Healthcare payers and prescription plan administrators may have more power than they think when it comes to improving medication adherence among diabetic seniors, according to research from Kaiser Permanente, and exercising that power may be as simple as shifting a few costs around to make access to medications a little easier.

In a study published this month in Medical Care, Julie Schmittdiel, PhD, Research Scientist at Kaiser Permanente’s Northern California Division of Research led an investigation into health system-level factors that significantly influence medication adherence.

She found that adherence rose significantly for Medicare-aged seniors when out-of-pocket costs and co-payments were kept to a minimum and diabetic medications were delivered by mail in longer-term amounts.


“A lot of medication adherence literature and research really focuses on individual patient factors, like reminders, and individual patient barriers to medication adherence,” Schmittdiel told

“And while those are certainly issues, we felt that there were a lot of health system-level aspects that might facilitate ease of access and ease of use to really help diabetes patients, particularly older diabetes patients, really help improve their health.”

With medication adherence taking on extra significance as part of the Medicare Five Star Quality Rating System for health plans, payers must ensure that patients have access to their prescriptions at least 80 percent of the time, or risk those patients being classified as non-adherent.  Patients who do not take their medications as prescribed are at significantly higher risk of hospitalization, and incur millions of dollars in avoidable costs across the healthcare system.

Stakeholders have been attempting to tackle the problem from multiple angles, including providing more reminders and alerts for providers through their EHRs, easing the process of prior authorization through improved data exchange, and making it simpler to refill prescriptions through online patient portals.

But none of these interventions will be truly effective if patients cannot afford their co-pays or cannot leave their homes to travel to the pharmacy for pick-up. “Our understanding of how to help people take those medications regularly so that they can maximize the clinical efficacy of those medications can sometimes be really challenging,” Schmittdiel says.

The researchers used EHR and pharmacy data from more than 130,000 diabetic patients over the age of 65 in West Coast sections of the Kaiser Permanente system, focusing on three common diabetes prescriptions, including hypertension medications, statins, and antihyperglycemics.  Schmittdiel and her team found a 90 percent medication adherence rate among patients who experienced the following four factors:

• Prescription of a 90-day supply of their medications

• Use of a mail-order pharmacy for direct delivery of products

• Maximum annual out-of-pocket expenses of $2000 or less

• Medication co-pays of $10 or less per prescription for a 30-day supply

It may not seem like much of a surprise that providing cheap, easy access to medications will increase adherence, but quantifying the impact of specific interventions for a diabetic population at high risk of complications like cardiovascular disease is an important step for the industry, Schmittdiel says.  Research generally focuses on increasing a patient’s sense of responsibility, not on how health plans can effectively reduce barriers to active engagement from their end.

“One of the unique aspects of this study was the ability to focus in on the older population, which not a lot of medication adherence medication research has had the chance to do,” she said.  “Based on other work that we’ve done, we had thought it possible that more mail order pharmacy use might lead to better adherence.  When we actually focused in for the first time on this older population, we found that was indeed the case.”

“Price sensitivity does tend to happen at a pretty low level.  Five, ten, fifteen dollars per prescription – that can make a huge difference to individual patients, particularly older diabetes patients.  Many patients, when they’re diagnosed with diabetes, they might already have an average of five medications that they’re on.  So, it doesn’t take very much to have an impact on their ability to pay for their prescriptions.”

Kaiser Permanente researchers have the benefit of leveraging the data from the health system’s integrated services and standardized databases, which allow access to large data sets that encompass many different aspects of care delivery. “We had the very unique ability to look at all these factors at once and see how they really kind of had a synergistic effect,” agreed Schmittdiel.

schmittdiel_julie (1)“Our ability to measure medication adherence really came from the electronic health record pharmacy files,” she explained. “The vast majority of patients at Kaiser Permanente, and certainly the vast majority of older diabetes patients, have a Kaiser Pharmacy benefit.  We’re able to leverage that information to look at people’s refill history to know when a physician ordered a medication.”

“We were also able to use the pharmacy-based EHR information to look at mail order pharmacy use, because when we track a refill, we also track the mechanism by which it happens.  So we know if it happened at a brick and mortar facility, and which one of those it went to,” she continued.  “We know if it came from our mail order distribution system, and how many days’ supply was issued. The health care cost information came from administrative and benefit databases that we have.”

While Kaiser Permanente often leads the way with big data analytics and population health management research, other health systems and benefit plans can conduct their own analyses with tools that they likely already have on hand, Schmittdiel says.  “We do have a very rich and extensive EHR system, and some things we collect through our EHR I think maybe are less common than in some other setting, like BMI. But I think a lot of the variables that we were using are available through a lot of different EHRs.  There’s a wide range of data analytics and technology platforms that can support this particular type of work.”

“I also think it’s very likely, even though we did not study it, that this kind of work could generalize to medication adherence for other chronic diseases,” she added.  “We were very focused on diabetes and cardiovascular disease risk factor medications in this project.  But I would think, particularly for lower cost generic drugs that people are expected to take for a long time to help control chronic illnesses, these findings may also apply if you ask the right questions.”

“What are the things that work to help people stay on their medications for a long period of time?  What are really going to help them improve their quality of life and their clinical outcomes in the long run?  I do think that a lot of the findings from this work could be applied by health care systems and quality improvement programs to other types of chronic diseases.”

Full article: