Association Between FDA Drug Recall Classification and Product-Related Characteristics
摘要
This study examined associations between FDA drug recall classification (Class I–III) and selected product-related characteristics using publicly available recall data.
MethodsA secondary analysis of 13,248 FDA drug recalls issued from 2015 to 2024 was conducted. Data on recall classification, recall cause, dosage form, therapeutic category, and geographic origin were extracted from publicly available FDA databases. Descriptive statistics were used to summarize recall distributions. Pearson’s chi-square tests assessed associations between recall classification and categorical variables, and Cramér’s V was used to quantify the magnitude of association.
ResultsSignificant associations were identified between recall classification and recall cause (χ2 = 8,003.44; p < 0.001), therapeutic category (χ2 = 3,973.44; p < 0.001), dosage form (χ2 = 3,040.18; p < 0.001), and geographic origin (χ2 = 823.82; p < 0.001). Recall cause exhibited the largest effect size (Cramér’s V = 0.550), followed by therapeutic category (0.387), dosage form (0.339), and geographic origin (0.176). Microbial contamination, adulteration, and unapproved drug status accounted for a substantial proportion of Class I recalls. Injectables and oral liquid dosage forms, as well as homeopathic products, were disproportionately represented among severe recall classifications.
ConclusionRecall classification was significantly associated with multiple product-related characteristics, particularly recall cause. While these findings reflect statistical associations rather than causal relationships, they identify recurring patterns within FDA recall data that may inform risk-based regulatory surveillance and pharmaceutical quality management activities. Because exposure-adjusted denominator data were unavailable, the findings should be interpreted as patterns within recalled products rather than measures of comparative product risk.