Depression-Anxiety Coupling Strength as a predictor of relapse in major depressive disorder: A CAN-BIND wellness monitoring study report.

BACKGROUND: A critical challenge in the study and management of major depressive disorder (MDD) is predicting relapse. We examined the temporal correlation/coupling between depression and anxiety (called Depression-Anxiety Coupling Strength, DACS) as a predictor of relapse in patients with MDD. METHODS: We followed 97 patients with remitted MDD for an average of 394 days. Patients completed weekly self-ratings of depression and anxiety symptoms using the Quick Inventory of Depressive Symptoms (QIDS-SR) and the Generalized Anxiety Disorder 7-item scale (GAD-7). Using these longitudinal ratings we computed DACS as random slopes in a linear mixed effects model reflecting individual-specific degree of correlation between depression and anxiety across time points. We then tested DACS as an independent variable in a Cox proportional hazards model to predict relapse. RESULTS: A total of 28 patients (29 %) relapsed during the follow-up period. DACS significantly predicted confirmed relapse (hazard ratio [HR] 1.5, 95 % CI [1.01, 2.22], p = 0.043; Concordance 0.79 [SE 0.04]). This effect was independent of baseline depressive or anxiety symptoms or their average levels over the follow-up period, and was identifiable more than one month before relapse onset. LIMITATIONS: Small sample size, in a single study. Narrow phenotype and comorbidity profiles. CONCLUSIONS: DACS may offer opportunities for developing novel strategies for personalized monitoring, early detection, and intervention. Future studies should replicate our findings in larger, diverse patient populations, develop individual patient prediction models, and explore the underlying mechanisms that govern the relationship of DACS and relapse.

Predicting which patients with cancer will see a psychiatrist or counsellor from their initial oncology consultation document using natural language processing.

BACKGROUND: Patients with cancer often have unmet psychosocial needs. Early detection of who requires referral to a counsellor or psychiatrist may improve their care. This work used natural language processing to predict which patients will see a counsellor or psychiatrist from a patient's initial oncology consultation document. We believe this is the first use of artificial intelligence to predict psychiatric outcomes from non-psychiatric medical documents. METHODS: This retrospective prognostic study used data from 47,625 patients at BC Cancer. We analyzed initial oncology consultation documents using traditional and neural language models to predict whether patients would see a counsellor or psychiatrist in the 12 months following their initial oncology consultation. RESULTS: Here, we show our best models achieved a balanced accuracy (receiver-operating-characteristic area-under-curve) of 73.1% (0.824) for predicting seeing a psychiatrist, and 71.0% (0.784) for seeing a counsellor. Different words and phrases are important for predicting each outcome. CONCLUSION: These results suggest natural language processing can be used to predict psychosocial needs of patients with cancer from their initial oncology consultation document. Future research could extend this work to predict the psychosocial needs of medical patients in other settings.

Patients with cancer often need support for their mental health. Early detection of who requires referral to a counsellor or psychiatrist may improve their care. This study trained a type of artificial intelligence (AI) called natural language processing to read the consultation report an oncologist writes after they first see a patient to predict which patients will see a counsellor or psychiatrist. The AI predicted this with performance similar to other uses of AI in mental health, and used different words and phrases to predict who would see a psychiatrist compared to seeing a counsellor. We believe this is the first use of AI to predict mental health outcomes from medical documents written by clinicians outside of mental health. This study suggests this type of AI can predict the mental health needs of patients with cancer from this widely-available document.

A Randomized Evaluation of MoodFX, a Patient-Centred e-Health Tool to Support Outcome Measurement for Depression: Une évaluation randomisée de MoodFX, un outil de santé en ligne centré sur le patient pour soutenir la mesure du résultat dans la dépression.

