摘要：We examined the ability of MedisGroups, a severity measure based on clinical data abstracted from the medical record, to predict mortality 30 days following admission. MedisGroups measures severity both on admission and approximately one week into the hospital stay. The data base was a random sample of 20,985 admissions of Medicare beneficiaries with one of six conditions from 833 hospitals in seven states between January 1985 and May 1986. In all six conditions, higher admission and mid-stay severity scores were generally associated with higher risk of death. Across the six conditions, the R-squared values for predicting death using admission scores ranged from 0.01 to 0.16; R-squared values using mid-stay scores ranged from 0.03 to 0.34; and R-squared values from combining admission and mid-stay scores ranged from 0.05 to 0.41. Admission MedisGroups score was thus only modestly predictive of 30-day mortality. While the mid-stay score was more powerful, it may not be an appropriate severity adjuster to screen for quality using hospital mortality rates because it could be influenced by substandard care.
