The Challenge of Medication Adherence: A Physician’s Perspective

As a primary care physician, I have a number of tools at my disposal. Coaching around lifestyle management and helping rectify social barriers to achieving optimal health are always my first choice. Sometimes referrals are needed for surgical or other types of interventions and often to my mental health colleagues. But in many cases I must prescribe medications.

Medications can be miraculous. When used appropriately they reduce morbidity, mortality, medical complication, and suffering, and they can improve quality of life.
But all this comes at a high financial price – a price that is rising inexorably.
Frequently, a decision to start a medicine is not taken lightly. On the medical side I need to take into consideration the patient’s other medical conditions, their other medications, their drug allergies and intolerances, and their specific ability to metabolize these drugs (such as their liver and kidney function), as well as the specific side effect profile of the drug–both short- and long-term. Financially, I need to determine whether their drug plan covers the medication and how costly it will be to the patient. And of course I must take into consideration my patient’s preferences in all of this.
After I prescribe a drug, I need to assess its impact. Depending on the situation, I may do this by having the patient contact me (via phone or e-messaging through our practice patient portal), by my staff reaching out to them, or by seeing them back in my office.
But in order to assess its impact, a number of things need to occur:

1. The prescription needs to get to the pharmacy (less of a problem with e-prescribing, in which prescriptions are electronically transmitted to the pharmacy, but still an issue with controlled substances, which in many states need to be paper prescriptions);
2. The patient’s drug plan needs to determine whether and how much they will pay for the prescription, and how much the patient must pay. Sometimes this requires additional information from the prescriber or even a different prescription;
3. The prescription needs to be filled by the pharmacist;
4. The patient needs to go to the pharmacy to pick it up (and this has to occur every time the patient needs a refill); and finally…
5. The patient must take the medication as prescribed.

Most of these transactions are managed by computers, with minimal human intervention. But the last two are related, and both involve patient behavior: the patient must retrieve the medication and the patient must take the medication as prescribed.
These are related, because if a patient decides to take a medication, he or she will need to pick it up. But patients don’t pick up their prescriptions from the pharmacy more than 40% of the time.
Once the patient has the medication many potentially problematic patient behaviors are possible:

1. Not taking the medication some or all of the time
2. Taking an incorrect dose
3. Taking it at incorrect intervals
4. Not refilling when it runs out
5. Not contacting physician for a renewal when the refills have run out

If any of these problematic behaviors occur, then all of the work we put into prescribing the medication is for naught.
Let’s examine a typical scenario. I prescribe a new medication (or make a dosage change to a medication the patient is already taking) and ask the patient to schedule a follow-up so that we can evaluate the patient’s response to the medication, assess for side effects, and determine next steps. If the patient returns for his or her follow-up visit but has not taken the medication as prescribed or has missed some doses, I cannot assess the impact of the change.
What is the impact of this? We wasted the time deciding on the medication change, we wasted time in the follow-up visit (and prevented another patient from seeing me during that time), and if the patient filled the prescription but did not take it, then we’ve also wasted the new medication itself. In addition, this increases the length of time the patient has not been adequately treated for his or her condition, which affects both the patient and me, since I am responsible for clinical outcomes. This is tremendously frustrating.
Unfortunately, this scenario plays out hundreds of thousands of times each day in physicians’ offices in the US alone (and many times this globally).
Even worse, patients who start out taking their medication as prescribed often don’t continue taking it, particularly for asymptomatic conditions (like heart disease, hypertension, and diabetes)–but reducing intervals between visits does improve outcomes, likely because of increased frequency of face-to-face adherence reminders. And adherence behavior often varies over time.
How can we improve this situation? In my next post I will describe approaches that physicians can use to improve medication adherence.