Pharmacoepidemiology Reveals Hidden Safety Signals in New Antibiotics and Vaccines

Pharmacoepidemiology, the discipline that studies the use and effects of drugs at the population scale, has become one of the most powerful lenses for understanding how infectious disease interventions perform outside the controlled boundaries of clinical trials. A new editorial published in Frontiers in Antibiotics by Cesar Henriquez-Camacho of Hospital Universitario de Mostoles in Madrid brings together six original studies that span the full breadth of the field, from large-scale pharmacovigilance of newly approved antibiotics to early warning outbreak surveillance, resistance mutation tracking in HIV, adjunctive therapies for chronic rhinosinusitis, and the long-term effectiveness of pertussis vaccination. Taken together, the collection illustrates how real-world data can expose safety signals, resistance patterns, and epidemiological trends that randomized trials were never designed to capture.

The core rationale behind post-marketing safety evaluation lies in a fundamental statistical limitation of randomized controlled trials. Because trials typically enroll hundreds to a few thousand participants over limited follow-up periods, they are powered to detect common adverse events but systematically exclude rare or long-term harms that may only emerge once a drug reaches tens of thousands of patients. Two of the studies in the Research Topic addressed this gap by mining the FDA Adverse Event Reporting System, or FAERS, a database containing millions of spontaneous reports submitted by healthcare professionals, manufacturers, and patients. Spontaneous reporting databases are noisy by nature, but when analyzed with disproportionality methods, which compare the observed frequency of a drug-event pair against its expected background frequency, they can reveal genuine safety signals hidden in the statistical structure of the reports.

The first of these pharmacovigilance analyses, led by Xie and colleagues, focused on oritavancin, a long-acting lipoglycopeptide antibiotic used against serious gram-positive infections including methicillin-resistant Staphylococcus aureus. The study corroborated known safety signals such as chills and infusion-related reactions, but it also identified multiple underemphasized or entirely unlabeled signals, including dyspnea, chest discomfort, and anaphylactic reactions. Particularly striking was the temporal pattern of the reports: the majority of adverse events occurred within the first few days of treatment. This finding underscores the need for intensified vigilance during the earliest phase of therapy, a window that clinicians may overlook when a drug has already passed regulatory review and entered routine use.

A second FAERS-based study by Li and colleagues applied disproportionality and Bayesian analysis to a newer class of tetracycline derivatives, examining coagulation dysfunction associated with tigecycline, eravacycline, and omadacycline. The analysis identified reporting signals for tigecycline and eravacycline, including hypofibrinogenemia and prolonged prothrombin time, which emerged as the predominant coagulation abnormalities across the three antibiotics. Omadacycline, by contrast, showed a generally more favorable coagulation profile, although discrete positive signals for hypofibrinogenemia appeared in subgroup analyses. The distinction matters clinically because coagulation abnormalities can escalate into serious bleeding events, and recognizing which members of a drug class carry the highest risk allows prescribers to weigh alternatives more precisely, particularly in patients already vulnerable to bleeding complications.

Beyond drug safety, the Research Topic also explored how pharmaceutical sales data can function as an early warning system for respiratory infectious disease outbreaks. Li and colleagues investigated the effectiveness of over-the-counter pharmaceutical sales as a timely indicator for predicting respiratory infection trends in Pinghu, China. The underlying logic is that people typically self-medicate at the first appearance of symptoms, well before they seek hospital care or receive a confirmed diagnosis. By monitoring sales of cough suppressants, antipyretics, and other over-the-counter remedies, public health authorities can detect potential outbreaks before confirmed hospital data become available, enabling better medical resource allocation and more rapid epidemic responses. The authors cautioned, however, that such sales signals should be integrated with other data sources, including school absence records and online search queries, to improve early monitoring and overall surveillance accuracy, since sales patterns alone can be distorted by seasonal purchasing behavior and marketing effects.

Drug resistance surveillance formed another pillar of the collection. Zhang and colleagues investigated resistance mutations in people living with HIV experiencing low-level viremia in Chongqing, China, a clinical scenario in which standard genotyping often fails because viral loads are too low for conventional RNA-based assays. By employing a combined strategy that genotyped both plasma RNA and proviral DNA integrated into the host genome, the team detected resistance mutations in 26.32 percent of the cohort, predominantly against non-nucleoside reverse transcriptase inhibitors. Low baseline CD4-positive T-cell counts below 100 cells per microliter were significantly associated with the presence of resistance. The combined genotyping strategy enhanced resistance detection even when viral load was minimal, demonstrating its value for resistance surveillance and for optimizing clinical management in patients whose treatment failure would otherwise go undetected until viremia rebounded to higher levels.

Adjunctive disease management was the focus of a meta-analysis by Nie and colleagues, who evaluated the role of supplementary therapies for recurrent rhinosinusitis, a condition in which standard treatment alone often fails to prevent repeated episodes. The analysis examined nasal irrigation, steroid-eluting implants, and concurrent dental management, and found that these modalities produced a 30 to 50 percent improvement in symptom scores measured with validated instruments such as the Sino-Nasal Outcome Test, alongside reduced recurrence rates compared with conventional treatment alone. The findings support a multimodal, patient-centered approach to disease management, in which mechanical, pharmacological, and dental interventions work in concert rather than in isolation, reflecting a broader trend in infectious disease care toward combination strategies that address the anatomical and behavioral factors driving chronic recurrence.

The long-term impact of immunization programs rounded out the Research Topic. Liu and colleagues analyzed the epidemiological characteristics and vaccine effectiveness of pertussis vaccination in Linping District, China. The diphtheria-tetanus-acellular pertussis vaccine series provided high protection, exceeding 90 percent effectiveness for children through the preschool age. However, the study identified a distinct spring peak in incidence, and delayed or missed vaccination emerged as a key risk factor for infant cases. The findings suggest that waning immunity after the primary series may necessitate evaluation of booster doses at school age, a consideration that aligns with international experience showing that acellular pertussis vaccines induce strong but non-durable protection whose decay can leave adolescents and adults susceptible to infection and capable of transmitting the bacterium to unvaccinated infants.

Henriquez-Camacho concludes that pharmacoepidemiology provides indispensable insights into the real-world use and impact of infectious disease interventions, thereby ensuring safe, effective, and equitable healthcare strategies. The editorial emphasizes that the discipline’s integration with emerging technologies and global data networks will further enhance preparedness for future health crises. As the six studies demonstrate, whether the question is an unlabeled anaphylaxis signal for a newly approved antibiotic, a coagulation risk hidden within a drug class, an outbreak brewing behind rising pharmacy sales, a resistance mutation at undetectable viral load, or the slow decay of vaccine-induced immunity, population-scale data analysis consistently reveals what trials cannot. In an era of rising antimicrobial resistance and recurring epidemic threats, that capacity to see the full picture of drug use and disease control at the population level has never been more essential.

Subject of Research: Pharmacoepidemiology in infectious diseases

Article Title: Editorial: Pharmacoepidemiology in infectious diseases

Article References: Henriquez-Camacho, C. (2026). Editorial: Pharmacoepidemiology in infectious diseases. Frontiers in Antibiotics, 5, Article 1965761. https://doi.org/10.3389/frabi.2026.1965761

Image Credits: AI Generated

DOI: 10.3389/frabi.2026.1965761

Keywords: pharmacoepidemiology, infectious diseases, pharmacovigilance, FAERS, antibiotics, drug resistance, vaccine effectiveness, public health, Editorial, infectious, diseases, scientific research