Patient Nonadherence: A Symptom, Not a Disease State

Prescription fill history may not tell the full story of patient nonadherence.

Over the past few years, medication nonadherence has come to be recognized as an urgent health care problem that needs to be addressed. Despite roughly 2 decades of compelling studies indicating that wide swaths of our patient population do not fill their prescriptions regularly or at all, it was only when the economic consequences of medication nonadherence were more fully understood that the larger health care policy and advocacy community took up the cause.

The push to recognize medication nonadherence as one of the most pressing problems in our health care system has been so successful in recent years that it has become increasingly utilized as an outcomes measure in accountable care and performance-based contracting, most notably in the Medicare star ratings program. As far as widely recognized threats to public health and well-being go, medication nonadherence has seen a meteoric rise in prominence compared with typically decades-in-the-making health policy foci such as smoking cessation, vaccinations, and preventive health screenings.

Adherence as an Outcome

With all of this attention now paid to medication nonadherence, it still remains unclear exactly what we are trying to accomplish when we focus on the pattern of a patient’s procurement of medication from a dispensing pharmacy as the measure of adherence. Prescription fill history is a widely available and precise set of data that describes how often a patient receives medication supplies, but is this really an outcome? There is relatively solid evidence to suggest that if patients fill their prescriptions for chronic conditions, such as hypertension and diabetes, more frequently, their respective therapeutic outcomes (blood pressure and glycated hemoglobin) improve.^1,2 However, the fill history in and of itself is not a therapeutic end point. Knowing a patient is not filling one of their medications simply provides a data point indicative of a potential problem—and does nothing to tell us how to solve it.

To use an analogy: are we selling ourselves short by focusing on the cough (the symptom) and not its root cause?

Cause and Effect

There are hundreds of causes of the physiological phenomena we call a “cough.” Examples include postnasal drip, asthma, gastroesophageal reflux disease, infection, blood pressure drugs, chronic bronchitis, bronchiectasis, bronchiolitis, chronic obstructive pulmonary disease, cystic fibrosis, foreign body aspiration, laryngopharyngeal reflux, lung cancer, nonasthmatic eosinophilic bronchitis, and sarcoidosis,³ to name a few. Do we treat all of them with dextromethorphan? Probably not. More importantly, would that treatment (even if it eliminated the cough) lead to the desired therapeutic outcome? Absolutely not.

So, what is the goal of interventions that are designed to improve medication adherence? Ostensibly, the goal is to increase prescription fill frequencies on the assumption that doing so leads to improvements in therapeutic end points and global outcomes, such as reduction in hospitalizations, improved quality of life, and, ultimately, the prevention of premature death. Medication adherence holds great promise to improve the lives of millions of patients, while reducing overall health care costs.^4,5

Yet, although evidence suggests that patients who fill their medications on schedule reach therapeutic goals on average more consistently than patients who do not, we know far less about the effects of increasing fill rates for patients who are not taking their medications on schedule prior to an intervention, particularly for those whose medication regimen is not optimized to reach a therapeutic end point or treatment goal prior to the increase in fill rates. This raises the question, can therapeutic failures be strongly associated with both nonadherence as well as the root cause of the nonadherence? If so, what are the implications?

Causal Pathway

To highlight the importance of a more granular view of the root causes of nonadherence, consider a patient prescribed a dose that is too low to achieve the therapeutic end point. The “causal pathway” to therapeutic failure may look like the following: a suboptimal asthma controller therapy dose results in a lack of symptom improvement, which leads to a patient’s lack of confidence in the medication’s efficacy and a failure to persist in filling medications, with the result being failure to achieve the therapeutic end point or patient goal.

Consider another example in which a patient experiences adverse effects from diabetes medication. This can lead to a lack of desire to take the medication, failure to persist in filling medications, and, ultimately, failure to achieve therapeutic end point or patient goal. Moreover, any increase in fill rates absent a change in regimen or therapeutic approach may lead to a worsening of adverse effects.

In a third example, a patient who feels a stigma associated with antidepressant medication may lose the desire to take the medication, resulting in depression and low self-efficacy. This, in turn, can lead to the patient failing to persist in filling prescriptions for other conditions (eg, hypertension) and a failure to achieve therapeutic goals for those other conditions.

In all 3 of these examples, the patient’s likelihood of achieving the desired therapeutic end point or goal would not necessarily be met by greater fidelity to the medication flagged as having a lower fill rate than prescribed. There are dozens of examples of causal pathways where patient nonadherence is in the middle of the causal pathway, not at the beginning. Without resolution of the root cause of nonadherence, therapeutic goals may not be achieved. It is not simply a matter of “getting patients to take their meds” for many patients who struggle adhering to their current regimen.

Choosing the Right Intervention

The causes of medication nonadherence are heterogeneous. This is relatively well known and well understood. What’s less clear (but perhaps far more important) is which interventions should be paired with which root causes of medication nonadherence and whether that chosen intervention leads to a manner of adherence to a regimen that will actually improve patient outcomes.

This leads to the most controversial question of all: are all increases in medication adherence created equal? Would an intervention designed to send a prescription by mail to a patient with a waived co-pay—regardless of patient need, desire, or actual use— have the same effect as an intervention that assessed patient barriers and beliefs about their medications, offered appropriate resolutions to the barriers, and provided follow-up monitoring? In most cases, the latter intervention is more likely to lead to meaningful, patientcentered outcomes.

This presents policy makers with a dilemma. How do you create broadbased performance measures related to the problem of nonadherence (measuring its severity) to stimulate appropriate interventions (treating its causes) while ensuring that the desired outcome is achieved (solving the problem)?

The answer? Measure the desired outcome— which is achievement of therapeutic goals, not higher fill rates.

I often remind colleagues that my father spends the summer in a remote area of Minnesota, roughly an hour’s drive from the nearest functional emergency department. Over the past year, he has received prescriptions from physicians in northern Minnesota, southern Minnesota, northern Iowa, and south Florida. If today I told him to take everything in his medication cabinet as prescribed, he would be dead by nightfall. He is not atypical as far as Medicare recipients go.

The therapeutic end points that I care about for him? Glycated hemoglobin and blood pressure readings. The outcomes I care about? Avoiding hospitalization and loss of ability to ambulate.

The achievement of optimal adherence will require more than reaching certain fill-rate thresholds. It will require the insight and expertise of health care professionals who can spot and address the true obstacles to adherence along a patient’s pathway to better health. Helping patients to optimize medication use, not just procure their medications, will likely lead to better outcomes.

Troy Trygstad, PharmD, PhD, MBA, is the director of the Network Pharmacist Program and Pharmacy Projects for Community Care of North Carolina (CCNC), a parent organization of 14 regional care management networks. These networks bring together medical practices, county health departments, hospital systems, and mental health providers to integrate care delivery for Medicaid, Medicare, private plans, employers, and the uninsured. CCNC and its networks are responsible for developing and evaluating accountable care systems in North Carolina.Under his direction at CCNC, the Network Pharmacist Program has grown to include pharmacists who are involved in a number of diverse activities ranging from patient-level medication reconciliation to practice-level e-prescribing facilitation to network-level management of pharmacy benefits. Dr. Trygstad also plays an integral role in health information technology adoption and proliferation with CCNC practices and across the state, leading e-prescribing adoption efforts as well as the development and deployment of a statewide medication management platform.He has been involved in novel adherence implementations as well as the development of adherence technologies that use administrative claims data to predict, intervene, and triage adherence interventions and coaching opportunities. Dr. Trygstad received his PharmD and MBA degrees from Drake University and a PhD in pharmaceutical outcomes and policy from the University of North Carolina. He is co-editor-in-chief of the Pharmacy Times series Directions in Pharmacy^®.

References

• Cranor CW, Christensen DB. The Asheville Project: Short-term outcomes of a community pharmacy diabetes care program. J Am Pharm Assoc (2003). 2012;52(6):838-50. doi: 10.1331/JAPhA.2012.12542.
• Choudhry NK, Glynn RJ, Avorn J, et al. Untangling the relationship between medication adherence and post-myocardial infarction outcomes: medication adherence and clinical outcomes. Am Heart J. 2014;167(1):51-8. doi: 10.1016/j.ahj.2013.09.014.
• Mayo Foundation for Medical Education and Research. Diseases and conditions: chronic cough. Mayo Clinic web site. www.mayoclinic.org/diseases-conditions/chronic-cough/basics/causes/con-20030883.
• Stuart B, Loh FE, Roberto P, Miller LM. Increasing Medicare Part D enrollment in medication therapy management could improve health and lower costs. Health Aff (Millwood). 2013;32(7):1212-20.
• Congressional Budget Office. Offsetting effects of prescription drug use on Medicare’s spending for medical services. November 2012. www.cbo.gov/sites/default/files/cbofiles/attachments/43741-MedicalOffsets-11-29-12.pdf. Accessed June 6, 2014.