Potential drug interactions in patients with a history of cancer.

BACKGROUND

Cancer survivors (css) are frequently exposed to polypharmacy, which might increase their risk of drug interactions. Our study aimed to determine the relative prevalence of potential drug interactions (pdis) among css compared with non-cancer respondents (ncrs).

METHODS

Self-reported prescription data from 4975 patients were extracted from the U.S. National Health and Nutrition Examination Survey and screened for pdis using iFacts: Drug Interaction Facts (Facts and Comparisons, St. Louis, MO, U.S.A.). The clinical significance of each pdi was graded on a 5-point scale based on the severity of the interaction and the level of evidence documenting the interaction. Summary statistics and logistic regression models were used to assess the impact of cancer history on the risk of pdis.

RESULTS

Of patients eligible for the analyses, the css (n = 302) indicated using 4.4 ± 0.22 prescriptions on average, and the ncrs (n = 908), 3.8 ± 0.09. Nearly half of both cohorts (40% of css, 43% of ncrs) had at least 1 pdi. In both cohorts, 12% were at risk for fatal or permanently debilitating effects. In multivariate analyses, css were significantly less likely than ncrs to be at risk for any pdis (odds ratio: 0.65; 95% confidence interval: 0.46 to 0.92; p = 0.02). Advanced age and low household income were associated with pdis among css. Medications most commonly prescribed to css with a pdi included metoprolol (15.6%), levothyroxine (13.6%), and furosemide (11.9%).

CONCLUSIONS

Although css appear to be less susceptible than ncrs to pdis, the prevalence of pdis among css remains suboptimal. Specific subgroups of css may be particularly prone to pdis, underscoring the importance of increased vigilance.