Oral anticoagulants (OAC) are effective and safe if the international normalized ratio (INR) is maintained within a narrow therapeutic range.  Hospitalization is independently associated with poor anticoagulation control.  The objective of this study is to describe how anticoagulation control changes in the peri-hospitalization period.

This study is a retrospective cohort study using population-based administrative databases.  INR results were retrieved from a population-based laboratory database.  INR levels between laboratory measures were estimated using linear interpolation.  Auto-regressive, integrated, moving average (ARIMA) time-series modeling was used to determine how anticoagulation control changed in the peri-hospitalization period.  The study included 5,380 elderly patients in Eastern Ontario between September 1, 1999 and September 1, 2000 taking OACs.

Results showed that 951 (17.7%) were hospitalized during the OAC therapy [thrombotic, n=52 (1.0%); hemorrhagic, n=140 (2.6%); other hospitalization types, n=759 (14.1%)].  All measures of anticoagulation control changed significantly in the peri-hospitalization period.  Before hemorrhagic admissions, mean INR and proportion with INR >5 increased significantly (daily increase 0.024, P=.03 and 0.2%, P=.01).  Following other hospitalization types, the proportion of patients with INR < 1.5 was significantly increased (daily increase 0.19%, P=.02).

Patients admitted to the hospital for a variety of indications have significantly worse anticoagulation control in the peri-hosptialization period.  Anticoagulated patients discharged after medical hospitalizations could be targeted for improved anticoagulation control.