Prediction of Acute-Phase Treatment Outcomes by Adding a Single-Item Measure of Activity Impairment to Symptom Measurement: Development and Validation of an Interactive Calculator from the STAR*D and CO-MED Trials

Trajectories of depression symptoms, anxiety symptoms and functional impairment during internet-enabled cognitive-behavioural therapy

Underlying classes capture differences between patient symptom trajectories during psychological therapy. This has not been explored for one-to-one internet-delivered therapy or functional impairment trajectories. Patients experiencing depression or anxiety received cognitive-behavioural therapy with a therapist using an online chat platform (N = 52,029). Trajectory classes of depression symptoms (PHQ9), anxiety symptoms (GAD7) and functional impairment (WSAS) were investigated using growth mixture modelling. Multinomial regressions tested associations between baseline variables and trajectory class. A four-class trajectory model was selected for each outcome, and these were highly similar. Each outcome showed three classes with initially moderate-severe symptoms or impairment: one demonstrated no change, one gradual improvement and one fast improvement. A fourth class had mild baseline scores and minimal improvement. In the moderate-severe classes, patients in the two with improvement were more likely to be employed and not to have obsessive-compulsive disorder. Fast improvement was likelier than gradual improvement or no change for patients with older age, no disability (e.g., physical, learning), or lower comorbid symptom or impairment scores. Associations with functional impairment classes were more similar to associations with depression classes than anxiety classes. Results were largely consistent with findings from face-to-face therapy. This study is an important step towards personalising therapy in terms of suitability and continuation.

Cross-Sectional Associations Among Symptoms of Pain, Irritability, and Depression and How These Symptoms Relate to Social Functioning and Quality of Life: Findings From the EMBARC and STRIDE Studies and the VitalSign6 Project

OBJECTIVE

The aim of this report was to evaluate the psychometric properties of the Pain Frequency, Intensity, and Burden Scale (P-FIBS), a brief measure of pain, as well as the association of pain with irritability and depression and how these symptoms relate to functional impairments.

METHODS

Participants of 2 randomized controlled trials (Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care [EMBARC; n = 251 with DSM-IV diagnosis of major depressive disorder; study duration: August 2011-December 2015] and STimulant Reduction Intervention Using Dosed Exercise [STRIDE; n = 302 with DSM-IV diagnosis of stimulant abuse or dependence; study-duration: July 2010-February 2013]) and treatment-seeking patients in primary care clinics from an ongoing quality-improvement project (VitalSign6; n = 4,370; project duration: August 2014-July 2019) were included. Psychometric properties of the P-FIBS were evaluated with confirmatory factor and item response theory analyses in EMBARC and VitalSign6. The approach of Baron and Kenny was used to assess whether irritability accounted for the effect of pain on depression.

RESULTS

Cronbach α (0.84-0.89) and model fits for single-factor structure of P-FIBS were acceptable. Pain was positively correlated with irritability (r = 0.22-0.29) and depression (r = 0.10-0.33). Irritability accounted for 40.7%-65.5% of the effect of pain on depression. Higher irritability and depression were associated with poorer social functioning, quality of life, and productivity in work- and non-work-related activities. Pain was associated with non-work-related activity impairments even after controlling for irritability and depression.

CONCLUSIONS

The P-FIBS is a brief and reliable measure of pain. Irritability is associated with pain and accounts for a large proportion of the effect of pain on depression. Symptoms of pain, irritability, and depression are associated with functional impairments.

TRIAL REGISTRATION

ClinicalTrials.gov identifiers: NCT01407094 (EMBARC), NCT01141608 (STRIDE).

Assessment of a Prediction Model for Antidepressant Treatment Stability Using Supervised Topic Models

Importance

In the absence of readily assessed and clinically validated predictors of treatment response, pharmacologic management of major depressive disorder often relies on trial and error.

Objective

To assess a model using electronic health records to identify predictors of treatment response in patients with major depressive disorder.

Design, Setting, and Participants

This retrospective cohort study included data from 81 630 adults with a coded diagnosis of major depressive disorder from 2 academic medical centers in Boston, Massachusetts, including outpatient primary and specialty care clinics from December 1, 1997, to December 31, 2017. Data were analyzed from January 1, 2018, to March 15, 2020.

Exposures

Treatment with at least 1 of 11 standard antidepressants.

Main Outcomes and Measures

Stable treatment response, intended as a proxy for treatment effectiveness, defined as continued prescription of an antidepressant for 90 days. Supervised topic models were used to extract 10 interpretable covariates from coded clinical data for stability prediction. With use of data from 1 hospital system (site A), generalized linear models and ensembles of decision trees were trained to predict stability outcomes from topic features that summarize patient history. Held-out patients from site A and individuals from a second hospital system (site B) were evaluated.

Results

Among the 81 630 adults (56 340 women [69%]; mean [SD] age, 48.46 [14.75] years; range, 18.0-80.0 years), 55 303 reached a stable response to their treatment regimen during follow-up. For held-out patients from site A, the mean area under the receiver operating characteristic curve (AUC) for discrimination of the general stability outcome was 0.627 (95% CI, 0.615-0.639) for the supervised topic model with 10 covariates. In evaluation of site B, the AUC was 0.619 (95% CI, 0.610-0.627). Building models to predict stability specific to a particular drug did not improve prediction of general stability even when using a harder-to-interpret ensemble classifier and 9256 coded covariates (specific AUC, 0.647; 95% CI, 0.635-0.658; general AUC, 0.661; 95% CI, 0.648-0.672). Topics coherently captured clinical concepts associated with treatment response.

Conclusions and Relevance

The findings suggest that coded clinical data available in electronic health records may facilitate prediction of general treatment response but not response to specific medications. Although greater discrimination is likely required for clinical application, the results provide a transparent baseline for such studies.