Impact of institution of a stroke program upon referral bias at a rural academic medical center

Context: Referral bias reflecting the preferential hospital transfer of patients with intracerebral hemorrhage (ICH) has been demonstrated as the major contributing factor for an observed high nonrisk-adjusted in-hospital crude acute stroke mortality rate at a rural academic medical center. Purpose: This study was done to assess the impact of a clinical acute stroke program upon referral bias in August 2000. Methods: A chart review of acute stroke (DRG 14) discharges during 2001 from a rural academic medical center was compared with the same data from 1999. Results: The odds ratio of ICH in hospital-transfer patients compared with nonhospital-transfer patients decreased from 11.7 in 1999 to 3.2 in 2001 (P