3027 — Atrial Fibrillation Management in Patients with Mental Illness: Are They Anticoagulated? Are They Monitored?
Frayne SM (COE, VA Palo Alto), Turakhia MP
(COE, VA Palo Alto), Schmitt S
(COE, VA Palo Alto), Friedman SA
(COE, VA Palo Alto), Chiu V
(COE, VA Palo Alto), Moos R
(COE, VA Palo Alto), Heidenreich P
(COE, VA Palo Alto), Berlowitz D
(COE, VA Bedford), Phibbs CP
(COE and HERC, VA Palo Alto)
Warfarin anticoagulation can reduce stroke risk in atrial fibrillation (AF), but carries bleeding risk. We examined, (1) in patients with AF, whether mental health conditions (MHC) are associated with lower warfarin receipt; (2) among warfarin-treated, whether those with MHC complete guideline-recommended monitoring of International Normalized Ratio (INR).
We identified 148,942 VA patients with AF in FY02/FY03 VA/Medicare databases. Applying AHRQ’s mapping, we identified 14 specific ICD9-based MHCs. Excluding indeterminate MHC cases, the final cohort was 123,567. We estimated adjusted odds ratios (AOR) for FY04 warfarin receipt as a function of MHC, controlling for CHADS2 stroke risk index. Among 34,959 warfarin-treated patients, we used FY04 dates of INRs (VA, Medicare, or Fee Basis) to calculate “INR Monitoring Rate,” accounting for monitoring-ineligible days: 100% reflects the equivalent of perfect monitoring. We examined INR Monitoring Rate by MHC.
21,829 (17.7%) had MHC. 49.6% with MHC vs. 52.7% without MHC received warfarin; AOR (MHC vs no MHC) was 0.88 (0.85, 0.90). Common MHCs were PTSD, depressive disorders, other anxiety disorders, psychotic disorders, and alcohol use disorders. Among patients with these conditions, 57.3%, 51.3%, 49.0%, 45.8%, and 44.3%, respectively, received warfarin, with AOR (specific MHC vs no MHC) 1.24 (1.15, 1.34), 0.93 (0.89, 0.97), 0.85 (0.80, 0.90), 0.75 (0.68, 0.82), and 0.69 (0.63, 0.76), respectively. Among warfarin-treated, median INR Monitoring Rate was high for patients with and without MHC (89.2% vs 88.6%, p = 0.004), and varied modestly by specific MHC (86.7% for psychotic disorders to 90.1% for PTSD).
AF patients with MHC were less likely to receive warfarin, even controlling for stroke risk. This varied by MHC type. Warfarin-treated MHC patients had high INR monitoring rates, comparable to those without MHC.
Those with MHC are less likely to receive warfarin, perhaps reflecting patient factors (e.g., access, contraindications) or provider factors (e.g., competing demands, perceived adherence). However, over 10,000 AF patients with MHC do receive warfarin in VA; most are monitored regularly. If further inquiry supports the ability of some with MHC to take warfarin safely, this may inform clinical guidelines about whether MHC should be considered a warfarin contraindication.