Prevalence and predictors of patients reporting adverse drug reactions to health care providers during oral targeted cancer treatment

Abstract

BackgroundPharmacovigilance is a critical component to facilitate clinicians’ decision-making to alter or discontinue therapy. However, self-administration of oral targeted therapy (OTT) requires fewer clinical visits than parenteral infusions, potentially leading to an increase in the under-reporting of adverse drug reactions (ADRs). Objective(s)To identify factors associated with patients reporting ADRs to their health care provider (HCP) and to identify the prevalence of unreported ADRs while on OTT. MethodsPatients aged ≥18 years who received care from a community oncology clinic and newly prescribed an OTT between August 1, 2018, and October 31, 2018, were included. Six-monthly follow-up calls were conducted by the pharmacy staff to assess for gradable ADRs—validated by the NCI Common Terminology Criteria for Adverse Events—and ungradable ADRs. Descriptive analysis was used to analyze the prevalence of unreporting ADRs, and a multivariate logistic regression model was utilized to evaluate predictors of reporting ADRs to an HCP. Predictors included sociodemographic factors, severity of ADRs, insurance type, pharmacy setting, type of OTT, and the number of prescribed medications ResultsOf the 76 patients analyzed, the mean age was 63.32 ±11.55 years, 84.2% were women, 68.8% were non-Hispanic white, and 76.3% had breast cancer. During the follow-up calls, 306 ADRs were identified and 22.2% were not previously reported to an HCP. Of the unreported gradable ADRs, 63.2% were grade 1, 19.3% were grade 2, and 17.5% were grade 3. We found that for every 1-year increase in age, there was a 5% decrease in the likelihood of reporting ADRs (95% CI, 0.91–0.99), and men were 11.4 times more likely to report ADRs (95% CI, 1.29–100.8). ConclusionFollow-up calls served as an outlet to collect pharmacovigilance data by identifying over 20% of unreported ADRs to HCPs, in which over one-third were moderate to severe. However, future studies are needed to further understand the statistically significant differences found in under-reporting for women and the older population.