Background: When patients do not take medications as prescribed, adverse events often occur. Patients may misunderstand medication directions because the physician fails to give adequate oral or written instructions or because of variations in patient literacy skills. The increasing number of prescription medications also may make errors more likely. Although brief instructions on prescription drug labels should make it easier for patients to take medications properly, it is uncertain if these instructions help patients with low literacy levels. Davis and colleagues examined the relationship between literacy levels and a patient's understanding of common prescription drug labels.

The Study: The cross-sectional study was conducted in three outpatient primary care clinics in medically underserved communities. In the study, 395 English-speaking patients 19 to 85 years of age were recruited from the clinics' waiting rooms in July 2003 and July 2004. Patients were excluded if they had hearing difficulties, severe visual impairment, or severe illness.

In a structured interview, research assistants collected information on patient demographics, socioeconomic status, and current medication use. Patients were then shown labeled prescription bottles of amoxicillin, trimethoprim (Proloprim), guaifenesin (Humibid), felodipine (Plendil), and furosemide (Lasix) and asked to describe how they would take each medication. Those who correctly stated the instructions for guaifenesin were asked to count out the number of pills to be taken in a day. The interview concluded with a brief standardized health literacy assessment to determine the patient's literacy level: low (sixth grade and lower), marginal (seventh to eighth grade), or adequate (ninth grade and higher).

Three general internal medicine attending physicians who were blinded to patient information rated responses as correct or incorrect. A correct response required an accurate description of dosage, timing, and duration of treatment. The raters disagreed on 147 responses. These responses were reviewed independently by three primary care physicians and two behavioral scientists and rated by a majority vote. Research assistants coded incorrect responses according to the type of misunderstanding. A statistical analysis controlling for age, sex, race, education, and number of daily medications examined the relationship between literacy level and understanding of drug labels.

Results: Out of 1,975 responses, 374 were rated as incorrect. Most incorrect responses were dosage errors (51.8 percent) and dose frequency errors (28.2 percent). Fewer patients were unable to find the instructions on the label (5.8 percent) or admitted to being unable to read (3.2 percent). Incorrect responses were associated, to a statistically significant degree, with low literacy levels (relative risk [RR] = 2.32; 95% confidence interval [CI], 1.26 to 4.28) and marginal lit-eracy levels (RR = 1.94; 95% CI, 1.14 to 3.27). The risk of an incorrect response increased with more medications used. Compared with patients who had adequate literacy levels, patients with low and marginal literacy levels who were able to read instructions were less likely to count out the correct number of daily guaifenesin pills.

Conclusion: The authors conclude that patients commonly misunderstand the instructions on prescription drug labels, and patients with low literacy are at the highest risk. They also conclude that the ability to read the label instructions does not necessarily correlate with being able to take medications properly. Further research is needed to evaluate ways to simplify the instructions and improve patient understanding of prescription drug labels.