BACKGROUND: e-Health tools using validated questionnaires to assess outcomes may facilitate measurement-based care for psychiatric disorders. MoodFX was created as a free online symptom tracker to support patients for outcome measurement in their depression treatment. We conducted a pilot randomized evaluation to examine its usability, and clinical utility. METHODS: Patients presenting with a major depressive episode (within a major depressive or bipolar disorder) were randomly assigned to receive either MoodFX or a health information website as the intervention and control condition, respectively, with follow-up assessment surveys conducted online at baseline, 8 weeks and 6 months. The primary usability outcomes included the percentage of patients with self-reported use of MoodFX 3 or more times during follow up (indicating minimally adequate usage) and usability measures based on the System Usability Scale (SUS). Secondary clinical outcomes included the Quick Inventory of Depressive Symptomatology, Self-Rated (QIDS-SR) and Patient Health Questionnaire (PHQ-9). RESULTS: Forty-nine participants were randomized (24 to MoodFX and 25 to the control condition). Of the 23 participants randomized to MoodFX who completed the user survey, 18 (78%) used MoodFX 3 or more times over the 6 months of the study. The mean SUS score of 72.7 (65th-69th percentile) represents good usability. Compared to the control group, the MoodFX group had significantly better improvement on QIDS-SR and PHQ-9 scores, with large effect sizes and higher response rates at 6 months. There were no differences between conditions on other secondary outcomes such as functioning and quality of life. CONCLUSION: MoodFX demonstrated good usability and was associated with reduction in depressive symptoms. This pilot study supports the use of digital tools in depression treatment.

E-health tools may be useful for measuring and tracking symptoms and other outcomes during treatment for depression. This study is a randomized evaluation of MoodFX, a free web-based app that helps patients track their symptoms using validated questionnaires, and also offers depression information and self-management tips. A total of 49 participants with clinical depression were randomized to using MoodFX or a health information website, for 6 months. In a survey, the participants that used MoodFX found it easy and useful to use. In addition, the participants that used MoodFX had greater improvement in depressive symptoms after 6 months, compared to those who used the health information website. These results suggest that MoodFX may be a useful tool to monitor outcomes and support depression treatment.

Supportive care and healthcare service utilisation in older adults with a new cancer diagnosis: a population-based review.

OBJECTIVES: Older adults have unique needs and may benefit from additional supportive services through their cancer journey. It can be challenging for older adults to navigate the siloed systems within cancer centres and the community. We aimed to document the use of supportive care services in older adults with a new cancer diagnosis in a public healthcare system. METHODS: We used population-based databases in British Columbia to document referrals to supportive care services. Patients aged 70 years and above with a new diagnosis of solid tumour in the year 2015 were included. Supportive care services captured were social work, psychiatry, palliative care, nutrition and home care. Chart review was used to assess visits to the emergency room and extra calls to the cancer centre help line. RESULTS: 2014 patients were included with a median age of 77, 30% had advanced cancer. 459 (22.8%) of patients accessed one or more services through the cancer centre. The most common service used was patient and family counselling (13%). 309 (15.3%) of patients used community home care services. Patients aged 80 years and above were less likely to access supportive care resources (OR 0.57) compared with those 70-79 years. Patients with advanced cancer, those treated at smaller cancer centres, and patients with colorectal, gynaecological and lung cancer were more likely to have received a supportive care referral. CONCLUSIONS: Older adults, particularly those above 80 years, have low rates of supportive care service utilisation. Barriers to access must be explored, in addition to novel ways of holistic care delivery.

Machine Learning Prediction of Quality of Life Improvement During Antidepressant Treatment of Patients With Major Depressive Disorder: A STAR*D and CAN-BIND-1 Report.

Background: Quality of life (QoL) is an important patient-centric outcome to evaluate in treatment of major depressive disorder (MDD). This work sought to investigate the performance of several machine learning methods to predict a return to normative QoL in patients with MDD after antidepressant treatment.Methods: Several binary classification algorithms were trained on data from the first 2 weeks of the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study (n = 651, conducted from 2001 to 2006) to predict week 9 normative QoL (score ≥ 67, based on a community normative sample, on the Quality of Life Enjoyment and Satisfaction Questionnaire-Short Form [Q-LES-Q-SF]) after treatment with citalopram. Internal validation was performed using a STAR*D holdout dataset, and external validation was performed using the Canadian Biomarker Integration Network in Depression-1 (CAN-BIND-1) dataset (n = 175, study conducted from 2012 to 2017) after treatment with escitalopram. Feature importance was calculated using SHapley Additive exPlanations (SHAP).Results: Random Forest performed most consistently on internal and external validation, with balanced accuracy (area under the receiver operator curve) of 71% (0.81) on the STAR*D dataset and 69% (0.75) on the CAN-BIND-1 dataset. Random Forest Classifiers trained on Q-LES-Q-SF and Quick Inventory of Depressive Symptomatology-Self-Rated variables had similar performance on both internal and external validation. Important predictive variables came from psychological, physical, and socioeconomic domains.Conclusions: Machine learning can predict normative QoL after antidepressant treatment with similar performance to that of prior work predicting depressive symptom response and remission. These results suggest that QoL outcomes in MDD patients can be predicted with simple patient-rated measures and provide a foundation to further improve performance and demonstrate clinical utility.Trial Registration: ClinicalTrials.gov identifiers NCT00021528 and NCT01655706.

