are highly overlapping across disorders and heterogeneous within disorders. It is unknown whether coherent subtypes exist that span multiple diagnoses and are expressed functionally (in underlying cognition and brain function) and clinically (in daily function). The identification of cohesive subtypes would help disentangle the symptom overlap in our current diagnoses and serve as a tool for tailoring treatment choices.

OBJECTIVE

To propose and demonstrate 1 approach for identifying subtypes within a transdiagnostic sample.

DESIGN, SETTING, AND PARTICIPANTS

This cross-sectional study analyzed data from the Brain Research and Integrative Neuroscience Network Foundation Database that had been collected at the University of Sydney and University of Adelaide between 2006 and 2010 and replicated at Stanford University between 2013 and 2017. The study included 420 individuals with a primary diagnosis of major depressive disorder (n = 100), panic disorder (n = 53), posttraumatic stress disorder (n = 47), or no disorder (healthy control participants) (n = 220). Data were analyzed between October 2016 and October 2017.

MAIN OUTCOMES AND MEASURES

We followed a data-driven approach to achieve the primary study outcome of identifying transdiagnostic subtypes. First, machine learning with a hierarchical clustering algorithm was implemented to classify participants based on self-reported negative mood, anxiety, and stress symptoms. Second, the robustness and generalizability of the subtypes were tested in an independent sample. Third, we assessed whether symptom subtypes were expressed at behavioral and physiological levels of functioning. Fourth, we evaluated the clinically meaningful differences in functional capacity of the subtypes. Findings were interpreted relative to a complementary diagnostic frame of reference.

RESULTS

Four hundred twenty participants with a mean (SD) age of 39.8 (14.1) years were included in the final analysis; 256 (61.0%) were female. We identified 6 distinct subtypes characterized by tension (n=81; 19%), anxious arousal (n=55; 13%), general anxiety (n=38; 9%), anhedonia (n=29; 7%), melancholia (n=37; 9%), and normative mood (n=180; 43%), and these subtypes were replicated in an independent sample. Subtypes were expressed through differences in cognitive control (F5,383 = 5.13, P < .001, ηp2 = 0.063), working memory (F5,401 = 3.29, P = .006, ηp2 = 0.039), electroencephalography-recorded β power in a resting paradigm (F5,357 = 3.84, P = .002, ηp2 = 0.051), electroencephalography-recorded β power in an emotional paradigm (F5,365 = 3.56, P = .004, ηp2 = 0.047), social functional capacity (F5,414 = 21.33, P < .001, ηp2 = 0.205), and emotional resilience (F5,376 = 15.10, P < .001, ηp2 = 0.171).

CONCLUSIONS AND RELEVANCE

These findings offer a data-driven framework for identifying robust subtypes that signify specific, coherent, meaningful associations between symptoms, behavior, brain function, and observable real-world function, and that cut across DSM-IV-defined diagnoses of major depressive disorder, panic disorder, and posttraumatic stress disorder.

The centrality of DSM and non-DSM depressive symptoms in Han Chinese women with major depression

INTRODUCTION

We compared DSM-IV criteria for major depression (MD) with clinically selected non-DSM criteria in their ability to represent clinical features of depression.

METHOD

We conducted network analyses of 19 DSM and non-DSM symptoms of MD assessed at personal interview in 5952 Han Chinese women meeting DSM-IV criteria for recurrent MD. We estimated an Ising model (the state-of-the-art network model for binary data), compared the centrality (interconnectedness) of DSM-IV and non-DSM symptoms, and investigated the community structure (symptoms strongly clustered together).

RESULTS

The DSM and non-DSM criteria were intermingled within the same symptom network. In both the DSM-IV and non-DSM criteria sets, some symptoms were central (highly interconnected) while others were more peripheral. The mean centrality of the DSM and non-DSM criteria sets did not significantly differ. In at least two cases, non-DSM criteria were more central than symptomatically related DSM criteria: lowered libido vs. sleep and appetite changes, and hopelessness versus worthlessness. The overall network had three sub-clusters reflecting neurovegetative/mood symptoms, cognitive changes and anxiety/irritability.

LIMITATIONS

The sample were severely ill Han Chinese females limiting generalizability.

CONCLUSIONS

Consistent with prior historical reviews, our results suggest that the DSM-IV criteria for MD reflect one possible sub-set of a larger pool of plausible depressive symptoms and signs. While the DSM criteria on average perform well, they are not unique and may not be optimal in their ability to describe the depressive syndrome.

