Spontaneous reporting of adverse drug reactions as an outlet for patient dismay? The case of Levothyrox® change of excipients

Abstract

Following minor changes of excipients of Levothyrox®, the French Pharmacovigilance Database was overwhelmed by patients' spontaneous reports of adverse drug reactions associated with the new formula. After noticing that most of these reports differed from those related to other drugs, we aimed to characterize their features and compared them with spontaneous reports associated with other chronic treatments as comparators. We randomly sampled patient reports associated with either Levothyrox® new formula (n = 200) or comparator drugs (n = 200) from March 2017 till March 2018 from the National Pharmacovigilance Database. We evaluated the number of incriminated drugs and adverse drug reactions per report and verified whether they were “expected” or not according to the Summary of Product Characteristics. Levothyrox®‐associated reports included, on average, more adverse drug reactions (8 ± 4) than comparators (2 ± 2, P < 0.01) and mentioned mostly one drug (98.5% of reports), whereas comparators mentioned two at least (P < 0.001). The quantitative distribution of adverse drug reactions per report differed quite significantly, appearing almost Gaussian for Levothyrox® whereas Poisson‐like for comparators (P < 0.0001). Age did not differ significantly in the two groups (54.2 vs. 49.7, NS), but female predominated in Levothyrox® group (94.5%) as compared with comparators (60.8%, P < 0.001). A mere third of the Levothyrox®‐associated adverse drug reactions were deemed “expected,” versus two thirds for comparators (P < 0.001). The pattern of spontaneous reports associated with Levothyrox®, whether fueled by media or influenced by social networks, appears atypical, as compared with that of comparators. Such reports, by their abundance, may impair the automatic detection of relevant concomitant signals.

An artificial intelligence algorithm for co‐clustering to help in pharmacovigilance before and during the COVID‐19 pandemic

“…Time clusters 4, 5 and 6 spread from July 2017 to June 2018. Time cluster 4 began in July 2017, corresponding to the start of the massive upsurge of declarations arising from the launch of a new levothyroxine formulation 26,27 . Time cluster 5 began in August 2017 and lasted 3 months, coinciding with another surge of declarations related to the same issue of concern, probably triggered by a new press release from the ANSM.…”

“…Time cluster 6 began in November 2017 and lasted 2 months. It coincided with the introduction of generic drugs of the previous levothyroxine formulation, which failed to reassure the population 26,27 . Finally, time cluster 4 reoccurred in January 2018 and lasted 6 months.…”

“…26 Indeed, in the case of levothyroxine, up to 246 different ADRs (almost 42% of all ADRs) were ascribed to a single drug. The atypical nature of many of the corresponding reports 27 evidently reflects a possible patient mistrust.…”