Evidence of factors influencing delays in the diagnosis and treatment of bipolar disorder in adolescents and young adults. Protocol for a systematic scoping review.

BACKGROUND: Bipolar Disorder (BD) is a complex psychiatric condition that typically manifests during late adolescence and early adulthood. Over the past two decades, international studies have reported that BD often goes unrecognized and untreated for several years, which can lead to negative clinical and functional outcomes. However, the components of delay in the diagnosis and treatment of BD and various factors influencing those components have not been systematically explored. OBJECTIVES: The scoping review described in this protocol aims to map the existing literature on potential factors that influence delays in the treatment of BD in adolescents and young adults, in order to identify the knowledge gaps and future research and policy priorities. METHODS: This protocol for a systematic scoping review will be reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guideline (PRISMA-ScR). We will search the electronic databases of MEDLINE (OVID), EMBASE, PsycINFO and CINAHL for peer-reviewed primary research articles published in academic journals. Grey literature will not be explored due to resource limitations. A conceptual framework based on the Model of Pathways to Treatment by Scott and colleagues was used as a foundation for our search and extraction strategy to ensure all components of delay and potential factors influencing each component are explored. Two independent reviewers will screen the references retrieved by the literature search and select relevant studies based on our inclusion criteria. The data from included studies will be synthesized into a narrative summary, and implications for future research, practice and policy will be discussed. DISCUSSION: To the best of our knowledge, this will be the first scoping review to explore the potential factors that influence delays in the treatment of BD in adolescents and young adults. We intend to disseminate the review results through academic conferences and publication in a peer-reviewed journal.

Response trajectories during escitalopram treatment of patients with major depressive disorder.

Depression is a leading global cause of disability, yet about half of patients do not respond to initial antidepressant treatment. This treatment difficulty may be in part due to the heterogeneity of depression and corresponding response to treatment. Unsupervised machine learning allows underlying patterns to be uncovered, and can be used to understand this heterogeneity by finding groups of patients with similar response trajectories. Prior studies attempting this have clustered patients using a narrow range of data primarily from depression scales. In this work, we used unsupervised machine learning to cluster patients receiving escitalopram therapy using a wide variety of subjective and objective clinical features from the first eight weeks of the Canadian Biomarker Integration Network in Depression-1 trial. We investigated how these clusters responded to treatment by comparing changes in symptoms and symptom categories, and by using Principal Component Analysis (PCA). Our algorithm found three clusters, which broadly represented non-responders, responders, and remitters. Most categories of features followed this response pattern except for objective cognitive features. Using PCA with our clusters, we found that subjective mood state/anhedonia is the core feature of response with escitalopram, but there exists other distinct patterns of response around neurovegetative symptoms, activation, and cognition.

Reduction in Substance-Related Composite Harm Scores Through Street Soccer.

Introduction Street soccer makes the sport accessible to people affected by homelessness or precarious housing. There is overwhelming evidence that exercise improves physical and mental health. In addition, sport facilitates positive peer pressure that leads to beneficial life changes. Method To examine participants' accounts of the effects of street soccer in a sample of socially disadvantaged players from Western Canada, we collected 73 cross-sectional self-reports of life changes via a questionnaire. The questionnaire included questions on social, mental, and physical health, including substance use. This allowed the calculation of a modified composite harm score. Results Participants reported improved physical (46% of participants) and mental (43% of participants) health, reduced cigarette (50% of smokers), alcohol (45% of users), cannabis (42% of users), and other non-prescribed drug use, increased number of friends (88% of participants), improved housing (60% of participants), increased income (19% of participants), increased community medical supports (40% of participants), and decreased conflicts with police (47% of those with prior recent conflict). Perceived reductions in substance use were supported by significant changes in composite harm score. Conclusion Street soccer appears to promote improved physical, mental, and social health among people affected by homelessness or precarious housing, with reduction in substance use likely to be a key factor. This work builds upon past qualitative research showing the benefits of street soccer and supports future research which may help elucidate the mechanisms by which street soccer has beneficial effects.

