Medication nonadherence among patients, particularly those with chronic diseases such as diabetes, has become an expensive problem for the American medical community, with a direct cost of approximately $100 billion annually that may range as high as $300 billion in potentially avoidable spending. Nonadherence in the United States is associated with 125,000 deaths annually and at least 10 percent of hospitalizations.

Hayden B. Bosworth, PhD, professor and vice chair of education in Duke’s Department of Population Health Sciences, and colleagues, have contributed to several studies on nonadherence demonstrating that 50 percent of patients do not take medications as prescribed. More than one in five new prescriptions go unfilled, the research finds, and 50 percent of patients discontinue medication within the first year. A search for effective interventions is underway on several fronts.

As much as 80 percent of nonadherence may be intentional, Bosworth says, due to such factors as uninformed decision-making, limited or poor communication with health professionals, patient disengagement, complex drug regimens, financial and cognitive constraints, and mismatched preferences between patient and provider.

“Nonadherence affects all disease conditions and remains a fundamental problem across the board,” Bosworth says. “For cardiovascular and oral diabetic medications, adherence is particularly poor.”

David A. D’Alessio, MD, chief of the Duke Division of Endocrinology, Metabolism, and Nutrition, describes nonadherence as “the number one problem in diabetes care.” Research indicates that patients with diabetes generally keep their appointments, Bosworth says, but they are not diligent in monitoring glucose. “For patients with diabetes, the problem seems to be one of measuring the level of adherence,” he says. “Unfortunately, it’s not easy to do.”

Physicians may not be aware of patient nonadherence; additionally, the limited time physicians can spend with patients during office visits may be a factor. Concerns about nonadherence have intensified as prescription rates increased 39 percent during the past decade, Bosworth adds, citing two statistics:

• Four out of every five American adults take at least one medication per week.
• On average, 20 percent of an employer’s population have at least one chronic condition and take four or more medications. The numbers account for 80 percent of health care costs for an employer.

The effort to develop interventions has proven challenging, Bosworth says, because of the difficulty in identifying nonadherent patients and finding the most effective juncture to help them. Many options are being explored, but post-appointment phone calls to patients by care managers has demonstrated initial effectiveness.

One brief intervention involving tailored disease/self-management support for patients on Medicaid—care managers placed at least one call to patients with increased risk of cardiovascular disease—resulted in improvements in hypertension medication adherence. Bosworth says personal interaction by phone allowed the intervention to be tailored to participants’ concerns, health goals, and specific barriers to achieving goals.

Other ideas include incorporating blister packaging designed to schedule or synchronize dosages. Bosworth encourages providers to screen for nonadherent patients by using EHRs to monitor follow-up visits or prescription pick-ups. Providers should also prompt patients to take medications when performing a routine daily task such as brushing teeth.