PM360 asked experts in helping patients adhere to their treatment regimens how COVID will impact the ability of patients to do that moving forward and how data can help. Specifically, we wanted them to address one of two questions:
- What long-term impact will COVID-19 have on adherence? Are there any specific barriers, social determinants of health (SDOH) factors, or other issues life sciences companies should pay particularly close attention to?
- What data sets or types of research can companies best rely on to determine the reasons behind patient nonadherence? What should companies look for to better monitor nonadherence or identify potential barriers?
Digital health programs and solutions are helping providers bridge the many gaps of SDOH that prevent patients from adherence—and consumers are more willing than ever to participate in such programs. A September 2020 ORCHA report found a 25% increase in daily health app downloads—up from four million to five million a day.
As market-wide consumer participation increases, insurers and providers should consider leveraging that interest into building incentive programs that reward daily engagement in healthy practices. According to a 2019 Change Healthcare survey, nearly half (49.3%) of healthcare industry leaders said that offering incentives for healthy behaviors would turn their patients into more active healthcare consumers.
The macro effects of greater digital health solution adoption could create more opportunities for providers to capitalize on value-based care models that pay based upon the quality of their care, rather than the quantity of services they provide. According to a Deloitte report, 36% of U.S. physicians drew some form of compensation from value-based payments in 2020, and that number is expected to grow.
Addressing SDOH has many facets, but providers have an opportunity to lead the way in what they do best—connect with their patients for improved outcomes.
First, many living with chronic health conditions have long felt isolated due to various aspects of their conditions, which can significantly affect adherence. But now that people globally have experienced similar levels of isolation due to the pandemic, there may be both greater empathy for people with chronic conditions and a greater dedication to combating adherence issues on a larger scale.
Second, the pandemic helped illuminate the many factors with long-lasting effects that can complicate adherence—from stress to an inability to obtain prescriptions or access treatments. In a recent Health Union survey, 37% said overall control of their conditions has worsened since the start of the pandemic and more than a tenth cited increased difficulties with taking prescriptions exactly as directed. The illumination—and magnification—of these challenges will hopefully motivate pharma companies to find solutions.
In terms of barriers to adherence, some of the evergreen challenges remain; however, life sciences companies should pay attention to the role relationships with HCPs play in treatment adherence. Finding ways to improve those experiences will remain extremely important.
In a good and rational economic model of human behavior, a diagnosis and prescription would compel a good and rational patient to adhere to good and rational treatments. But human behaviors in the real world don’t align with good and rational economic models. We must examine the act of prescribing to truly improve adherence.
Traditional medical models rely too much on directives from experts and not enough on the holistic challenges of messy, random, beautiful human life. Adherence is more than a twice-daily pill; it signals intent from a patient and an agreement with a physician (and the manufacturer). It is a commitment, and as such should be viewed through the lens of social contracts.
Solutions to adherence are not about gamification models, tricks, gimmicks, or complex systems of reminder emails. Start with a shared decision between prescriber and patient, along with a statement of intent. A plan. It should be an investment, one that creates a sense of responsibility and pride of ownership.
The physician scribbles a recommended treatment on an Rx pad. Why not a “contract”? Agreeing to terms and co-signing will create a vested interest. After that? Gamify whatever; at least the players have signed up for it.
The gold standard for health data related to adherence is and remains Proportion of Days Covered (PDC); however, there are gaps in the accuracy PDC provides. For starters, PDC uses one metric for polypharmacy and does not account for gaps and delays with regards to multiple medications. Additionally, PDC is simply a measure of medication on hand or delivered and not necessarily taken.
As a result, barriers to nonadherence can be challenging to quantify, including health literacy, mood, cognitive, and psychological barriers. These can, however, be elicited through well accepted and validated means of assessment. Similarly, these data may be available in complementary data sets related to mental health, refills, and/or claims data. For example, an instance of a mental health medication within a claim file can be indicative of/predictive of adherence behavior and used to manage an intervention. Much work has also been done in using historical refill data and polypharmacy and provider switching to predict future medicine-taking behavior.
Advanced analytics coupled with domain knowledge in reverse-engineering clinical data, claims, and ambient mobile phone data can be very powerful in this new endeavor.
Reasons for nonadherence to chronic medications is multifactorial and predicting human behavior is quite difficult, but having rich data sets such as administrative, pharmacy and medical claims, geographic determinants, patient surveys, and counseling interviews/clinical consultations help provide a better understanding of adherence barriers.
Monitoring prescription claims data for disruptions in filling patterns should be done at least weekly, or even daily if administratively possible. Statistical models predicting nonadherence greatly improve odds of finding at-risk populations, while well-designed analytic evaluations provide program effectiveness information necessary to guide operations; however, patients and programs are not created equal, in that what works for one will not necessarily work for another.
Some patients may forget only on occasion to take their medications and a simple automated refill reminder phone call is enough. Others may face more difficult challenges and require direct engagement when trying to understand and improve medication nonadherence. And of course, costs and associated return on investments vary greatly along this gradient of outreach modalities. The best solution to improving adherence is one that is versatile, yet highly targeted to the specific needs and/or barriers of selected cohorts, and scalable to large populations while maintaining cost effectiveness for program sustainability.
Research shows that the strongest indicator for patient nonadherence is tied to behavior, but for a multitude of different reasons. In fact, there are more than 190 reasons why patients don’t take their medications, ranging from medication fatigue to financial issues to social stigma to medication side effects. Not all of these issues have a simple solution, but they don’t necessarily have to be barriers to medication adherence if patients have the right resources and support to overcome them. Each individual is different, meaning that each person’s motivations will be different as well.
The key to truly bettering patient medication adherence is to deliver personalized guidance and timely reinforcement at the time that specific obstacles occur so that a single issue or challenge doesn’t completely derail the patient journey. By integrating technologies such as artificial intelligence, medication management tools are able to create the individualized journey that analyzes a patient’s behavior and present the right support to overcome those roadblocks and stay adherent. This data can then be used to better personalize the journeys of other patients by continuously learning from the data collected in order to provide the most individualized experience possible.