Adverse drug reactions, defined by the World Health Organization as any unintended and harmful response to a medication administered at normal therapeutic doses, are a recognized burden on health systems worldwide. They are estimated to account for roughly 5 to 10 percent of hospital admissions and contribute to prolonged hospital stays, generating substantial clinical and economic costs. But pregnant and postpartum women occupy a uniquely vulnerable position. The profound physiological transformations of pregnancy, including altered gastrointestinal motility, expanded plasma volume, and shifts in hepatic metabolism and renal clearance, can fundamentally change how drugs are absorbed, distributed, metabolized, and eliminated. These pharmacokinetic and pharmacodynamic shifts mean that a dose that is safe and effective for most adults may behave very differently in a pregnant patient, raising the risk of harm to both mother and fetus. At the same time, ethical constraints have long limited the inclusion of pregnant women in clinical trials, leaving clinicians with a paucity of evidence-based guidance for medication use in this population.

The new study, conducted between February and May 2025, sought to fill a critical evidence gap. While reported prevalence rates for adverse drug reactions among obstetric and gynecologic patients globally range from 10 to 30 percent, most data come from high-income countries with robust pharmacovigilance systems. In Ethiopia and similar resource-constrained settings, the true burden has likely been underestimated due to weak monitoring systems, limited training among healthcare workers, medication shortages, and suboptimal documentation practices. The research team at Debre Tabor Comprehensive Specialized Hospital, a facility serving approximately 3.5 million people in the country’s northwest, designed their investigation to generate context-specific data that could directly inform clinical practice and health policy.

The methodology was rigorous and multi-layered. The researchers enrolled 354 women aged 18 or older who were hospitalized for obstetric care, either during pregnancy or within six weeks of delivery, using a single population proportion formula to determine sample size and consecutive sampling to recruit participants. Data were gathered through systematic daily chart reviews, structured interviews with patients and their families, and consultations with healthcare professionals. To strengthen detection, the team also employed the medication module of the Institute for Healthcare Improvement’s Global Trigger Tool, which uses “triggers,” or clues in the medical record, to flag potential adverse events that might otherwise go unnoticed. Each suspected reaction was then confirmed using the Naranjo Adverse Drug Reaction Probability Scale, a ten-question instrument that assigns point values to establish causal relationships between a clinical event and a drug. Scores of 9 or higher indicate a definite reaction, 5 to 8 a probable one, and 1 to 4 a possible one.

Severity was assessed using the modified Hartwig and Siegel scale, which grades reactions from mild, involving levels 1 and 2 on the scale, through moderate, at levels 3 and 4, to severe, at levels 5 through 7. Preventability was evaluated with the Schumock and Thornton criteria, a structured questionnaire that classifies reactions as definitely preventable, probably preventable, or non-preventable based on answers across three sections. To ensure reliability, two trained pharmacists independently reviewed every suspected adverse drug reaction, with discrepancies resolved through consensus alongside a senior clinician. Data were analyzed using SPSS version 27, with bivariate and multivariate logistic regression employed to identify factors independently associated with reaction occurrence.

The results were striking. Of the 354 patients studied, 52 experienced at least one adverse drug reaction, yielding a prevalence of 14.7 percent. The overwhelming majority of these reactions, 88.5 percent, ranged from mild to severe in intensity. Most concerning of all, 82.7 percent were judged to be definitely preventable under the Schumock and Thornton criteria, meaning that straightforward changes in prescribing, monitoring, or documentation could have averted them entirely. In a population where the health of two patients, the mother and the fetus, hangs in the balance, this figure represents an enormous and largely unclaimed opportunity for harm reduction.

When the researchers turned to the question of who was most at risk, three factors emerged as independent predictors. The most powerful was polypharmacy: women receiving four to six medications simultaneously had more than four times the odds of experiencing an adverse drug reaction compared with those on fewer drugs, with an adjusted odds ratio of 4.02 and a 95 percent confidence interval of 1.63 to 9.93. This finding aligns with longstanding pharmacological principles, since each additional medication multiplies the potential for drug-drug interactions and cumulative toxicity, and it carries particular weight in obstetric care, where women may simultaneously receive treatment for pregnancy complications, labor management, and preexisting conditions.

Equally alarming was the role of reproductive history. Women with a history of adverse pregnancy outcomes had 6.28 times the odds of suffering an adverse drug reaction, with a 95 percent confidence interval of 2.64 to 14.95, the strongest association observed in the study. The authors suggest that such patients may present with more complex clinical pictures, requiring more aggressive pharmacotherapy, but the association also hints that women already traumatized by pregnancy loss or complications may face heightened medication risks in subsequent hospitalizations. Meanwhile, patients with underlying medical comorbidities such as hypertension, diabetes, or infectious disease had nearly three times the odds of experiencing a reaction, with an adjusted odds ratio of 2.95 and a confidence interval of 2.23 to 9.76, reflecting the added pharmacological complexity that chronic disease brings to pregnancy care.

The predominance of preventable events is, paradoxically, the study’s most hopeful finding. Unlike many drivers of maternal mortality in low-resource settings, medication-related harm is amenable to targeted, relatively inexpensive interventions. The researchers argue that their results make a compelling case for strengthening pharmacovigilance systems in Ethiopia and similar settings, including better training for healthcare workers in medication safety, improved documentation practices, and systematic medication review to curb unnecessary polypharmacy. Clinical decision support, mandatory reconciliation of medication histories at admission, and closer monitoring of high-risk patients, particularly those with comorbidities or prior adverse pregnancy outcomes, could all meaningfully reduce preventable harm.

As maternal health outcomes increasingly become a global benchmark of health system performance, this study delivers a clear message: the drugs meant to protect mothers are, in a meaningful fraction of cases, harming them, and almost all of that harm is within our power to prevent. For the millions of women who give birth each year in low- and middle-income countries, closing this medication safety gap may be one of the most direct and achievable paths to safer motherhood.