Medical Examiner

Doctors Without Orders

The mystery of patients who fail to follow prescriptions.

Many patients do not take medicine as prescribed

Earlier this month, scientists at Georgia Tech announced their invention of a necklace that records the date and time at which a person swallows his prescription medicine. The device (which looks more like a dog collar than jewelry) responds to a tiny magnet in the pill as it travels down the esophagus. Other recently developed similar technologies include a drug-filled prosthetic tooth that slowly drips medicine into the mouth and a pill bottle that sends a wireless message to your pharmacist every time it’s opened.

Are we so bad at taking medicine that we need false teeth to do it for us and pill bottles that tattle on us when we don’t? It would seem so. About 50 percent of patients fail to correctly follow prescriptions: We forget to take pills, we alter doses, we take breaks. Nonadherence—the medical term for neglecting to abide by a doctor’s orders—is rampant, resulting in up to one-quarter of all hospital and nursing home admissions. It’s also expensive. The problem persists despite monumental efforts to prevent it. Why? For one thing, it’s impossible to predict which patients are likely to deviate from their orders. And while the problem seems like it should have a simple solution, it doesn’t. Nonadherence, it turns out, is one more reason to heed the call for better American health care.

Blowing off a doctor’s instructions might seem like the act of a basically healthy person. Who hasn’t neglected to take that last antibiotic or exercised less than the doctor said to? But treatment drop-off rates are high among the seriously ill, too. About half the peoplewho undergo kidney transplants do not adequately adhere to the regimen necessary to thwart rejection of their new organ. A 1970s study found that 43 percent of glaucoma patients refused to take the doctor-ordered measures necessary to prevent blindness, even when that refusal had already led to blindness in one eye.

Cost of medication is an obvious consideration. A recent study in Health Affairs reported that 29 percent of patients with chronic kidney failure in the United States did not purchase needed drugs because they were too expensive. It’s understandable: Their co-pay of $114 for a month of medication was the highest of all 12 countries included in the analysis. By contrast, British patients, who had the lowest out-of-pocket costs, were the best at sticking with treatment.

But cost wasn’t the only factor that determined whether patients took their medicine. Swedish patients also had high monthly co-pays, but they were great at following their prescriptions. And good adherence among Japanese patients was only partially explained by low costs. The authors of the Health Affairs study wondered about the influence of cultural factors but also pointed to the health system structure: Since Japanese doctors earn income by selling medicine directly to patients, they have an incentive to make sure prescriptions get filled.

In the United States, where doctors don’t play that middleman role, they have a devil of a time predicting which patients are likely to deviate. Psychiatrist Allan Showalter, who writes about nonadherence at, explains that no matter how long they’ve been practicing, doctors tend to drastically overestimate their patients’ likelihood of sticking to regimens. It’s the medical version of the Lake Wobegon syndrome: Doctors consider their own patients to be above average. One study asked physicians to predict adherence only among patients they knew well, and the doctors still grossly overestimated.

Patients who don’t follow doctors’ orders defy prediction because as a group, they lack defining characteristics. Intelligence, age, gender, and economic background all have no bearing. Even doctors are a mixed bag. In a study of medical students who were prescribed a regimen of Tic Tacs, fewer than half followed instructions fully, even though they knew it was candy and that the exercise was specifically designed to teach them about this issue. Basically, decades of research to identify measures of nonadherence have yielded just one fact: There are no solid measures.

Clearly, some medical disobedience is no different from any other deviant behavior, like speeding or jaywalking. Sometimes the treatment being prescribed is just too much. Few diabetics follow the complete roster of lifestyle recommendations, mainly because doing so is extremely difficult. Other times, patients decide that because of side effects, a treatment just isn’t worth it—many patients with life-threatening diseases would rather live better than live longer. But these explanations only account for a small percentage of nonadherents. Why aren’t the rest of us better at following doctors’ orders?

As it turns out, there is one predictive factor: experience with the health care system. A study of 186 doctors and their patients with diabetes, heart disease, and hypertension found that whether patients got their questions answered correlated strongly with whether they would stick with treatment. Other factors that mattered included the doctor’s level of job satisfaction, how many patients he or she sees per week, and whether patients scheduled a follow-up appointment. As health care consultant Kip Piper explains, the average doctor’s visit clocks in at less than 20 minutes, leaving little time for discussion. And when patients do ask questions, they are usually interrupted within 18 seconds. With little explanation, tricky regimens may not be followed correctly, or a person may take a break from a drug, not understanding the importance of completing the regimen. Many times, patients simply don’t understand the doctor’s orders. Requiring patients to use mail-order companies to order some drugs, as health insurers are increasingly doing, will probably make matters worse.

The legion of gadgets helps some patients but doesn’t make a big enough dent because the contraptions don’t address these underlying issues. Concierge medicine is a more successful fix: For an annual fee, doctors promise to limit the number of patients they treat and provide a higher level of care. But, while effective, this approach is inherently limited. The fee of up to $1,500 is prohibitive for many people, and in any case, there aren’t that many VIP doctors out there. Adherence, then, is unlikely to improve much unless something changes dramatically in the health care system. Or else, we need to invent a drug for nonadherence. If only anyone would take it.