Abstract:
This study attempted to determine the effectiveness of
rehabilitation for TBI using a non-randomized design. Participants
consisted of a consecutive series of 365 individuals with TBI who were
discharged from a Level I trauma center either to inpatient
rehabilitation or home (78% follow-up). Outcome measures included
Glasgow Outcome Scale (GOS), Sickness Impact Profile (SIP), Burden
Inventory, and Perceived Quality of Life (PQOL). We found that
discharge to inpatient rehabilitation was associated with significantly
poorer functioning on the GOS, non-significantly poorer functioning on
the SIP, non-significantly increased care-giver burden as measured by
the Burden Inventory, and non-significant improvement on the PQOL.
Similar results were found for longer lengths of inpatient
rehabilitation stay. However, these findings do not appear to result
from the effects of rehabilitation, but rather from the effects of
confounding. The confounding primarily stems from unmeasured or
difficult to measure aspects of the clinical decisions for discharge
placement and rehabilitation length. This confounding could not be
controlled for by the observational study design and data analysis,
i.e., regression methodology. Furthermore, typical severity indices
were inadequate to control for injury severity and recovery. Matching
designs investigating TBI rehabilitation also appear to be at risk for
inadequate confounder control.