Hot Spotting as a Strategy to Identify High-Cost Surgical Populations.

Abstract

The aim of this study was to determine the feasibility of "hot spotting" in elective surgical populations. Prospective identification of high-cost patients, known as "hot spotting," is well developed in medical populations, but has not been performed in surgical populations. Population-based management of surgical expenditures requires identification of high-cost surgical patients to allow for effective implementation of cost-saving strategies. Using 100% Medicare claims data for 2010 to 2013, we identified patients aged 65 to 99 years undergoing elective surgical procedures. We calculated price-standardized Medicare payments for the surgical episode from the index admission through 30 days after discharge. Patient-level factors associated with payments were analyzed by multivariable linear regression. Medicare patients in the highest decile of spending accounted for a disproportionate share of aggregate costs: 30% in Colectomy (COL), 22% in coronary artery bypass grafting (CABG), 19% in Total Hip Arthroplasty, and 18% in Total Knee Arthroplasty. Medicare expenditure differences between the highest and lowest deciles were because of a 5-fold difference for COL and 3-fold difference for CABG in index hospitalization cost. In contrast, for orthopedic procedures, there were 47- to 80-fold post-acute care expenditures between highest and lowest deciles. In multivariable analyses, patients with ≥3 comorbidities had significantly higher costs than healthier patients. We found that a subset of multimorbid patients was responsible for a disproportionate share of total Medicare spending, but the individual components of spending vary by procedure. These findings suggest that targeting high-cost Medicare patients (ie, hot spotting) for cost containment efforts would be a potentially effective strategy to reduce costs in surgical populations.