Observational Studies Ask How Much Medication Adherence Matters After MI

BARCELONA, Spain—Two studies presented August 31, 2014 at the annual European Society of Cardiology Congress found discordant results on the importance of medication adherence within 1 year of myocardial infarction (MI). A US study found that adherent patients can reap better long-term outcomes, while another suggested no added risk of 5-year mortality when French patients with preserved left ventricular (LV) function did not take beta-blockers. 

More Adherence Better than Less or None

For the US-based study, Sameer Bansilal, MD, of Mount Sinai Hospital (New York, NY), and colleagues identified 4,015 MI patients in the Aetna Commercial and Medicare Advantage databases between 2010 and 2013 who refilled both an ACE inhibitor and a statin within 6 months of their index event. Exclusion criteria included mental disorders, living in a nursing home, and refilling an ARB prescription post-MI (indicating possible ACE intolerance). Occurrence of MI was documented based on ICD codes along with length of stay more than 2 days.

Patients were classified according to the proportion of days they were taking the statin/ACE inhibitor combo as:

  • Fully adherent (≥ 80% of days; n = 1,721)
  • Partially adherent (40-79% of days; n = 1,031)
  • Non-adherent (< 40% of days; n = 1,263)
Baseline characteristics differed substantially among the 3 groups, so adjustment was made for sociodemographics, copayment and medication use, use of medical services, and comorbidities.  


The rate of MACE (death, hospitalization for MI, stroke, and coronary revascularization) was highest in the non-adherent group (18.1%) compared with the partially (17.2%) and fully (12.8%) adherent groups. Similar patterns were seen for coronary/MI hospitalization (4.8%, 4.4%, and 2.3%, respectively) and revascularization (14.4%, 13.1%, and 10.8%, respectively).

After multivariable adjustment, the fully adherent group had reduced risk of MACE and coronary/MI hospitalization compared with the partially and non-adherent groups; risk was similar between the 2 less-adherent groups. For revascularization, the difference between patients with full and partial adherence did not reach significance (table 1). Degree of adherence did not affect hospitalization related to stroke or to angina/cardiovascular atherosclerosis. 

Table 1. Risk According to Level of Adherence to Statins/ACE Inhibitor


Adjusted RR

P Value


    Full vs No

    Full vs Partial

    Partial vs No









Coronary/MI Hospitalization

    Full vs No

    Full vs Partial

    Partial vs No










    Full vs No

    Full vs Partial

    Partial vs No










Based on these and other findings, methods to maximize adherence are needed, Dr. Bansilal said. “We need to get beyond the 80% threshold,” he stressed, suggesting strategies such as the ‘polypill’ and new drug packaging.

Panelist Christopher Granger, MD, of Duke University (Durham, NC), said, however, that new pills may or may not be needed. “Probably our biggest opportunity to improve cardiovascular health is to simply better apply what we know is effective,” he said. 

Dr. Granger also questioned whether the metric used in the current study, proportion of days covered, is something that clinicians would find user-friendly.

Other than directly observing whether patients are on a given therapy, Dr. Bansilal acknowledged, “most other methods of trying to assess adherence or persistence do fall short, whether it’s pill counts or bottle opening, or blister packs…. I could not agree with you more that health delivery of proven therapies is probably going to be the biggest bang for our buck in the United States and elsewhere over the next 2 decades.”

Panel moderator, Udo Sechtem, MD, of Robert-Bosch Medical Center (Stuttgart, Germany), suggested that such questions also represent “a philosophical issue.”

 For example, he said, German patients are largely suspicious of statins. “I spend hours every week trying to explain [their benefits],” Dr. Sechtem reported. “At the end of the day, it’s the patient’s right to believe what they want to believe after [being informed]…. [T]he question is, do we need to police that patient?”

No Harm from Stopping Beta-Blockers? 

Yet findings from FAST-MI 2005 cast doubt on whether beta-blockers in particular are necessary after MI for patients with preserved LV function and no history of heart failure.

Etienne Puymirat, MD, of Hôpital Européen Georges-Pompidou (Paris, France), and colleagues examined outcomes of 1,630 such patients in the 223-center registry who were on beta-blockers at the time of discharge post-MI in 2005 and had known prescription status at 2 years. In all, 89% had remained on beta-blockers and 11% had not.

Patients taking beta-blockers were more likely to have multivessel CAD and to be on other medications such as clopidogrel, statins, or an ACE inhibitor at 1 year, while those not taking beta-blockers were more likely to have diabetes and PAD.

After propensity score matching (n = 280 beta-blocker and n = 142 non-beta-blocker), survival at 1 year was equivalent regardless of beta-blocker use, with some suggestion of benefit from adherence. By 5 years, no added risk of death was seen for patients who did not take beta-blockers at 1 year (table 2).

Table 2. FAST-MI 2005: Survival With vs Without Beta-Blockers at 1 Year


Adjusted HR

95% CI








Dr. Puymirat said the lack of harm at 5 years lends support to the most recent ESC guidelines, which lessen the emphasis on beta-blockers in STEMI patients.

However, panel members pointed to several study limitations, such as the question of whether patients continued to receive beta-blockers from 1 to 5 years and the possibility of confounding.

1. Bansilal S. Assessing the impact of medication adherence on long-term outcomes post myocardial infarction. Presented at: European Society of Cardiology Congress; August 31, 2014; Barcelona, Spain.

2. Puymirat E. Can beta-blockers be stopped in patients with preserved left ventricular function after acute myocardial infarction? Presented at: European Society of Cardiology Congress; August 31, 2014; Barcelona, Spain.


  • Dr. Bansilal reports no relevant conflicts of interest.
  •  Dr. Puymirat reports receiving fees for lectures and/or consulting from AstraZeneca, Bayer, Eli-Lilly, MSD, and Servier.

Related Stories:

Caitlin E. Cox is News Editor of TCTMD and Associate Director, Editorial Content at the Cardiovascular Research Foundation. She produces the…

Read Full Bio