Olfactory bulbectomy in mice triggers transient and long-lasting behavioral impairments and biochemical hippocampal disturbances

Major depressive disorder (MDD) is a neuropsychiatric disease that is associated with profound disturbances in affected individuals. Elucidating the pathophysiology of MDD has been frustratingly slow, especially concerning the neurochemical events and brain regions associated with disease progression. Thus, we evaluated the time-course (up to 8weeks) behavioral and biochemical effects in mice that underwent to a bilateral olfactory bulbectomy (OBX), which is used to modeling depressive-like behavior in rodents. Similar to the symptoms in patients with MDD, OBX induced long-lasting (e.g., impairment of habituation to novelty, hyperactivity and an anxiety-like phenotype) and transient (e.g., loss of self-care and motivational behavior) behavioral effects. Moreover, OBX temporarily impaired hippocampal synaptosomal mitochondria, in a manner that would be associated with hippocampal-related synaptotoxicity. Finally, long-lasting pro-oxidative (i.e., increased levels of reactive oxygen species and nitric oxide and decreased glutathione levels) and pro-inflammatory (i.e., increased levels of pro-inflammatory cytokines IL-1, IL-6, TNF-α and decreased anti-inflammatory cytokine IL-10 levels) effects were induced in the hippocampus by OBX. Additionally, these parameters were transiently affected in the posterior and frontal cortices. This study is the first to suggest that the transient and long-lasting behavioral effects from OBX strongly correlate with mitochondrial, oxidative and inflammatory parameters in the hippocampus; furthermore, these effects show a weak correlation with these parameters in the cortex. Our findings highlight the underlying mechanisms involved in the biochemical time course of events related to depressive behavior.

Effects of levomilnacipran ER on noradrenergic symptoms, anxiety symptoms, and functional impairment in adults with major depressive disorder: Post hoc analysis of 5 clinical trials

OBJECTIVE

To evaluate the effects of levomilnacipran extended-release (LVM-ER; 40-120mg/day) on noradrenergic (NA) and anxiety-related symptoms in adults with major depressive disorder (MDD) and explore the relationship between these symptoms and functional impairment.

METHODS

Data were pooled from 5 randomized, double-blind, placebo-controlled trials (N=2598). Anxiety and NA Cluster scores were developed by adding selected item scores from the Montgomery-Åsberg Depression Rating Scale (MADRS) and 17-item Hamilton Depression Rating Scale (HAMD₁₇). A path analysis was conducted to estimate the direct effects of LVM-ER on functional impairment (Sheehan Disability Scale [SDS] total score) and the indirect effects through changes in NA and Anxiety Cluster scores.

RESULTS

Mean improvements from baseline in NA and Anxiety Cluster scores were significantly greater with LVM-ER versus placebo (both P<0.001), as were the response rates (≥50% score improvement): NA Cluster (44% vs 34%; odds ratio=1.56; P<0.0001); Anxiety Cluster (39% vs 36%; odds ratio=1.19; P=0.041). Mean improvement in SDS total score was also significantly greater with LVM-ER versus placebo (-7.3 vs -5.6; P<0.0001). LVM-ER had an indirect effect on change in SDS total score that was mediated more strongly through NA Cluster score change (86%) than Anxiety Cluster score change (18%); the direct effect was negligible.

LIMITATIONS

NA and Anxiety Cluster scores, developed based on the face validity of individual MADRS and HAMD₁₇ items, were not predefined as efficacy outcomes in any of the studies.

CONCLUSION

In adults with MDD, LVM-ER indirectly improved functional impairment mainly through improvements in NA symptoms and less so via anxiety symptoms.

Using patient self-reports to study heterogeneity of treatment effects in major depressive disorder

BACKGROUNDS

Clinicians need guidance to address the heterogeneity of treatment responses of patients with major depressive disorder (MDD). While prediction schemes based on symptom clustering and biomarkers have so far not yielded results of sufficient strength to inform clinical decision-making, prediction schemes based on big data predictive analytic models might be more practically useful.

METHOD

We review evidence suggesting that prediction equations based on symptoms and other easily-assessed clinical features found in previous research to predict MDD treatment outcomes might provide a foundation for developing predictive analytic clinical decision support models that could help clinicians select optimal (personalised) MDD treatments. These methods could also be useful in targeting patient subsamples for more expensive biomarker assessments.

