TOPLINE:

Researchers developed a scoring tool, known as the Pregnancy Risk Evaluation of Adverse Drug Reactions (PREG‑ADR), to predict which pregnant women were at risk of developing adverse drug reactions (ADRs). The tool incorporated eight independent risk factors, including gestational hypertensive syndrome, self‑reported drug allergy, and polypharmacy.

METHODOLOGY:

• Researchers conducted a case-control study to develop and validate the PREG-ADR tool, involving 330 pregnant women admitted to the high-risk maternity ward who received at least one medication during a hospital stay lasting more than 24 hours.
• Participants included those who developed ADRs (n = 110; mean age, 29.8 years) and matched control individuals who did not (n = 220; mean age, 30 years).
• Researchers assessed how well predicted risks matched observed outcomes using the Hosmer-Lemeshow test (P > .05 suggested a good fit) and measured overall predictive performance using the Brier score (0 indicated perfect and 1 indicated poor performance).
• They also assessed how well the tool distinguished high-risk from low-risk pregnant women using the area under the receiver operating characteristic curve (0.5 indicated no discrimination).
• An ADR was defined as any harmful, unintended response to a medication at doses normally used in humans for the prevention, diagnosis, and treatment of a disease or for alteration of a physiologic function.

TAKEAWAY:

• The PREG-ADR tool demonstrated good calibration (P = .528) and acceptable predictive performance (Brier score, 0.27), with an area under the receiver operating characteristic curve of 0.77 (95% CI, 0.71-0.82).
• Of all the 63 variables tested, only eight showed significant associations with ADRs: gestational hypertensive syndrome, self-reported drug allergy, blurred vision, excessive sedation, nausea, activated partial thromboplastin time of 27.9 or more seconds, polypharmacy, and hospitalization lasting 7 or more days (P < .05 for all).
• Using a threshold of 34 or more points, the tool achieved a sensitivity of 58.9% and a specificity of 79.8%.

IN PRACTICE:

“[The] PREG-ADR is the first validated predictive tool specifically developed for hospitalized high-risk pregnant women, demonstrating both clinical utility and strong discriminative performance,” the authors of the study wrote. “It enables the early identification of patients at increased risk for ADRs during hospitalization, thereby supporting personalized therapeutic management and enhancing pharmacotherapy safety,” they added.

SOURCE:

Rand Randall Martins, PharmD, Federal University of Rio Grande do Norte, Natal, Brazil, was the corresponding author of the study, which was published online on November 14, 2025, in Expert Review of Clinical Pharmacology.

LIMITATIONS:

The single-center design and use of institution-specific clinical and laboratory reference values may have limited the study’s generalizability. Control individuals were matched only by maternal age and length of hospital stay and not by factors such as gestational age or diagnosis at admission, which may have reduced matching precision. Also, restricting ADRs to cases rated probable or definite by the Naranjo algorithm may have underestimated the incidence of ADRs and reduced the sensitivity of the tool in complex cases.

DISCLOSURES:

The study was funded by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Finance Code 001). The authors declared having no conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.