Medication adherence and the psychology behind it

By Naveed Saleh, MD, MS
Published November 5, 2021

Key Takeaways

As a prescriber, here’s an interesting statistic for you: In the United States, 3.8 billion prescriptions are written every year. Guess how many are never filled? One of five, according to the CDC. Moreover, 50% of those that are filled are taken incorrectly, with errors in timing, dosage, frequency, and duration.

The costs of nonadherence are gargantuan on two fronts. Nonadherence is correlated with higher rates of hospital admissions, suboptimal health outcomes, and higher morbidity/mortality for chronic conditions. On the other front, direct healthcare costs linked to nonadherence account for $100 billion-$300 billion of US healthcare dollars spent annually, even though nonadherence rates have held steady.

Considering these factors, it’s important that physicians do all they can to promote medication adherence. But, to help your patients adhere to their prescribed medication regimens, it helps to understand the psychology and barriers that contribute to the phenomenon. Let’s dive a little deeper.

ABC taxonomy 

Authors of a review article in American Psychologist described the ABC taxonomy of medication adherence as a time-varying, three-pronged process including initiation, implementation, and discontinuation.

The first phase is initiation and signifies the time from prescription until the patient takes the first dose of the medication. The second phase, or implementation, is the period from initiation until the last dose is taken. The third phase is discontinuation and demarks the end of therapy, when the patient quits the last dose and no more doses are subsequently taken. Of note, the term persistence refers to the duration of time a patient is in the implementation phase. 

“A failure to identify barriers to adherence, adapt an intervention, and identify the self-regulatory processes involved in these phrases is likely to result in poor long-term adherence,” wrote the authors.

Most medication-adherence studies have investigated the investigation phase, with examination of a current—instead of new—prescriptions. In these studies, outcomes entail the proportion of drug taken, measured at various time points. Less research effort has focused on the initiation or re-initiation of a medication wherein a patient discontinues a drug or a provider prescribes a new medication in lieu of, or in addition to, an existing regimen. Research strategies should be focused on which stage of adherence is examined, according to the authors.


According to the aforementioned review, barriers can be broken down into four categories: patient-related, medication-related, clinician-related, and health system issues.

  • Patient-related barriers include forgetfulness, symptom self-management, and schedule disruption (eg, vacations, holidays).

  • Medication-related barriers include frequency of dosing, adverse effects of the medication, polypharmacy, or length of stay.

  • Clinician-related barriers are factors like lack of trust in the physician, poor communication with the doctor, and hurried office visits.

  • Health-system barriers include lack of/inadequate insurance or drug coverage, trouble making appointments, lack of ongoing dialogue, and lack of feedback.


Strategies to improve rates of adherence typically include multiple efforts spanning the breadth of care. The CDC outlines various approaches including the following:

  • Using pillboxes or blister packs, which is cost-effective

  • Team-based or coordinated care, such as counseling involving the pharmacist and physician

  • Decreasing cost, such as eliminating copays

  • Identifying and re-engaging nonadherent patients in treatment by linking them via the health department and their care providers (eg, HIV care)

  • Electronic prescribing, which can keep track of whether a prescription was filled or dispensed, as well as send the patient reminders

Artificial intelligence (AI) is also playing an important role in boosting rates of medication adherence. For instance, authors of a review article published in Frontiers in Digital Medicine wrote that automated systems have been employed to provide patient reminders to take their medication. They also cited research that showed text messages reminding patients with non-communicable diseases to refill their medication significantly improved rates vs those of a control group.

Lastly, in patients with Crohn disease, AI has been used by patients who record their medication adherence, activity, and symptoms on a daily or hourly basis. The data were then provided to providers to enhance decision-making.

“Patients reported that the AI components of the system enabled them to track, understand, and monitor their ODLs [observations of daily living] more easily and accurately while also helping them to remember to take their medications,” stated the authors.

Share with emailShare to FacebookShare to LinkedInShare to Twitter