RESULTS

Approximately two dozen baseline variables obtained from medical records or patient reports have been found repeatedly in MDD treatment trials to predict overall treatment outcomes (i.e., intervention v. control) or differential treatment outcomes (i.e., intervention A v. intervention B). Similar evidence has been found in observational studies of MDD persistence-severity. However, no treatment studies have yet attempted to develop treatment outcome equations using the full set of these predictors. Promising preliminary empirical results coupled with recent developments in statistical methodology suggest that models could be developed to provide useful clinical decision support in personalised treatment selection. These tools could also provide a strong foundation to increase statistical power in focused studies of biomarkers and MDD heterogeneity of treatment response in subsequent controlled trials.

CONCLUSIONS

Coordinated efforts are needed to develop a protocol for systematically collecting information about established predictors of heterogeneity of MDD treatment response in large observational treatment studies, applying and refining these models in subsequent pragmatic trials, carrying out pooled secondary analyses to extract the maximum amount of information from these coordinated studies, and using this information to focus future discovery efforts in the segment of the patient population in which continued uncertainty about treatment response exists.

The efficacy of levomilnacipran ER across symptoms of major depressive disorder: a post hoc analysis of 5 randomized, double-blind, placebo-controlled trials

OBJECTIVE

A post hoc analysis evaluated the effects of levomilnacipran ER on individual symptoms and symptom domains in adults with major depressive disorder (MDD).

METHODS

Data were pooled from 5 Phase III trials comprising 2598 patients. Effects on depression symptoms were analyzed based on change from baseline in individual Montgomery-Åsberg Depression Rating Scale (MADRS) item scores. A1dditional evaluations included resolution of individual symptoms (defined as a MADRS item score ≤1 at end of treatment) and concurrent resolution of all 10 MADRS items, all MADRS6 subscale items, and all items included in different symptom clusters (Dysphoria, Retardation, Vegetative Symptoms, Anhedonia).

RESULTS

Significantly greater mean improvements were found on all MADRS items except Reduced Appetite with levomilnacipran ER treatment compared with placebo. Resolution of individual symptoms occurred more frequently with levomilnacipran ER than placebo for each MADRS item (all P<.05), with odds ratios (ORs) ranging from 1.26 to 1.75; resolution of all 10 items was also greater with levomilnacipran ER (OR=1.57; P=.0051). Significant results were found for the MADRS6 subscale (OR=1.73; P<.0001) and each symptom cluster (OR range, 1.39 [Vegetative Symptoms] to 1.84 [Retardation]; all clusters, P<.01).

CONCLUSION

Adult MDD patients treated with levomilnacipran ER improved across a range of depression symptoms and symptom domains.

Testing a machine-learning algorithm to predict the persistence and severity of major depressive disorder from baseline self-reports

Heterogeneity of major depressive disorder (MDD) illness course complicates clinical decision-making. Although efforts to use symptom profiles or biomarkers to develop clinically useful prognostic subtypes have had limited success, a recent report showed that machine-learning (ML) models developed from self-reports about incident episode characteristics and comorbidities among respondents with lifetime MDD in the World Health Organization World Mental Health (WMH) Surveys predicted MDD persistence, chronicity and severity with good accuracy. We report results of model validation in an independent prospective national household sample of 1056 respondents with lifetime MDD at baseline. The WMH ML models were applied to these baseline data to generate predicted outcome scores that were compared with observed scores assessed 10-12 years after baseline. ML model prediction accuracy was also compared with that of conventional logistic regression models. Area under the receiver operating characteristic curve based on ML (0.63 for high chronicity and 0.71-0.76 for the other prospective outcomes) was consistently higher than for the logistic models (0.62-0.70) despite the latter models including more predictors. A total of 34.6-38.1% of respondents with subsequent high persistence chronicity and 40.8-55.8% with the severity indicators were in the top 20% of the baseline ML-predicted risk distribution, while only 0.9% of respondents with subsequent hospitalizations and 1.5% with suicide attempts were in the lowest 20% of the ML-predicted risk distribution. These results confirm that clinically useful MDD risk-stratification models can be generated from baseline patient self-reports and that ML methods improve on conventional methods in developing such models.

