Identifying medications underlying communication atypicalities in psychotic and affective disorders: A pharmacovigilance study within the FDA Adverse Event Reporting System

Abstract

Background: Disruptions in language and speech are considered promising markers of affective and psychotic disorders. However, little is known about the mechanisms and confounders underlying such communicative atypicalities. Medications might have a crucial, relatively unknown role both as potential confounders and relatedly offering an insight about the mechanisms at work. The integration of regulatory documents with pharmacovigilance techniques could provide a more comprehensive picture to account for in future investigations of communication-related biomarkers. Objectives: Our aim was to identify a list of drugs potentially associated with speech and language atypicalities within psychotic and affective disorders. Methods: To structure a search for potential drug-induced communicative atypicalities on the FDA Adverse Event Reporting System (FAERS, updated June 2021), we developed a query using the Medical Dictionary for Regulatory Activities (MedDRA). We performed a Bonferroni corrected disproportionality analysis (Reporting Odds Ratio) on three separate populations: psychotic, affective, and non-neuropsychiatric disorders, to account for the confounding role of different underlying conditions. Unexpected drug adverse event associations, which were not already reported in the SIDER database of labeled adverse drug reactions, were subjected to further robustness analyzes to account for expected biases. Results: We identified a list of 291 expected and 91 unexpected potential confounding medications. We corroborated known/suspected associations: e.g., corticosteroids-related dysphonia and immunosuppressant-related stuttering. We also identified novel signals: e.g., domperidone-associated aphasia or VEGFR inhibitors-related dysphonia. Conclusions: We provide a list of medications to account for in future studies of communication-related biomarkers in affective and psychotic disorders. The methodological tools here implemented for large scale pharmacosurveillance investigations will facilitate future investigations of communication-related biomarkers in other conditions and provide a case study in more rigorous procedures for digital phenotyping in general. Objectives: Our aim was to identify a list of drugs potentially associated with speech and language atypicalities within psychotic and affective disorders, to account for in future investigations of communicative markers and to provide tools for similar future endeavors. Methods: We identified terms from the Medical Dictionary for Regulatory Activities (MedDRA) related to speech and language adverse drug reactions and clustered them by partial semantic overlap to structure a search on the FDA Adverse Event Reporting System (FAERS, updated June 2021). A Bonferroni corrected disproportionality analysis was applied to three separate populations in the FAERS: psychotic, affective, and non-neuropsychiatric disorders, to account for the confounding role of different underlying conditions. Unexpected drug adverse event associations, which were not already reported in the SIDER database of labeled adverse drug reactions, were subjected to further robustness analyzes to account for expected biases. Results: We identified a list of 291 expected and 91 unexpected potential confounding medications. We also developed methodological tools for large-scale pharmacosurveillance investigations: a MedDRA query proposal for speech and language impairments, formalization of possible biases, and related analyzes to account for them. Conclusions: We provide a list of medications to account for in future studies of communicative behavioral biomarkers in affective and psychotic disorders. The developed methodological tools will facilitate future investigations of communicative biomarkers in other conditions and provide a case study in more rigorous procedures for digital phenotyping in general.