Prediction of short-term antidepressant response using probabilistic graphical models with replication across multiple drugs and treatment settings

AbstractHeterogeneity in the clinical presentation of major depressive disorder and response to antidepressants limits clinicians’ ability to accurately predict a specific patient’s eventual response to therapy. Validated depressive symptom profiles may be an important tool for identifying poor outcomes early in the course of treatment. To derive these symptom profiles, we first examined data from 947 depressed subjects treated with selective serotonin reuptake inhibitors (SSRIs) to delineate the heterogeneity of antidepressant response using probabilistic graphical models (PGMs). We then used unsupervised machine learning to identify specific depressive symptoms and thresholds of improvement that were predictive of antidepressant response by 4 weeks for a patient to achieve remission, response, or nonresponse by 8 weeks. Four depressive symptoms (depressed mood, guilt feelings and delusion, work and activities and psychic anxiety) and specific thresholds of change in each at 4 weeks predicted eventual outcome at 8 weeks to SSRI therapy with an average accuracy of 77% (p = 5.5E-08). The same four symptoms and prognostic thresholds derived from patients treated with SSRIs correctly predicted outcomes in 72% (p = 1.25E-05) of 1996 patients treated with other antidepressants in both inpatient and outpatient settings in independent publicly-available datasets. These predictive accuracies were higher than the accuracy of 53% for predicting SSRI response achieved using approaches that (i) incorporated only baseline clinical and sociodemographic factors, or (ii) used 4-week nonresponse status to predict likely outcomes at 8 weeks. The present findings suggest that PGMs providing interpretable predictions have the potential to enhance clinical treatment of depression and reduce the time burden associated with trials of ineffective antidepressants. Prospective trials examining this approach are forthcoming.

Download Full-text

Prevalence, detection and associations of depression in Australian dialysis patients

Australasian Psychiatry ◽

10.1177/1039856219859281 ◽

2019 ◽

Vol 27 (5) ◽

pp. 444-449 ◽

Cited By ~ 1

Author(s):

Elaine Kwan ◽

Brian Draper ◽

Samuel B Harvey ◽

Zoltan H Endre ◽

Mark A Brown

Keyword(s):

Depressive Symptoms ◽

Depression Scale ◽

Linguistically Diverse ◽

Sociodemographic Factors ◽

Cross Sectional Study ◽

Dialysis Patients ◽

Cross Sectional ◽

History Of ◽

Poor Outcomes ◽

History Of Depression

Objective:Depression in dialysis patients is often undetected despite associations with poor outcomes. The aim was to determine the prevalence and associations of depressive symptoms and physician recognition of depression within a typical Australian dialysis population.Method:A cross-sectional study examined haemodialysis and peritoneal dialysis patients of two hospitals in Sydney. Participants were screened for depression using the Hospital Anxiety and Depression Scale (HADS).Results:One hundred and ten patients completed the HADS. Subjects had a mean age of 63.7 years, 37% from a culturally and linguistically diverse background, and median dialysis duration of 2 years. Forty-one per cent of participants had significant depressive symptoms, of whom 42% had been diagnosed with depression by their clinicians. After adjustment for sociodemographic factors, having >10 medications prescribed, >3 hospitalisations in the last 12 months, and a history of depression were associated with depression.Conclusion:Depressive symptoms are prevalent in Australian dialysis patients. Robust evidence is needed on the effectiveness of treatments for depression in changing outcomes in chronic kidney disease.

Download Full-text

Exploring Probabilistic Graphical Models in a GNSS Software Receiver

Proceedings of the 30th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2017) ◽

10.33012/2017.15168 ◽

2017 ◽

Author(s):

Xin Zhang ◽

Xingqun Zhan

Keyword(s):

Graphical Models ◽

Probabilistic Graphical Models ◽

Software Receiver

Download Full-text

Bereavement or Breakup: A Network Approach to Modeling Differential Pathways into Depression

10.31234/osf.io/38r4j ◽

2019 ◽

Author(s):

Julian Burger ◽

Margaret S. Stroebe ◽

Pasqualina Perrig-Chiello ◽

Henk A.W. Schut ◽

Stefanie Spahni ◽

...

