(1020) Effect of the FDA Statement on Low-dose Aspirin Recommendation for Primary Prevention of Cardiovascular Disease Among Older Adults with Diabetes: A Population-based Analysis of NHANES 2011-2018

Objectives: Despite the 2014 FDA statement that “aspirin should not be used to prevent a first heart attack or stroke in patients with no history of cardiovascular disease”, more than half of low-dose aspirin (LDA) users are older adults who take aspirin for primary prevention. Notably, before the FDA statement, the 2013 American Diabetes Association (ADA) had recommended LDA for primary prevention in patients with diabetes who were at an increased risk of cardiovascular disease (CVD). However, in line with the 2014 FDA statement, meta-analyses have demonstrated unfavorable outcomes of LDA use in primary prevention trials. Given the conflicting recommendations by the FDA and ADA, our objective was to determine whether the LDA recommendation for primary prevention among older adults (≥60 years old) with diabetes in the US was affected by the FDA statement.

Methods: Using data from the National Health and Nutrition Examination Survey (NHANES), we conducted cross-sectional analyses from 2011 to 2018. LDA recommendation (yes/no) was determined from participant’s self-reported receipt of physician’s advice for LDA. We included older adults (≥60 years old) who responded to the Preventative Aspirin Use and Medical Conditions Questionnaires. Those who reported borderline, or no diabetes diagnosis were excluded. Multivariable logistic regressions were done to examine the associations between LDA recommendation and history of CVD, adjusted for age, gender, race, high blood pressure and high cholesterol. Subgroup analysis based on age (60-74 and 75-84) was conducted to assess the effect of age on LDA recommendation. A difference-in-difference regression approach was used to compare the change in LDA recommendation between patients with versus without history of CVD. Due to the timing of the FDA statement and two-year cycles of the NHANES data, we defined the pre2014 period as from 2011 to 2012, the post2014 period from 2015 to 2018, and the wash-in period from 2013 to 2014. Descriptive statistics measuring total counts, percentages, and means were calculated for demographic variables. All analyses were done with NHANES survey weights to generalize findings to the US population. Statistical significance was determined at alpha of 0.05.

Results: A total of 2,001 participants (weighted n = 13,498,433) aged 60 or older with diabetes were evaluated (mean [SD] age, 70[0.22] years; 53% men; 65.7% White participants; 74.6% without history of CVD). LDA was recommended to 68.3% of those without a history of CVD, corresponding to 6,879,535 participants. Multivariable logistic regression showed that no previous history of CVD was associated with lower odds of being recommended LDA (OR 0.49, 95%CI 0.35-0.69), whereas high cholesterol was associated with higher odds of receiving LDA recommendation (OR 1.4, 95%CI 1.03-1.91). In a subgroup analysis, patients of the 75-84 age group had higher odds of being recommended LDA than those of the 60-74 age group (OR 1.34, 95%CI 1.01-1.79). In adjusted analyses, there was non-significant interaction between the 2014 FDA warning and the group without history of CVD (OR 0.84, 95%CI 0.35-1.99).

Conclusions: The 2014 FDA statement did not affect the LDA recommendation to older adults with diabetes who had no prior history of CVD. Recently, the U.S. Preventive Services Task Force (USPSTF) has joined the FDA’s stance against the use of LDA for primary prevention of CVD in older adults. With a high prevalence of LDA recommendation for primary prevention, albeit with limited benefits, our study highlights an urgent need for effective strategies to improve guideline implementation. It is important to raise awareness and resolve conflicting clinical recommendations, exemplified by the varying statements from the FDA, USPSTF, and ADA on LDA recommendation for primary prevention. Addressing these discrepancies is critical to support physicians in adopting guideline-concordant practices, and ultimately, improve patient’s clinical outcomes and safety.