Beyond Lumping and Splitting: A Review of Computational Approaches for Stratifying Psychiatric Disorders

Heterogeneity is a key feature of all psychiatric disorders that manifests on many levels, including symptoms, disease course, and biological underpinnings. These form a substantial barrier to understanding disease mechanisms and developing effective, personalized treatments. In response, many studies have aimed to stratify psychiatric disorders, aiming to find more consistent subgroups on the basis of many types of data. Such approaches have received renewed interest after recent research initiatives, such as the National Institute of Mental Health Research Domain Criteria and the European Roadmap for Mental Health Research, both of which emphasize finding stratifications that are based on biological systems and that cut across current classifications. We first introduce the basic concepts for stratifying psychiatric disorders and then provide a methodologically oriented and critical review of the existing literature. This shows that the predominant clustering approach that aims to subdivide clinical populations into more coherent subgroups has made a useful contribution but is heavily dependent on the type of data used; it has produced many different ways to subgroup the disorders we review, but for most disorders it has not converged on a consistent set of subgroups. We highlight problems with current approaches that are not widely recognized and discuss the importance of validation to ensure that the derived subgroups index clinically relevant variation. Finally, we review emerging techniques-such as those that estimate normative models for mappings between biology and behavior-that provide new ways to parse the heterogeneity underlying psychiatric disorders and evaluate all methods to meeting the objectives of such as the National Institute of Mental Health Research Domain Criteria and Roadmap for Mental Health Research.

PSYCHIATRIC COMORBIDITY DOES NOT ONLY DEPEND ON DIAGNOSTIC THRESHOLDS: AN ILLUSTRATION WITH MAJOR DEPRESSIVE DISORDER AND GENERALIZED ANXIETY DISORDER

BACKGROUND

High rates of psychiatric comorbidity are subject of debate: To what extent do they depend on classification choices such as diagnostic thresholds? This paper investigates the influence of different thresholds on rates of comorbidity between major depressive disorder (MDD) and generalized anxiety disorder (GAD).

METHODS

Point prevalence of comorbidity between MDD and GAD was measured in 74,092 subjects from the general population (LifeLines) according to Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) criteria. Comorbidity rates were compared for different thresholds by varying the number of necessary criteria from ≥ 1 to all nine symptoms for MDD, and from ≥ 1 to all seven symptoms for GAD.

RESULTS

According to DSM thresholds, 0.86% had MDD only, 2.96% GAD only, and 1.14% both MDD and GAD (odds ratio (OR) 42.6). Lower thresholds for MDD led to higher rates of comorbidity (1.44% for ≥ 4 of nine MDD symptoms, OR 34.4), whereas lower thresholds for GAD hardly influenced comorbidity (1.16% for ≥ 3 of seven GAD symptoms, OR 38.8). Specific patterns in the distribution of symptoms within the population explained this finding: 37.3% of subjects with core criteria of MDD and GAD reported subthreshold MDD symptoms, whereas only 7.6% reported subthreshold GAD symptoms.

CONCLUSIONS

Lower thresholds for MDD increased comorbidity with GAD, but not vice versa, owing to specific symptom patterns in the population. Generally, comorbidity rates result from both empirical symptom distributions and classification choices and cannot be reduced to either of these exclusively. This insight invites further research into the formation of disease concepts that allow for reliable predictions and targeted therapeutic interventions.

Multiple risk factors predict recurrence of major depressive disorder in women

BACKGROUND

It is difficult to predict recurrence of depressive episodes in patients with major depression (MD): evidence for many risk factors is inconsistent and general prediction algorithms are lacking. The aim of this study was to develop a prediction model for recurrence of depressive episodes in women using improved methodology.

METHODS

We used prospective data from a general population sample of female twins with a last-year MD episode (n=194). A rich set of baseline predictors was analyzed with Cox proportional hazards regression subject to elastic net regularization to find a model predicting recurrence of depressive episodes. Prediction accuracy of the model was assessed in an independent test sample (n=133), which was limited by the unavailability of a number of key predictors.

RESULTS

A wide variety of risk factors predicted recurrence of depressive episodes in women: depressive and anxiety symptoms during the index episode, the level of symptoms at the moment of interview, psychiatric and family history, early and recent adverse life events, being unmarried, and problems with friends and finances. Kaplan Meier estimated survival curves showed that the model differentiated between patients at higher and lower risk for recurrence; estimated areas under the curve were in the range of 0.61-0.79.

LIMITATIONS

Despite our rich set of predictors, certain potentially relevant variables were not available, such as biological measures, chronic somatic diseases, and treatment status.

CONCLUSIONS

Recurrence of episodes of MD in women is highly multifactorial. Future studies should take this into account for the development of clinically useful prediction algorithms.

