Real world hypertension treatment patterns analysis identifies putative gaps and risk-associated patterns
Bandhakavi, S.; McCammon, J.; Karigowde, S.; Liu, Z.; Ni, X.; Rahmanian, F.
Show abstract
ObjectiveAlthough treatment patterns analyses at scale can provide insights into associated health outcomes, they remain relatively uncharacterized for most chronic diseases, including hypertension (HTN). To address this gap, we analyzed HTN treatment patterns among US health-insured patients. Materials and MethodsNew (n = 200,786) or all (n = 4.1 million) HTN patients were identified from 2015, 2016 enrollments in a nationwide administrative claims database and compared for HTN-specific treatment (Rx) choices, Rx count, and distinct rounds of treatment options (ROTO), respectively. Selected treatment patterns were risk-assessed using predictive modeling and/or literature-based recommendations. ResultsFor 2016, ACE inhibitors/ARBs were most frequent choices in new HTN patients vs more diverse Rx-choices among all HTN patients. All HTN patients had [~]3-fold and [~]5-fold-higher prevalence of (same-year) polytherapy and Rx interventions, respectively. New HTN patients with [≥]2 rounds of treatment options (vs single/initial round) were associated with [~]5-fold or higher predicted HTN complications risk (p-value < 0.0001). All HTN patients with [≥]3 rounds of treatment options (vs single/initial round) had 3.8-fold higher next-year HTN complications risk (p-value < 0.0001). Co-morbidities/persistence of [≥]3 rounds of treatment options over 2 years further increased these odds and total medications/chronic disease score correlated with ROTO counts. Discussion[~]95% of new HTN patients in 2016 did not start treatment with current literature-recommended first option, Thiazides. Assuming "sticky" prescription patterns, opportunity exists to improve (current) initial HTN treatments. Additionally, ROTO counts can inform HTN complications risk/management thereof. ConclusionWe highlight opportunities to improve initial HTN treatments and treatment patterns associated with higher risk among HTN patients.
Matching journals
The top 8 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Automated Interpretable Discovery of Heterogeneous Treatment Effectiveness: A Covid-19 Case Study 92%
- Disease Network Delineates the Disease Progression Profile of Cardiovascular Diseases 91%
- Natural language processing for scalable feature engineering and ultra-high-dimensional confounding adjustment in healthcare database studies 91%
Similar papers in this journal
- Pharmacogenomic and drug interaction risk associations with hospital length of stay among Medicare Advantage members with COVID-19 93%
- The Contribution of Pharmacogenetic Drug Interactions to 90-Day Hospital Readmissions: Preliminary Results from a Real-World Healthcare System 92%
- Development and validation of decision rules models to stratify coronary artery disease, diabetes, and hypertension risk in preventive care: cohort study of returning UK Biobank participants 91%
Similar papers in this journal
- Large Language Models Facilitate the Generation of Electronic Health Record Phenotyping Algorithms 91%
- Clinical Utility of Automatable Prediction Models for Improving Palliative and End-Of-Life Care Outcomes: Towards Routine Decision Analysis Before Implementation 91%
- Learning Decision Thresholds for Risk-Stratification Models from Aggregate Clinician Behavior 90%
Similar papers in this journal
- Pretrained Patient Trajectories for Adverse Drug Event Prediction Using Common Data Model-based Electronic Health Records 90%
- Clinical trial emulation can identify new opportunities to enhance the regulation of drug safety in pregnancy 89%
- Precision Prognostics for Cardiovascular Disease in Type 2 Diabetes: A Systematic Review and Meta-analysis 89%
Similar papers in this journal
- Development and Application of Pharmacological Statin-Associated Muscle Symptoms Phenotyping Algorithms Using Structured and Unstructured Electronic Health Records Data 93%
- Clinical interpretation of machine learning models for prediction of diabetic complications using electronic health records 92%
- Determining prescriptions in electronic health care (EHR) data: methods for development of standardised, reproducible drug codelists 91%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.