3063 — Use of Heart Failure Clinics and Quality of VA Heart Failure Care

Sahay A (CHF QUERI), Heidenreich PA (CHF QUERI)

Objectives:
Many medical practices have established clinics that specialize in heart failure care (HF). The prevalence of these clinics within the VA is unknown. We sought to determine the prevalence of HF clinics and their association with quality of VA HF care.

Methods:
We performed a web-based survey of either the Chief of Cardiology or Chief of Medicine at each of 144 VA Medical centers. Each site was asked if they have a clinic that specializes in heart failure. American Hospital Association data were used to determine teaching status defined as membership in Council on Teaching Hospitals. Hospital data were then linked to HF performance measure data from chart review (random sample during 2004-2005) collected as part of the VA’s External Peer Review Program. We determined the association between presence of a HF clinic and quality of care for established (documentation of left ventricular ejection fraction (EF), discharge instructions, outpatient use of angiotensin converting enzyme inhibitors (ACEi) use in those with EF < 40%), and new measures (outpatient use of recommended beta-blockers (BB) in those with EF < 40%) after adjustment for teaching status, patient demographics and clustering of patients within hospitals using generalized estimating equations.

Results:
Surveys were completed by 89 VA medical centers (62%) and 41 (46%) reported having a HF clinic. Hospitals with an HF clinic were more likely to be teaching (84% vs. 64%, p < 0.0001) and have a cardiologist specializing in HF (68% vs. 10% p < 0.0001), but were similar in use of home monitoring (73% vs. 60%, p=0.22). Patients from hospitals with a HF clinic were slightly older (71.5 vs. 69.9, p < 0.0001) and more likely to be smoking (25% vs. 21%, p=0.0001) than those without an HF clinic. Quality of care data from these hospitals was available for 10,983 patients. Patients from hospitals with HF clinics were more likely to be treated with recommended BB (51% vs. 43%, p < 0.0001) and receive discharge instructions (85.4% vs. 83.6%, p=0.02), but were similar in EF measurement (94% vs. 93%, p=0.5), and ACEi use (89% vs. 89%, p=0.3). After adjustment for clustering within hospitals, teaching status, and patient characteristics a HF clinic remained associated with use of recommended BB (odds ratio 1.43 95% CI 1.08 to 1.91).

Implications:
Performance on traditional HF quality measures is high in the VA. Outpatient use of recommended beta-blockers is not optimal but is higher in those hospitals with a HF clinic after adjustment for patient and hospital characteristics.

Impacts:
Additional studies should explore the degree to which a VA HF clinic is responsible for, or a marker of, high quality care.