The OPTIMIZE-HF database: heart failure core quality measures disappoint

An analysis of clinical outcomes in the OPTIMIZE-HF registry appeared in the January 3 issue of JAMA. An attempt to relate compliance with “core quality” measures for heart failure to clinical outcomes was somewhat disappointing, with most measures showing no effect. Angiotensin converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) were associated with no improvement in mortality through 90 days although a decrease in the composite endpoint of re-hospitalization and mortality did reach statistical significance. The most robust improvement in outcomes was associated with one evidence based treatment which was not one of the core measures: beta blockers.

DB has already ranted on this issue, stating “Perhaps the concept of performance measures has flaws.” (I would strike the word perhaps). He correctly points out “Medical care has greater complexity than simple rules would suggest.”

There are many reasons not to be surprised at these findings. The web based reporting system for the registry depended on chart documentation and was of questionable accuracy regarding eligibility for various interventions. The rush by hospitals to get a good report card may have diverted attention from other life saving therapies such as devices and aldosterone receptor antagonists.

The first quality measure was the provision of discharge instructions on medications, diet and other aspects of heart failure care. In one study on which this recommendation was based the instructions included a full hour of one-on-one verbal counseling. The intervention was associated with improved outcomes. The “core quality” measure, in contrast, required only that written instructions be given to the patient. It’s one thing to hand patients a ream of paper as they are rushed out the door and quite another to provide detailed counseling. Nominal compliance may earn the hospital a perfect report card while doing little of substance to help patients.

The second measure was documentation of left ventricular systolic function. This is important in the identification of patients who are candidates for certain therapies. I’m reminded of a recent anecdote (unconfirmed but believable) about a nurse-case manager who wouldn’t allow a heart failure patient to be discharged until an echocardiogram was done whether the patient needed one at that time or not.

The third measure was the prescription of an ACEI or ARB for patients with left ventricular systolic dysfunction. These medications have been proven to increase heart failure outcomes. The disappointing results in the OPTIMIZE-HF registry may be due to the short follow up period. The same reason may apply to the negative findings concerning the other two measures, smoking cessation counseling and anticoagulant use for patients with atrial fibrillation.

Can we make sense of the findings? It would be wrong to conclude that the measures are not worthwhile. On the other hand the findings add to growing skepticism of “quality” mandates as they are now implemented.