Enhancing Patients’ Treatment Adherence
When patients are not improving as expected with treatment, the possibility of poor adherence to the treatment should be considered. How can we tell if patients are poorly adherent?
First, we could ask patients how often they are taking their medications. A straightforward query runs the risk of making the patient defensive. To mitigate against this, we might ask in such a way that empower a patient to report non-adherence. Instead of asking a patient, “Are you taking the medication daily as recommended?” we might ask, “Do you take the medicine at least once or twice each week?” In response to such a question, they might feel very good about themselves and report accurately if they were taking the drug every other day.
Examination of patients’ pill bottles can help shed light on adherence. If the medication is to be taken twice a day for one month, if 60 pills were prescribed, and if the patient returns for the one month return visit with a bottle still nearly full of medication, poor adherence is obvious. If the pill bottle is empty, adherence may be very good; alternatively, the patient may have dumped the medication in the toilet and be very non-adherent. That’s why I like to prescribe 70 pills when prescribing twice daily therapy for one month; if the bottle comes back empty instead of having 10 pills left, it strongly suggests medication dumping.
Another approach for asking about adherence is by asking (or appearing to ask) about something else. I frequently prescribe home injection treatment for patients with severe psoriasis. On their return, I ask patients, “Are you keeping the extra syringes of medication you’ve accumulated refrigerated like you are supposed to?” The correct answer from an adherent patient is, “Doc, I don’t know what you are talking about. I haven’t accumulated any extra syringes.” If the patient reports that he or she is refrigerating the extra medicine, I know that adherence is not what it is supposed to be (and I don’t presume that the extras are being refrigerated).
Finally, behavioral economists have found that people tend to cheat less when they are primed to think about religious imagery. Studies of college students found that the students did not cheat on math tests after being asked to name the Ten Commandments. This effect did not depend on the student being religious or the number of commandments that they could name. I sometimes ask patients about how well they are taking their medication after asking social history items about their church attendance. I also have a Ten Commandments tie I can use as a prop.
There are some patients who are absolutely convincing in their prostrations about their excellent adherence to treatment. When treatment is not working in these patients as it is expected to, I think it is best to trust patients yet still consider the possibility of poor use of the medication. The Cold War “Trust, but Verify” advice may be apropos.
–Steven R Feldman MD PhD
