Almost a Third of Patients Don't Take Drugs as Directed

Richard Mark Kirkner

August 08, 2022

As many as one third of patients do not take their medications as directed, data indicate.

In a study that examined more than 200,000 patients and 91,000 unique prescriptions, overall nonadherence rates ranged from 13.7% for patients prescribed antidepressants to 30.3% for patients prescribed antihypertensive therapies.

"The eye-opening piece of information for me and our research team was how common nonadherence was," lead study author Alexander G. Singer, MB, BAO, BCh, associate professor of family medicine at the University of Manitoba, Winnipeg, Canada, told Medscape Medical News. "As physicians, we assume that people follow the instructions that we give them, and what we have shown is that, many times, they don't. As much as a quarter to a third of the time, they're not filling the prescriptions that we're giving them."

The study was published in the July issue of Canadian Family Physician.

No Predictors Identified

To investigate primary medication nonadherence, the researchers retrospectively examined primary care provider prescriptions that are linked to pharmacy-based dispensing data. The researchers included 91,660 prescriptions written from April 1, 2012, to December 31, 2014, in their analysis. The prescriptions were given to a cohort of more than 200,000 patients. The investigators examined the Manitoba Population Research Data Repository at the Manitoba Center for Health Policy to determine whether the prescriptions had been filled.

The researchers found that for conditions that typically are symptomatic (such as infections, depression, and anxiety), nonadherence ranged from 13.7% to 17.5%. For asymptomatic conditions (such as hypertension, osteoporosis, and diabetes), nonadherence rates ranged from 21.2% to 30.3%.

Lipid-lowering agents for asymptomatic hyperlipidemia and cardiovascular disease were an exception to this general trend. The primary nonadherence rate for these medicines was 15.2%.

One noteworthy aspect of the study was its failure to identify any demographic or clinical factors that predicted primary medication nonadherence, said Singer. "Our findings suggest that providers may not be able to use particular attributes to predict which patients will or will not fill a new prescription, adding to the notion that primary medication nonadherence is complex and often influenced by varying and competing factors," he added.

Drug cost may not have contributed to nonadherence. Manitobans with low income are eligible for the provincial Pharmacare program, which provides prescription drug coverage that is based on adjusted family income. "Our Pharmacare program probably blunted any of the major cost impacts, whereas other literature has shown that cost is an indicator," said Singer.

Certain patient and provider characteristics may play roles in nonadherence, but the dataset from administrative and electronic medical records that the investigators used does not account for all such variables, said Singer. "It doesn't have provider-reported outcomes, and there may be patient-experience outcomes that are missing," he added.

The most compelling findings from the study, said Singer, are how common nonadherence is "and how infrequently we ask about it. We assume that people are following our instructions. We're often giving people second and third prescriptions before we even check if they're taking their first one."

An Important Contribution

Commenting on the study for Medscape, Brady Bouchard, MBBS, president of the College of Family Physicians of Canada and a family physician in Saskatchewan, said that its large sample size makes it an important contribution to the literature. "Outside of an intensive research setting, which itself can introduce bias, capturing prescription fill data is likely the most viable method for tracking medication adherence," he said.

The study also confirms how serious and overlooked medication nonadherence is. "Just asking patients about their adherence is likely inaccurate," Bouchard said. "Each patient's recall bias would tend to overestimate adherence as a fact of human nature."

The study highlights the challenge of medication adherence for asymptomatic conditions. "When starting a new, necessary medication for an asymptomatic condition, hypertension being a common one, it is likely very helpful to carefully explain the motivations for treatment and in particular explaining long-term complications, such as heart attack or stroke, that may be avoided," Bouchard said. "Visual aids such as that used by the CVD Risk/Benefit Calculator could likely be helpful."

Although the study does not indicate that drug cost is a barrier to adherence, it "gives us more ammunition in important advocacy work in trying to convince the federal government to implement and fund a national Pharmacare for all," said Bouchard. "It would be very interesting to look at a similar dataset but one that's segmented based on the availability of public or private drug benefit plans. I suspect they would see a significant difference in rates of medication adherence."

The study was independently supported. Singer received a grant from the Canadian Institute for Military and Veteran Health Research with funding and in-kind support from IBM. Bouchard has disclosed no relevant financial relationships.

Can Fam Physician. 2022;68:520-527. Full text

For more news, follow Medscape on Facebook, Twitter, Instagram, and YouTube.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.