Outcome, efficiency, and time-trend pattern analyses for stroke rehabilitation.

We present a series of pattern analysis techniques using inpatient medical rehabilitation as an example setting, stroke as a model diagnosis, and various versions of the Functional Independence Measure-Function Related Groups as a case mix adjuster. The pattern analytic approach uses severity-adjusted benchmarks to distinguish among groups of patients whose outcomes and resource use patterns are lower, typical, or higher than established ranges at set points in treatment. Changing outcome and resource use patterns were illustrated for 1990, 1992, and 1995 discharges (n=34,734). These pattern analytic approaches are applicable to any postacute care rehabilitation setting, assuming the availability of an appropriate case mix adjuster and a reliable patient information system. These techniques provide clinicians with tools to help maintain quality in this era of managed care.