How do hospitals respond to cross price incentives inherent in diagnosis‐related groups systems? The importance of substitution in the market for sepsis conditions

This study addresses the question of how hospitals respond to the cross price incentives inherent in reimbursements based on diagnosis‐related groups (DRG). Unique market‐wide administrative data allow to exploit a natural experiment in Germany in which the relative attractiveness of greatly divergent reimbursements for clinically similar patients changes in the market for sepsis conditions on January 1, 2010. This natural experiment provides—unintentionally—extra reimbursements in cases in which hospitals reorganize transfers for deceasing patients to other facilities, alter the time of death, the choice of the condition being chiefly responsible for the hospital admission (primary diagnosis), or the intensity of mechanical ventilation. The differences‐in‐differences results demonstrate that hospitals primarily alter the primary diagnosis. As the choice of the primary diagnosis is the backbone of the design of modern DRG systems, the findings suggest that payment contracts between hospitals and payers based on modern DRG algorithms may not necessarily improve patient welfare.