Authors: Health Economics & Outcomes Research (HEOR) Cohort Study (Purdue University)
Applies two-part Generalized Linear Models (Log-Link Gamma GLM) and propensity score matching to skewed administrative claims data across CKD stages 1-5 and ESRD.
Key Quantitative & Theoretical Takeaway: Transitioning to Stage 5 CKD and ESRD dialysis drives a non-linear surge in inpatient hospitalizations and total annual per-patient-per-year (PPPY) direct medical costs.
Autosomal dominant polycystic kidney disease (ADPKD) is the most common hereditary kidney disorder and fourth leading cause of end-stage renal disease. This study quantifies incremental healthcare resource utilization and direct medical expenditures across CKD stages.
Medical claims expenditures exhibit heavy right-skewness and zero-mass spikes. A Log-Link Gamma GLM models E[Y|X] = exp(X*beta) directly on the dollar scale without requiring Duan smearing retransformation.
Replace arxiv.org with nujan.app on any arXiv URL (for example, https://nujan.app/abs/2006.11239) to open the PDF and AI research partner side-by-side.