Constructs of depression and distress in diabetes: time for an appraisal

Depression presents in roughly 20% of people with diabetes worldwide, and adversely affects quality of life and treatment outcomes. The causes of depression in diabetes are poorly understood, but research suggests a bi-directional association, at least for type 2 diabetes. Inconsistent findings regarding prevalence and depression treatment outcomes in patients with diabetes seem partly attributable to inconsistencies in the definition and measurement of depression and in distinguishing it from diabetes-distress, a psychological concept related to depression. We review evidence suggesting that diabetes-distress and depression are correlated and overlapping constructs, but are not interchangeable. Importantly, diabetes-distress seems to mediate the association between depression and glycaemic control. We propose a model to explain the direct and indirect effects of depression and diabetes-distress on glycaemic control. Additionally, using emerging insights from data-driven approaches, we suggest three distinct symptom profiles to define depression in patients with diabetes that could help explain differential associations between depression and metabolic abnormalities, and to tailor interventions for depression. Future research should focus on further refining depression profiles in patients with diabetes, taking into account the natural history of diabetes and depression, clinical characteristics, and diabetes-distress. The assessment of diabetes-distress and depression in research and clinical practice will be essential to identify high-risk patients with different mental health needs.

Depression sum-scores don't add up: why analyzing specific depression symptoms is essential

Most measures of depression severity are based on the number of reported symptoms, and threshold scores are often used to classify individuals as healthy or depressed. This method--and research results based on it--are valid if depression is a single condition, and all symptoms are equally good severity indicators. Here, we review a host of studies documenting that specific depressive symptoms like sad mood, insomnia, concentration problems, and suicidal ideation are distinct phenomena that differ from each other in important dimensions such as underlying biology, impact on impairment, and risk factors. Furthermore, specific life events predict increases in particular depression symptoms, and there is evidence for direct causal links among symptoms. We suggest that the pervasive use of sum-scores to estimate depression severity has obfuscated crucial insights and contributed to the lack of progress in key research areas such as identifying biomarkers and more efficacious antidepressants. The analysis of individual symptoms and their causal associations offers a way forward. We offer specific suggestions with practical implications for future research.

A biological mechanism for Bayesian feature selection: Weight decay and raising the LASSO

Biological systems are capable of learning that certain stimuli are valuable while ignoring the many that are not, and thus perform feature selection. In machine learning, one effective feature selection approach is the least absolute shrinkage and selection operator (LASSO) form of regularization, which is equivalent to assuming a Laplacian prior distribution on the parameters. We review how such Bayesian priors can be implemented in gradient descent as a form of weight decay, which is a biologically plausible mechanism for Bayesian feature selection. In particular, we describe a new prior that offsets or "raises" the Laplacian prior distribution. We evaluate this alongside the Gaussian and Cauchy priors in gradient descent using a generic regression task where there are few relevant and many irrelevant features. We find that raising the Laplacian leads to less prediction error because it is a better model of the underlying distribution. We also consider two biologically relevant online learning tasks, one synthetic and one modeled after the perceptual expertise task of Krigolson et al. (2009). Here, raising the Laplacian prior avoids the fast erosion of relevant parameters over the period following training because it only allows small weights to decay. This better matches the limited loss of association seen between days in the human data of the perceptual expertise task. Raising the Laplacian prior thus results in a biologically plausible form of Bayesian feature selection that is effective in biologically relevant contexts.

Rodent models of treatment-resistant depression

Major depression is a prevalent and debilitating disorder and a substantial proportion of patients fail to reach remission following standard antidepressant pharmacological treatment. Limited efficacy with currently available antidepressant drugs highlights the need to develop more effective medications for treatment- resistant patients and emphasizes the importance of developing better preclinical models that focus on treatment- resistant populations. This review discusses methods to adapt and refine rodent behavioral models that are predictive of antidepressant efficacy to identify populations that show reduced responsiveness or are resistant to traditional antidepressants. Methods include separating antidepressant responders from non-responders, administering treatments that render animals resistant to traditional pharmacological treatments, and identifying genetic models that show antidepressant resistance. This review also examines pharmacological and non-pharmacological treatments regimes that have been effective in refractory patients and how some of these approaches have been used to validate animal models of treatment-resistant depression. The goals in developing rodent models of treatment-resistant depression are to understand the neurobiological mechanisms involved in antidepressant resistance and to develop valid models to test novel therapies that would be effective in patients that do not respond to traditional monoaminergic antidepressants.

The effects of co-morbidity in defining major depression subtypes associated with long-term course and severity

BACKGROUND

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.

METHOD

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.

RESULTS

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.

CONCLUSIONS

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.