Phone-letter prompts improve first-time statin adherence

Calling and then sending follow-up letters to patients who failed to pick up newly prescribed statins improved medication adherence by 16 percentage points, according to a randomized trial published online Nov. 26 in the Archives of Internal Medicine. The results highlight both the opportunities and challenges of treating patients with coronary artery disease, one of the authors told Cardiovascular Business.

“Cholesterol medicine is probably one of the largest unmet opportunities in our system,” said Ronald D. Scott, MD, a family physician at Southern California Kaiser Permanente’s West Los Angeles Medical Center. As co-leader of cardiovascular disease prevention programs in the region, he provides clinical guidance in quality control initiatives such as the intervention detailed in the study. “If we can tweak our system and optimize use of cholesterol medicine, then we can prevent a lot of cardiovascular disease.”

Scott and colleagues conducted a randomized controlled trial that targeted primary adherence among Southern California Kaiser Permanente members who were prescribed statins as a new medication between April and mid-June 2010. They passively enrolled patients whose electronic records showed they failed to collect their medications after one to two weeks and randomized them to either an intervention group (2,606 patients) or a control group (2,610 patients).

Patients in the intervention group received an automated telephone call with a personalized message, information about the importance of taking the medication and encouragement to pick up the prescribed statin. If patients still had not obtained their medications one week later, they were sent a letter signed by the prescribing physician. The control group received no outreach calls or letters.

All patients were enrolled in batches weekly for 10 weeks. For the primary and secondary outcomes, researchers assessed statins collected within two weeks after delivery of the letter and refills at intervals during the following year.

In the intervention group, 42.3 percent of participants had statins dispensed compared with 26 percent of the control group. Secondary adherence was always greater in the intervention group up to one year out in the study. The refill rate was similar for both groups, at 35.1 percent in the intervention group and 35.5 percent in the control group.

“The key is that people who take statins have lower rates of heart attacks and stroke,” Scott said. “That has been consistently shown in the literature. In order to take the statin, they have to be primary adherent at some point. It is a first step.”

The authors wrote that the intervention included marginal costs, making it a feasible approach for improving primary adherence. They calculated the cost for the calls and letters as approximately $1.70 per person.

Scott said there were no plans to conduct a cost analysis. But he added that because the intervention was inexpensive, easy to implement and likely to improve outcomes, it would be cost saving. Southern California Kaiser Permanente already has implemented the intervention regionally and expanded it to other prevention medications.

Scott pointed to several opportunities to intervene and improve adherence, starting with initial communication with the patient, primary adherence once the physician orders the medicine and secondary adherence to coax patients back on their medications if they stop taking statins. “We’re making use of tools we’ve developed to try to promote adherence from the outset,” he said.

Kaiser Permanente offers resources that may not exist in other healthcare systems, though, wrote Michael A. Fischer, MD, MS, of the pharmacoepidemiology and pharmacoeconomics division at Brigham and Women’s Hospital in Boston. That includes an EHR that allows integration of medical and pharmacy data.

“Although the integration of the Kaiser Permanente system was likely helpful for this intervention, it also limits the generalizability of the findings to less integrated settings,” he wrote. “For example, in most clinical settings there is no mechanism available to link electronic prescriptions to pharmacy-filling records, as was done in this study.”

Nonetheless, Scott and colleagues have shown the benefits of data integration and proof of its value, Fischer wrote. They provided “a set of tools with which we can chip away at the challenge of nonadherence. Improved integration of clinical systems may allow for more efficient and practical implementation of multimodal interventions to address multiple components of adherence.”