Keyword(s):

Depressive Symptoms ◽

Graphical Models ◽

Contextual Factors ◽

Marital Disruption ◽

Depression Symptoms ◽

Depression Symptom ◽

Multiple Time ◽

Cross Sectional ◽

Network Analyses ◽

Within Subjects

Background: Prior network analyses demonstrated that the death of a loved one potentially precedes specific depression symptoms, primarily loneliness, which in turn links to other depressive symptoms. In this study, we extend prior research by comparing depression symptom network structures following two types of marital disruption: bereavement versus separation. Methods: We fitted two Gaussian Graphical Models to cross-sectional data from a Swiss survey of older persons (145 bereaved, 217 separated, and 362 married controls), and compared symptom levels across bereaved and separated individuals. Results: Separated compared to widowed individuals were more likely to perceive an unfriendly environment and oneself as a failure. Both types of marital disruption were linked primarily to loneliness, from where different relations emerged to other depressive symptoms. Amongst others, loneliness had a stronger connection to perceiving oneself as a failure in separated compared to widowed individuals. Conversely, loneliness had a stronger connection to getting going in widowed individuals. Limitations: Analyses are based on cross-sectional between-subjects data, and conclusions regarding dynamic processes on the within-subjects level remain putative. Further, some of the estimated parameters in the network exhibited overlapping confidence intervals and their order needs to be interpreted with care. Replications should thus aim for studies with multiple time points and larger samples. Conclusions: The findings of this study add to a growing body of literature indicating that depressive symptom patterns depend on contextual factors. If replicated on the within-subjects level, such findings have implications for setting up patient-tailored treatment approaches in dependence of contextual factors.

Download Full-text

Sociodemographic inequalities in lifetime treatment seeking and receipt for common mental health problems

10.31234/osf.io/vsu4g ◽

2020 ◽

Author(s):

Christopher Rayner ◽

Jonathan Richard Iain Coleman ◽

Kirstin Lee Purves ◽

Ewan Carr ◽

Rosa Cheesman ◽

...

Keyword(s):

Depressive Disorders ◽

Mental Health Problems ◽

Sociodemographic Factors ◽

Treatment Seeking ◽

Binary Outcomes ◽

Treatment Access ◽

Sociodemographic Correlates ◽

Common Mental Health Problems ◽

Poor Outcomes ◽

Lifetime Treatment

Background: Anxiety and depressive disorders can be chronic and disabling, and are associated with poor outcomes. Whilst there are effective treatments, access to these is variable and only a fraction of those in need receive treatment. Aims: The primary aim was to investigate sociodemographic correlates of lifetime treatment access and unpick the relationships between socioeconomic features and treatment inequalities. As such, we aimed to identify groups at greatest risk of never accessing treatment and targets for intervention. Methods: We tested for sociodemographic factors associated with treatment access in UK Biobank participants with lifetime generalised anxiety or major depressive disorder, performing multivariable logistic regressions on two binary outcomes: treatment-seeking (n=33,704) and treatment receipt (n=28,940). Results: Treatment access was less likely in those who were male, from a UK ethnic minority background and who self-medicated with alcohol or drugs. Treatment access was more likely in those who reported use of self-help strategies, with lower income (<£30,000) and greater neighbourhood deprivation, as well as those with a university degree. Conclusion: This work on lifetime treatment seeking and receipt replicates known correlates of treatment receipt during time of treatment need. Our focus on treatment-seeking and receipt highlights two targets for improving treatment access. More work is required to understand the psychosocial barriers to treatment, which mediate the associations observed here.

Download Full-text

Trajectories of Depressive Symptoms and Anxiety during Pregnancy and Associations with Pregnancy Stress

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph18052733 ◽

2021 ◽

Vol 18 (5) ◽

pp. 2733 ◽

Cited By ~ 1

Author(s):

Hyejung Lee ◽

Ki-Eun Kim ◽

Mi-Young Kim ◽

Chang Gi Park ◽

Jung Yeol Han ◽

...

