Who diagnosed and prescribed what? Using provider details to inform observational research

Bushnell, Greta A.; Stürmer, Til; Mack, Christina; Pate, Virginia; Miller, Matthew

doi:10.1002/pds.4685

Cited by 5 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent research indicates an interaction between age and sunshine hourssunshine hours as well as latitude lines [13]. Based on the results of previous research on the effects of various factors on the use of benzodiazepines [5], with additional education and revision of guidelines, it is necessary to precisely define which profile of experts has the right to prescribe this group of drugs [2,14].…”

Section: Discussionmentioning

confidence: 99%

Sunshine hours and utilization of benzodiazepines

Marković¹,

Nakarada²,

Mojović³

et al. 2021

Hosp Pharm Int Multi J

View full text Add to dashboard Cite

Introduction: Benzodiazepines belong to the group of anxiolytic sedatives and the most prescribed drugs in the world. The authors conducted this pilot study based n large number of patients reports that they were less anxious and felt better overall by migrating to sunnier environments. Aim: We estimated the impact of the total sunshine hours per month on daily exposure of the population to Benzodiazepine anxiolytic on a monthly basis in the Republic of Serbia. Material and Methods: We conducted a retrospective academic IV phase study, which tested the correlation of time series of monthly sold Benzodiazepine anxiolytics in pharmacies and total sunshine hours per month in the Republic of Serbia, from January 2014 to May 2019. Results: According to the seasons, we did not find any difference in the increase of the population exposure to benzodiazepines. A difference was found between bromazepam consumption compared to cumulative monthly. Bromazepam consumption was negatively correlated with the cumulative monthly sunshine hours. The cutoff value for bromazepam was 131.45 cumulative monthly sunshine hours. Conclusion: This is the first pharmaco-epidemiologic study to prove that the population's exposure to benzodiazepines, in addition to the implicit health and economic effects on physicians' prescribing habits for these drugs, can be independently linked to the cumulative effects of monthly sunshine hours on the population. No difference in Benzodiazepine consumption was found in relation to the seasons (spring/summer vs autumn/winter). In months where sunshine hours was ≥135.45, the need for bromazepam was significantly reduced. Seasonal variation of climatic effects of total monthly sunshine hours is significantly correlated with the population's need for benzodiazepines with anxiolytic effects longer than 12 hours.

show abstract

Section: Discussionmentioning

confidence: 99%

Sunshine hours and utilization of benzodiazepines

Marković¹,

Nakarada²,

Mojović³

et al. 2021

Hosp Pharm Int Multi J

View full text Add to dashboard Cite

show abstract

“…(3) What might be the best model architecture to do so? As most antidepressant prescriptions are initiated in non-psychiatric settings 44 , we focused on patients for whom treatment was started by a non-psychiatrist physician (e.g. primary care physician).…”

Section: Introductionmentioning

confidence: 99%

AI-assisted prediction of differential response to antidepressant classes using electronic health records

et al. 2023

Self Cite

View full text Add to dashboard Cite

Antidepressant selection is largely a trial-and-error process. We used electronic health record (EHR) data and artificial intelligence (AI) to predict response to four antidepressants classes (SSRI, SNRI, bupropion, and mirtazapine) 4 to 12 weeks after antidepressant initiation. The final data set comprised 17,556 patients. Predictors were derived from both structured and unstructured EHR data and models accounted for features predictive of treatment selection to minimize confounding by indication. Outcome labels were derived through expert chart review and AI-automated imputation. Regularized generalized linear model (GLM), random forest, gradient boosting machine (GBM), and deep neural network (DNN) models were trained and their performance compared. Predictor importance scores were derived using SHapley Additive exPlanations (SHAP). All models demonstrated similarly good prediction performance (AUROCs ≥ 0.70, AUPRCs ≥ 0.68). The models can estimate differential treatment response probabilities both between patients and between antidepressant classes for the same patient. In addition, patient-specific factors driving response probabilities for each antidepressant class can be generated. We show that antidepressant response can be accurately predicted from real-world EHR data with AI modeling, and our approach could inform further development of clinical decision support systems for more effective treatment selection.

show abstract

“…Here, we compare a total of eight deep-learning based NLP modeling approaches to classify antidepressant response during clinically relevant time windows after the first antidepressant treatment was given. Because the vast majority of antidepressant treatments are initiated by primary care clinicians rather than psychiatrists, 8 we focus on patients started on medication by non-psychiatrists. We show that, despite some differences between models, the deep-learning based NLP models overcome prior limitations that have constrained the classification of antidepressant response outcomes.…”

Section: Introductionmentioning

confidence: 99%

Phenotyping Antidepressant Treatment Response with Deep Learning in Electronic Health Records

Sheu

Magdamo

Miller

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Efficient, accurate phenotyping for antidepressant treatment response in electronic health records (EHRs) could facilitate precision psychiatry applications but remains a challenge. Increasingly, artificial intelligence methods using "deep learning" applied to clinical data have shown promise in complex classification problems. Here, we systematically evaluate the performance of eight deep-learning-based natural language processing models in classifying response to antidepressants in a large real-world healthcare setting. We obtained data spanning 1990-2018 for adults with depression and a co-occurring antidepressant prescription from the EHR data warehouse of the Mass General Brigham healthcare system (n=111,572). Clinical notes were collected for the following time windows after antidepressant initiation: (1) 2 days to 4 weeks, (2) 4-12 weeks, and (3) 12-26 weeks. A stratified random sample of these note sets (total 4,299 across time periods) were manually reviewed to classify response status as "improved" or "no evidence of improvement" in depression symptoms. All models performed well, with areas under the receiver operator curve (AUROC) of at least 0.80. Positive predictive values (PPVs) ranged from 0.72 - 0.91. In general, models incorporating more information-dense and longer text sequences performed better than others. The best performing model (Longformer-large with sliding window) had an AUROC = 0.88 and PPV = 0.84 at a specificity of 0.88. Our results indicate that deep learning methods applied to EHR data can accurately classify antidepressant response in a real-world healthcare setting. Automated treatment response classification may facilitate a range of research and clinical decision support applications.

show abstract

Who diagnosed and prescribed what? Using provider details to inform observational research

Cited by 5 publications

References 13 publications

Sunshine hours and utilization of benzodiazepines

Sunshine hours and utilization of benzodiazepines

AI-assisted prediction of differential response to antidepressant classes using electronic health records

Phenotyping Antidepressant Treatment Response with Deep Learning in Electronic Health Records

Contact Info

Product

Resources

About