Physical Activity Predicts Optimal Patterns In Hypertension
A study trained a machine-learning model on 71,637 UK Biobank participants and externally validated it in 5,104 NHANES adults to predict individualized accelerometer-derived physical activity (PA) patterns for people with high blood pressure. The model achieved a 10-year mortality AUC of 86.4%, identified active regular, active light, and weekend-warrior patterns as optimal across participants, and found inconsistency with the predicted pattern linked to a 28% higher all-cause mortality (HR 1.28).
Scoring Rationale
High novelty and strong external validation drive score, limited by focused hypertension scope and observational design.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