Keyword(s):

Depressive Symptoms ◽

Pregnant Women ◽

Healthcare Providers ◽

Sociodemographic Factors ◽

Survey Questionnaire ◽

Emotional Status ◽

Factors Associated ◽

Latent Trajectory ◽

Logistic Regressions ◽

Health Hospital

The purposes of this study were to investigate the trajectory groups of depressive symptoms and anxiety in women during pregnancy and to identify the factors associated with those groups. Participants were recruited from the outpatient clinic of a women’s health hospital in Seoul, Korea. Pregnant women (n = 136) completed a survey questionnaire that included questions on depressive symptoms, anxiety, and pregnancy stress; additionally, their saliva was tested for cortisol hormone levels three times during their pregnancies. The group-based trajectory modeling approach was used to identify latent trajectory groups. Ordinal logistic regressions were used to explore the association of latent trajectory groups with sociodemographic factors and pregnancy stress. Three trajectory groups of depressive symptoms were identified: low-stable (70%), moderate-stable (25%), and increased (5%). Four trajectory groups of anxiety were identified: very low-stable (10%), low-stable (67%), moderate-stable (18%), and high-stable (5%). The only factor associated with both the depressive symptoms and anxiety trajectory groups was pregnancy stress (p < 0.001). Most participants showed stable emotional status; however, some participants experienced higher levels of depressive symptoms and anxiety related to higher pregnancy stress. These pregnant women may need additional care from healthcare providers to promote their wellbeing during pregnancy.

Download Full-text

FRI0224 IDENTIFICATION OF RISK AND PROGNOSTIC FACTORS FOR POLYARTERITIS NODOSA PATIENTS WITH DIGITAL GANGRENE

Annals of the Rheumatic Diseases ◽

10.1136/annrheumdis-2020-eular.817 ◽

2020 ◽

Vol 79 (Suppl 1) ◽

pp. 695.2-695

Author(s):

D. Xu ◽

X. Tian ◽

X. Zeng ◽

F. Zhang ◽

L. Zhao ◽

...

Keyword(s):

Systemic Lupus Erythematosus ◽

Prognostic Factors ◽

Retrospective Study ◽

Polyarteritis Nodosa ◽

Lupus Erythematosus ◽

Disease Duration ◽

Response To Therapy ◽

Smoking History ◽

Systemic Lupus ◽

Poor Outcomes

Background:Polyarteritis nodosa (PAN) is a segmental, necrotizing vascular disease that primarily impacts medium-sized muscle arteries. The estimated annual incidence of PAN is still lacking in China. Digital gangrene is an ischemic manifestation of the limb. However, the causes and the treatment methods vary from case to case, and the outcome is unpredictable. These features emphasize the need to identify measurable variables that accelerate digital gangrene development in PAN patients. However, little effort has been made to identify the clinical and laboratory factors that affect PAN patients with digital gangrene to anticipate their natural history and response to therapy.Objectives:Many patients with polyarteritis nodosa (PAN) complicated with digital gangrene have poor outcomes and related research information is limited. This study was carried out to identify the associated risk and prognostic factors.Methods:We conducted a retrospective study of 148 PAN patients admitted to Peking Union Medical College Hospital (PUMCH) from September 1986 to December 2018. The characteristics, therapeutic regimens, and outcome data for patients with and without gangrene were compared. The Kaplan–Meier method and Cox hazard regression model were used to evaluate the prognostic factors.Results:Forty-seven (31.8%) PAN patients had digital gangrene complications. The average age was 40.4±17.9 years and the average disease duration was 11 (4-27) months. The presence of digital gangrene was correlated with smoking history [odds ratio (OR), 4.27; 95% confidence interval (95% CI), 1.56-11.66] and eosinophil elevation (28.12; 10.30-76.8). Thirty-two (68.1%) gangrene patients received methylprednisolone pulse therapy and all of these patients were treated with cyclophosphamide. Nine patients suffered irreversible organ injury and two died. Disease duration ≥ 24 months and elevated serum C-reactive protein (CRP) were identified as hazardous factors for poor prognosis in patients with gangrene (P=0.003, HR=8.668, 95% CI 2.11, 35.55 andP=0.042, HR=27.062, 95% CI 1.13, 648.57, respectively).Conclusion:Smoking history and eosinophil elevation in PAN patients were more prone to digital gangrene and high serum CRP level predicted poor outcomes. PAN patients with smoking history and elevated eosinophils need to be seriously evaluated by clinicians. Furthermore, the CRP level should be efficiently controlled for good prognosis.References:[1]De Virgilio A, Greco A, Magliulo G, Gallo A, Ruoppolo G, Conte M, et al. Polyarteritis nodosa: A contemporary overview. Autoimmun Rev. 2016;15:564-70.[2]Pagnoux C, Seror R, Henegar C, Mahr A, Cohen P, Le Guern V, et al. Clinical features and outcomes in 348 patients with polyarteritis nodosa: a systematic retrospective study of patients diagnosed between 1963 and 2005 and entered into the French Vasculitis Study Group Database. Arthritis Rheum. 2010;62:616-26.[3]Xu D, You X, Wang Z, Zeng Q, Xu J, Jiang L, et al. Chinese Systemic Lupus Erythematosus Treatment and Research Group Registry VI: Effect of Cigarette Smoking on the Clinical Phenotype of Chinese Patients with Systemic Lupus Erythematosus. PLoS One. 2015;10:e0134451.Acknowledgments:NoDisclosure of Interests:Dong Xu: None declared, Xinping Tian: None declared, Xiaofeng Zeng Consultant of: MSD Pharmaceuticals, Fengchun Zhang: None declared, Lin Zhao: None declared, Shangzhu Zhang: None declared, Jiaxin Zhou: None declared, Jiu-liang Zhao: None declared, Xiaodan Kong: None declared

