Is social media a reliable source of paediatric drug information?

10 Jun 2025 byStephen Padilla
Is social media a reliable source of paediatric drug information?

Using social media to obtain relevant clinical information is possible, with a recent study suggesting that data from child-rearing public parenting forums (CPPFs) may aid in assessing paediatric adverse drug events and in detecting unknown adverse drug reactions (ADRs) and medication errors.

“Consequently, CPPFs are a good source of complementary pharmacovigilance reporting via a holistic approach for paediatrics,” the investigators said. “Following this holistic approach, ADRs and medication errors of interest can be further researched via artificial intelligence tools.”

This study assessed negative outcomes due to medication (NOMs)–drug related problems (DRPs), ADRs, and medication errors in posts of CPPFs from inception until December 2021.

The investigators classified adverse drug events, examining causality by Liverpool Causality Assessment Tool and seriousness by the World Health Organization criteria. They determined ADR prevalence via summary of product characteristics.

Two CPPFs were accessed to retrieve a total of 3,573 posts. Of these, 906 (25 percent) contained descriptions of medicine, with 823 (91 percent) included in the analysis. [Br J Clin Pharmacol 2025;91:1760-1770]

More than half of the posts (n=425, 2 percent) described 636 NOM-DRPs (1 NOM-DRP median per child). Among these, 161 (26 percent) were ADRs from 105 posts (1.5 ADR median per child) and 95 (15 percent) were medication errors in 64 posts (1 medication error median per child).

Seventy percent of posts retrieved with medicines mentioned included NOM-DRPs, while 18 percent and 10 percent included ADRs and medication errors, respectively. More ADRs were noted among females and infants.

Social media boom

In addition, the majority of ADRs were deemed possible (n=158, 98 percent), while a few were serious (n=17, 11 percent). There were also uncommon (n=19, 12 percent), very rare (n=3, 2 percent), and rare (n=1, 1 percent) ADRs seen.

“These findings are examples of the high potential of social media to detect health data, which cannot be ignored,” the investigators said. [Br J Clin Pharmacol 2015;80:910-920]

“In recent years, the size and growth of data on social media have been unparalleled, and the number of actively involved patients, including parents who share and post health information, has been continuously growing,” they added. [J Biomed Inform 2015;54:202-212; J Med Internet Res 2021;23:e23205; Pediatrics 2022;22:526-539]

Specifically, 11 percent of caregivers and 6 percent of patients shared their experiences or posted questions online.

“In this context, social media is considered a valuable resource for pharmacovigilance, providing clinical insights beyond traditional communication methods,” the investigators said. [Drug Saf 2019;42:1393-1407; Ther Adv Drug Saf 2020;11:2042098620938595]

Premature

Although social media can be helpful, some researchers say that it remains “premature to replace traditional reporting systems.” [Br J Clin Pharmacol 2015;80:910-920; Ther Adv Drug Saf 2020;11:2042098620938595]

For instance, the process of extracting ADRs from social media is tricky, given that data extraction is guided by lexicons. User may misspell medicines or ADRs or write sentences with grammatical errors. [Br J Clin Pharmacol 2015;80:910-920; J Biomed Inform 2015;54:202-212]

In addition, data from social media posts have poor quality and insufficient detail to allow any deep evaluation. Finally, “a high volume of data is needed to obtain a significant amount of relevant information because of the high prevalence of noninformative posts,” according to the investigators. [Ther Adv Drug Saf 2020;11:2042098620938595]