Predicting the Survival of Patients With Cancer From Their Initial Oncology Consultation Document Using Natural Language Processing.

Importance: Predicting short- and long-term survival of patients with cancer may improve their care. Prior predictive models either use data with limited availability or predict the outcome of only 1 type of cancer. Objective: To investigate whether natural language processing can predict survival of patients with general cancer from a patient's initial oncologist consultation document. Design, Setting, and Participants: This retrospective prognostic study used data from 47 625 of 59 800 patients who started cancer care at any of the 6 BC Cancer sites located in the province of British Columbia between April 1, 2011, and December 31, 2016. Mortality data were updated until April 6, 2022, and data were analyzed from update until September 30, 2022. All patients with a medical or radiation oncologist consultation document generated within 180 days of diagnosis were included; patients seen for multiple cancers were excluded. Exposures: Initial oncologist consultation documents were analyzed using traditional and neural language models. Main Outcomes and Measures: The primary outcome was the performance of the predictive models, including balanced accuracy and receiver operating characteristics area under the curve (AUC). The secondary outcome was investigating what words the models used. Results: Of the 47 625 patients in the sample, 25 428 (53.4%) were female and 22 197 (46.6%) were male, with a mean (SD) age of 64.9 (13.7) years. A total of 41 447 patients (87.0%) survived 6 months, 31 143 (65.4%) survived 36 months, and 27 880 (58.5%) survived 60 months, calculated from their initial oncologist consultation. The best models achieved a balanced accuracy of 0.856 (AUC, 0.928) for predicting 6-month survival, 0.842 (AUC, 0.918) for 36-month survival, and 0.837 (AUC, 0.918) for 60-month survival, on a holdout test set. Differences in what words were important for predicting 6- vs 60-month survival were found. Conclusions and Relevance: These findings suggest that models performed comparably with or better than previous models predicting cancer survival and that they may be able to predict survival using readily available data without focusing on 1 cancer type.

Replication of machine learning methods to predict treatment outcome with antidepressant medications in patients with major depressive disorder from STAR*D and CAN-BIND-1.

OBJECTIVES: Antidepressants are first-line treatments for major depressive disorder (MDD), but 40-60% of patients will not respond, hence, predicting response would be a major clinical advance. Machine learning algorithms hold promise to predict treatment outcomes based on clinical symptoms and episode features. We sought to independently replicate recent machine learning methodology predicting antidepressant outcomes using the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) dataset, and then externally validate these methods to train models using data from the Canadian Biomarker Integration Network in Depression (CAN-BIND-1) dataset. METHODS: We replicated methodology from Nie et al (2018) using common algorithms based on linear regressions and decision trees to predict treatment-resistant depression (TRD, defined as failing to respond to 2 or more antidepressants) in the STAR*D dataset. We then trained and externally validated models using the clinical features found in both datasets to predict response (≥50% reduction on the Quick Inventory for Depressive Symptomatology, Self-Rated [QIDS-SR]) and remission (endpoint QIDS-SR score ≤5) in the CAN-BIND-1 dataset. We evaluated additional models to investigate how different outcomes and features may affect prediction performance. RESULTS: Our replicated models predicted TRD in the STAR*D dataset with slightly better balanced accuracy than Nie et al (70%-73% versus 64%-71%, respectively). Prediction performance on our external methodology validation on the CAN-BIND-1 dataset varied depending on outcome; performance was worse for response (best balanced accuracy 65%) compared to remission (77%). Using the smaller set of features found in both datasets generally improved prediction performance when evaluated on the STAR*D dataset. CONCLUSION: We successfully replicated prior work predicting antidepressant treatment outcomes using machine learning methods and clinical data. We found similar prediction performance using these methods on an external database, although prediction of remission was better than prediction of response. Future work is needed to improve prediction performance to be clinically useful.

