Post-myocardial infarction patients are 30% more likely to discontinue generic cardiovascular medications if their pills change in color or shape, researchers reported.

The medications included beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin II-receptor blockers, and statins.

More than one quarter of patients who started one of these drugs within 90 days of hospital discharge for myocardial infarction had a change in shape or color unrelated to a change in dose in the first year, wrote Aaron Kesselheim, MD, and colleagues at Brigham and Women’s Hospital in Boston, Mass., in a study published in the July 14 Annals of Internal Medicine.

The findings have implications for both individual healthcare providers and for policy.

“Patients know their medications as the little blue pill, or the diamond one,” Niteesh Choudhry, MD, a study co-author and internist at Brigham and Women’s Hospital, told MedPage Today.

But physicians often do not consider that generic drugs change in appearance or that the changes may affect how patients take pills, he added. Physicians and pharmacists can educate patients not to discontinue drugs that are biologically the same.

The findings may also change how the FDA regulates generic medications.

Currently, the FDA does not interfere with the visual appearance of drugs, which are a form of intellectual property. However, the U.S. Supreme Court ruled in 1995 that if the color of a medical pill served a specific function, then competitors were allowed to copy it.

By demonstrating that pill appearance is clinically relevant, the study’s authors hope that the FDA could require new generic applicants to conform pill aesthetics to the brand-name reference drug.

The authors looked at 11,513 patients discharged after hospitalization for myocardial infarction between 2006 and 2011 who initiated treatment with a generic beta-blocker, angiotensin-converting enzyme inhibitor, angiotensin II-receptor blocker, or statin. Researchers followed the patients’ refill habits for one year.

Case patients discontinued a medication for at least one month, and control patients continued the medication. The patients were matched on class of medication, number of dispensings before nonpersistence, sex, and age.

Researchers then evaluated the two refills preceding nonpersistence to see whether pill color or shape had changed. Appearance changes related to variations in dosage were excluded.

More than third of patients discontinued use of a medication (3,666 case patients, 6,519 control patients). Case and control groups had similar rates of coexisting illnesses, medication use before hospitalization, and healthcare utilization.

More than one quarter (28.5%) of patients had a change in pill color or shape unrelated to change in dose.

The odds ratio for nonpersistence after a change in color or shape was 1.49 (CI, 1.30-1.71) after adjusting for age, year, combined comorbidity score, revascularization procedure during the index hospitalization, number of drugs received before the index hospitalization, and prior use of non-index study drugs.

Case patients had a 34% increase in the odds of discordance in pill color preceding an episode of nonpersistence (adjusted OR 1.34, CI 1.12-1.59) and a 66% increase in the odds of nonpersistence after a change in shape (adjusted OR 1.67, CI 1.43-1.94).

Further adjusting the odds ratios for pharmacy changes or use of a mail-order pharmacy did not change the statistically significant associations with nonpersistence except for color alone.

Clinical outcomes were not measured in this study, the authors noted.

The findings are similar to those in a previous study conducted by Kesselheim and colleagues, which showed that changes in pill color among antiepileptic drugs were associated with a nearly 30% increased odds of nonpersistent use of medication.

Post-myocardial infarction patients are a larger chronic disease population than epilepsy patients, which makes the findings more generalizable, Kesselheim told MedPage Today.

Still, the group is somewhat homogenous in age (mean 57.7 years) and gender (72.1% male), and nearly all had commercial insurance (99.6%).

Relevant population demographics such as socioeconomic status and functional or cognitive disability were not included, since patient characteristics were gathered by insurance claims data.

Sounding a note of caution, Eiman Jahangir, MD, a cardiologist at Ochsner Clinical Center in New Orleans, La., told MedPage Today, that the study did not address reasons for medication discontinuation, such as side effects or difficulty obtaining the drug,

The findings underscore the importance of medication reconciliation at every office visit, Jahangir said. Physicians should continue to tell patients to look at the labels rather than relying on a pill’s appearance, he added.