Predicting charges for inpatient medical rehabilitation using severity, DRG, age, and function

We examined the effectiveness of using diagnosis related groups (DRGs), Severity of Illness Index (SII), age and function at admission to predict inpatient charges for medical rehabilitation. Data from our sample of 199 indicate that DRGs alone explained approximately 12 per cent of the variation in charges for inpatient rehabilitation while SII explained 26 per cent of the variation. SII, DRG, and age together yielded the highest regression coefficient, accounting for nearly 39 per cent of the variation in total charges; SII and age accounted for 36 per cent of the variation. Within DRG categories, SII was the only important predictor of inpatient charges accounting for 23 per cent of the variation in charges among stroke patients (DRG 014) and 28 per cent of the variation in charges among hip fracture patients (DRG 210). Function at admission was not a useful predictor of inpatient rehabilitation charges within DRGs. These results suggest that SII and age may be useful in developing a DRG-based prospective payment system for inpatient medical rehabilitation.