Download Full-text

Sociodemographic factors of families related to postnatal depressive symptoms of mothers

International Journal of Nursing Practice ◽

10.1046/j.1440-172x.2002.00373.x ◽

2002 ◽

Vol 8 (5) ◽

pp. 240-246 ◽

Cited By ~ 36

Author(s):

Tarja Tammentie ◽

Marja-Terttu Tarkka ◽

Paivi Astedt-Kurki ◽

Eija Paavilainen

Keyword(s):

Depressive Symptoms ◽

Sociodemographic Factors ◽

Postnatal Depressive Symptoms

Download Full-text

Pharmacogenomics of Antihypertensive Drugs in Brazil: recent progress and clinical implications

Endocrine Metabolic & Immune Disorders - Drug Targets ◽

10.2174/1871530322666211222122212 ◽

2021 ◽

Vol 22 ◽

Author(s):

Fabiana Dalla Vecchia Genvigir ◽

Carolina Dagli Hernandez ◽

Thiago Dominguez Crespo Hirata ◽

Yitian Zhou ◽

Volker M Lauschke ◽

...

Keyword(s):

Genetic Factors ◽

Antihypertensive Drugs ◽

Gestational Hypertension ◽

Pressure Control ◽

Response To Therapy ◽

Clinical Implications ◽

Genetic Predictors ◽

Antihypertensive Response ◽

Drug Responsiveness ◽

Multiple Drugs

Background: The available antihypertensive drugs are effective and well tolerated agents. However, only about half of patients with treated hypertension achieve appropriate blood pressure control. Genetic and non-genetic factors contribute to the interindividual variability of the therapeutic response Objective: This review constitutes a comprehensive update of the pharmacogenomics of antihypertensive drugs and their clinical implications in Brazil. Results: Twenty-five studies explored the influence of gene variants on drug response in patients with primary, resistant, or gestational hypertension. Variants in BDKRB2, NOS3, PRKCA, and VEGFA influenced the response to enalapril in patients with primary hypertension. AGT and MMP2 variants were associated with high risk of resistance to antihypertensive treatment, whereas NOS2 variants were related to low risk. Moreover, NAT2 slow acetylators showed an increased response to hydralazine in patients with resistant hypertension. HMOX1, NAMPT, MMP9, NOS3 and TIMP1 variants might be markers of drug responsiveness in hypertensive or preeclamptic pregnant women. Power and replication of studies, polygenic nature of response to therapy, and treatment with multiple drugs were important challenges to be overcome for identifying genetic predictors of antihypertensive response in Brazil. Conclusion: Pharmacogenomic studies in Brazilian cohorts provide some evidences of variants, mainly in pharmacodynamics genes, which influence the response to antihypertensive drugs. However, some findings are limited by cohort size or therapeutic scheme and may be influenced by interactions with other genetic and non-genetic factors. Therefore, further investigations are needed to elucidate the contribution of pharmacogenomics to the efficacy and safety of antihypertensive therapy.

Download Full-text