Public Perspectives of Using Social Media Data to Improve Adverse Drug Reaction Reporting: A Mixed-Methods Study

Abstract

Introduction Information on suspected adverse drug reactions (ADRs) voluntarily submitted by patients can be a valuable source of information for improving drug safety; however, public awareness of reporting mechanisms remains low. Whilst methods to automatically detect ADR mentions from social media posts using text mining techniques have been proposed to improve reporting rates, it is unclear how acceptable these would be to social media users. Objective The objective of this study was to explore public opinion about using automated methods to detect and report mentions of ADRs on social media to enhance pharmacovigilance efforts. Methods Users of the online health discussion forum HealthUnlocked participated in an online survey (N = 1359) about experiences with ADRs, knowledge of pharmacovigilance methods, and opinions about using automated data mining methods to detect and report ADRs. To further explore responses, five qualitative focus groups were conducted with 20 social media users with long-term health conditions. Results Participant responses indicated a low awareness of pharmacovigilance methods and ADR reporting. They showed a strong willingness to share health-related social media data about ADRs with researchers and regulators, but were cautious about automated text mining methods of detecting and reporting ADRs. Conclusions Social media users value public-facing pharmacovigilance schemes, even if they do not understand the current framework of pharmacovigilance within the UK. Ongoing engagement with users is essential to understand views, share knowledge and respect users' privacy expectations to optimise future ADR reporting from online health communities.

Social Media Role and Its Impact on Public Health: A Narrative Review

“…The included research showed that various social media outlets could raise the degree of women's health promotion [ 25 ], awareness of menstrual hygiene, understanding of breast cancer awareness [ 26 ], breastfeeding techniques [ 27 ], and adherence, self-perception and promotion of oral health [ 28 , 29 ], significant use of antibiotics [ 30 ], consistency with exercise, sexual health promotion [ 31 ], road safety awareness [ 32 ], smoking cessation, adverse drug reaction reporting [ 33 ], and many more in a row.…”

Scientific Opinion on development needs for the allergenicity and protein safety assessment of food and feed products derived from biotechnology

“…The current tools for text mining are imperfect, and many challenges remain for such approaches as identified by the Innovative Medicines Initiative (IMI) WEB‐RADR (Recognising Adverse Drug Reactions) project where a need for coordination to facilitate development, adoption and acceptance of such technology was clearly identified (Vivekanantham et al, 2020 ). Social media networks for health already exist, such as Health Unlocked, 7 but their use needs to be undertaken in a sensitive manner as studies have already shown that consumer trust is greater in activities undertaken for research or by regulatory authorities (Bulcock et al, 2021 ). In addition, some initial activities have been piloted as a reporting system.…”

Sample Bias in Web-Based Patient-Generated Health Data of Dutch Patients With Gastrointestinal Stromal Tumor: Survey Study

“…Third, future work into the sample bias of patient forums for other patient populations is necessary, as this study was limited to a single population in a single country. Nonetheless, our work is a stepping stone toward dissuading the concerns that researchers have expressed regarding the sample bias of social media [6,[16][17][18][19][20][21][22] by unraveling on which characteristics users differ significantly from the overall patient population. Future work could also investigate how compensatory measures can be implemented to statistically correct for sample bias.…”

“…In fact, social media would be able to provide a more patient-centric view of which ADRs are most salient to patients on a day-to-day basis [15]. However, researchers as well as patients have expressed concern about sample bias on social media [6,[16][17][18][19][20][21][22]. Previous research on social media use in general shows that young people, women, and people of a higher socioeconomic class are generally highly represented [23][24][25][26].…”