A Conserved Residue Cluster That Governs Kinetics of ATP-dependent Gating of Kir6.2 Potassium Channels.

ATP-sensitive potassium (KATP) channels are heteromultimeric complexes of an inwardly rectifying Kir channel (Kir6.x) and sulfonylurea receptors. Their regulation by intracellular ATP and ADP generates electrical signals in response to changes in cellular metabolism. We investigated channel elements that control the kinetics of ATP-dependent regulation of KATP (Kir6.2 + SUR1) channels using rapid concentration jumps. WT Kir6.2 channels re-open after rapid washout of ATP with a time constant of ∼60 ms. Extending similar kinetic measurements to numerous mutants revealed fairly modest effects on gating kinetics despite significant changes in ATP sensitivity and open probability. However, we identified a pair of highly conserved neighboring amino acids (Trp-68 and Lys-170) that control the rate of channel opening and inhibition in response to ATP. Paradoxically, mutations of Trp-68 or Lys-170 markedly slow the kinetics of channel opening (500 and 700 ms for W68L and K170N, respectively), while increasing channel open probability. Examining the functional effects of these residues using φ value analysis revealed a steep negative slope. This finding implies that these residues play a role in lowering the transition state energy barrier between open and closed channel states. Using unnatural amino acid incorporation, we demonstrate the requirement for a planar amino acid at Kir6.2 position 68 for normal channel gating, which is potentially necessary to localize the ϵ-amine of Lys-170 in the phosphatidylinositol 4,5-bisphosphate-binding site. Overall, our findings identify a discrete pair of highly conserved residues with an essential role for controlling gating kinetics of Kir channels.

Paradoxical activation of an inwardly rectifying potassium channel mutant by spermine: "(b)locking" open the bundle crossing gate.

Intracellular polyamines are endogenous blockers of inwardly rectifying potassium (Kir) channels and underlie steeply voltage-dependent rectification. Kir channels with strong polyamine sensitivity typically carry a negatively charged side chain at a conserved inner cavity position, although acidic residues at any pore-lining position in the inner cavity are sufficient to confer polyamine block. We have identified unique consequences of a glutamate substitution in the region of the helix bundle crossing of Kir6.2. Firstly, glutamate substitution at Kir6.2 residue F168 generates channels with intrinsic inward rectification that does not require blockade by intracellular polyamines or Mg(2+). In addition, these F168E channels exhibit a unique "spiked" tail phenotype, whereby large decaying inward tail currents are elicited upon spermine unbinding. This contrasts with the time-dependent recovery of current typically associated with blocker unbinding from ion channels. Interestingly, Kir6.2[F168E] channels exhibit a paradoxical biphasic conductance-voltage relationship in the presence of certain polyamines. This reflects channel blockade at positive voltages, channel stimulation at intermediate voltages, and exclusion of spermine from the pore at negative voltages. These features are recapitulated by a simple kinetic scheme in which weakly voltage-dependent spermine binding to a "shallow" site in the pore (presumably formed by the introduced glutamate at F168E) stabilizes opening of the bundle crossing gate. These findings illustrate the potential for dichotomous effects of a blocker in a long pore (with multiple binding sites), and offer a unique example of targeted modulation of the Kir channel gating apparatus.

A novel mechanism for fine-tuning open-state stability in a voltage-gated potassium channel.

Voltage-gated potassium channels elicit membrane hyperpolarization through voltage-sensor domains that regulate the conductive status of the pore domain. To better understand the inherent basis for the open-closed equilibrium in these channels, we undertook an atomistic scan using synthetic fluorinated derivatives of aromatic residues previously implicated in the gating of Shaker potassium channels. Here we show that stepwise dispersion of the negative electrostatic surface potential of only one site, Phe481, stabilizes the channel open state. Furthermore, these data suggest that this apparent stabilization is the consequence of the amelioration of an inherently repulsive open-state interaction between the partial negative charge on the face of Phe481 and a highly co-evolved acidic side chain, Glu395, and this interaction is potentially modulated through the Tyr485 hydroxyl. We propose that the intrinsic open-state destabilization via aromatic repulsion represents a new mechanism by which ion channels, and likely other proteins, fine-tune conformational equilibria.