Why Patients Won't Fill Your Prescriptions

Charles P. Vega, MD


September 02, 2014

In This Article

Review Highlights

The research was completed in Quebec, Canada, where a provincial health insurance agency provides medication coverage for approximately one half of the total population. The study used an electronic prescribing database that incorporated data from approximately 70,000 patients in 131 physician practices. The database required that each prescription be coded with an indication for the medicine.

All included practices were primary care offices. Patients eligible for study inclusion had received at least 1 electronic prescription from their physician between 2006 and 2009.

The main study outcome was primary nonadherence to treatment, which was defined as not filling an incident prescription within 9 months. Researchers queried multiple behaviors involved with failure to adhere to prescribed therapy, including treatment-, patient-, and physician-associated characteristics.

The database yielded a total of 37,506 new incident prescriptions for 15,961 patients. Novel indications accounted for 86.7% of prescriptions, and 13.3% were medication changes within the same class of drugs. The mean age of study subjects was 61.5 years, and 62.3% were female. One quarter of the cohort earned a low income.

Overall, the cohort was fairly unwell. Almost one third of patients had visited an emergency department in the past 6 months, and the mean number of medications taken per patient was 6.25. The overall rate of nonadherence at 9 months was 31.3%.

Diagnoses Associated With Nonadherence

Sorting by indication, the most common conditions associated with nonadherence were headache (51.0% of prescriptions unfilled), ischemic heart disease (51.3%), and depression (36.8%). Conversely, the lowest rates of nonadherence occurred with urinary tract infection (21.0%). Treatment for a new indication was associated with a threefold increase in the risk for nonadherence compared with switching medications within the same therapeutic class.

By drug class, the highest rates of nonadherence were for hormones and synthetic drugs (36.3%). For some reason, thyroid medications went unfilled nearly one half of the time. Cardiovascular medications and prescriptions for ear, nose, and throat conditions were unfilled 34.7% and 34.2% of the time, respectively. Anti-infective treatment had the lowest rates of nonadherence (24.2%).

Patient-Related Factors Associated With Nonadherence

Patient-related factors were far more important in promoting nonadherence compared with physician factors, although nonadherence did significantly cluster in both certain patients and certain physicians. There was a progressive and positive relationship between drug cost and nonadherence, with an 11% increase in the rate of nonadherence between the cheapest and most expensive quartiles of drugs.

Copayment for medications was an even stronger predictor of nonadherence. Patients who paid the maximum copayment for drugs experienced a mean 63% increase in the rate of nonadherence compared with the patient cohort without a copay.

There was an inverse relationship between patient age and nonadherence. Each additional medication received was associated with a 17% average reduction in the rate of nonadherence. More frequent primary care visits were associated with lower rates of nonadherence, and female physicians experienced a mean 17% higher rate of nonadherence compared with male physicians. However, a higher index of chronic illness or more emergency department visits were associated with higher rates of nonadherence.


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.