Circ: Medication adherence as a performance metric doesnt measure up
 |
| Source: American Heart Association |
The study was published in the May issue of Circulation: Cardiovascular Quality Outcomes.
Support for the use of chronic medication adherence and treatment intensification as measures for quality improvement has been mixed, Rebecca Vigen, MD, MPA, of the cardiology division at the University of Colorado School of Medicine in Denver, and colleagues wrote.
In the American College of Cardiology and the American Heart Association’s 2011 performance measures for coronary artery disease and hypertension, the writing committee decided against the inclusion of medication adherence, reasoning that it may be beyond the physician’s control, may prompt some physicians to avoid taking on nonadherent patient and may add a cost burden for providers that lack access to pharmacy records (Circulation 2011;124:248-270). The National Quality Forum, on the other hand, endorsed a measure on adherence to chronic medications in 2009.
Treatment intensification has been shown to improve blood pressure control at the patient and clinic level. “However, prior studies have not assessed the association between adherence with blood pressure control at the clinic level, nor the impact of adherence and treatment intensification together on blood pressure control beyond a patient-level analysis,” Vigen and colleagues wrote.
For their study, they used the Cardiovascular Research Network Hypertension Registry to identify 162,879 patients among 89 clinics with incident hypertension who were started on antihypertensive medications between 2002 and 2006. The primary outcome was blood pressure control at 12 months after the initial treatment.
They calculated medication adherence as the proportion of days covered for antihypertensive medications (beta-blockers, ACE inhibitors or ARBs, dihydropyridine calcium channel blockers, nondihydropyridine calcium channel blockers and diuretics). Treatment intensification was calculated using a validated standard-based method with scores from minus 1 (no treatment intensification), 0 (treatment intensification occurred when blood pressure was elevated) and plus 1 (treatment intensification was made at all visits, even when blood pressure was not elevated).
At the patient level, the mean adherence was 0.77. For each 0.25 increase in medication adherence, patients had a 28 percent increase in the odds of blood pressure control. When the researchers categorized clinics into quintiles, they found little variation in adherence among quintiles. The mean adherence was 0.78 at the clinic level. For every 0.05 increase in medication adherence, the odds of patients achieving blood pressure control decreased by 7 percent.
Mean treatment intensification was 0.026 at the patient level and 0.01 at the clinic level. At the patient level, for each 0.25 increase in treatment intensification, the odds of achieving blood pressure control increased by 55 percent. At the clinic level, each 0.04 increment increase in treatment intensification was associated with a 25 percent increased odds of achieving blood pressure control.
“At the level of the clinic, the association between adherence and blood pressure control was not in the expected direction, whereas TI [treatment intensification] was associated with blood pressure control,” the authors wrote. “These findings raise the question of whether adherence when aggregated to the clinic level can be used to distinguish between high- versus low-performing clinics, as commonly done with performance measurement.”
They suggested that several factors may explain their findings about adherence at the clinic level. Averaging adherence levels among patients may have decreased variability, for instance, and the clinics used in the study had little variability in adherence. If clinics with greater variability were included then an association between adherence and blood pressure control may have become apparent.
They noted that the clinics in the study were integrated, managed care organizations that may not be representative of other healthcare systems. They also pointed out that the patients in their study had higher adherence levels than reported in other analyses.
Their findings suggested that medication adherence should not be used as a performance measure for facilities but treatment intensification could be a possible measure if some challenges are addressed. Those included the feasibility of data capturing, overcoming providers’ reluctance to intensify treatment in nonadherent patients and providers’ concerns that intensifying treatment may have adverse effects for some patient groups such as the elderly or diabetics.