Comparative Pharmacovigilance: Adverse Drug Reactions in Elderly vs. Adults

Thesis Abstract

This study addresses the persistent challenge of differential pharmacovigilance reporting for adverse drug reactions (ADRs) between elderly and adult populations, framing a comparative investigation that informs safer prescribing and monitoring practices in geriatric pharmacotherapy. The aim is to quantify and compare the incidence, severity, and outcomes of ADRs in adults (18–64 years) and the elderly (65 years and above) within a national pharmacovigilance database and corroborate findings with hospital-based spontaneous reporting systems. Specific objectives include (1) estimating ADR incidence rates per 1,000 patient-days for each age group; (2) comparing severity and seriousness classifications of ADRs using the WHO-UMC causality and seriousness criteria; (3) identifying medication classes with the highest differential ADR risk between groups; (4) examining comorbidity and polypharmacy as effect modifiers through stratified analyses; (5) exploring reporting timeliness and outcome resolution across age strata; and (6) evaluating the predictive value of age-related pharmacokinetic and pharmacodynamic factors on ADR risk. The study employs a convergent mixed-methods design, integrating quantitative analysis of large-scale pharmacovigilance datasets with qualitative insights from clinician interviews. The quantitative component analyzes a national ADR dataset comprising 120,000 reports from 2015–2025 and hospital data from three tertiary centers, yielding a sample of 85,000 adult- and 40,000 elderly-reported ADR cases after de-duplication. Data collection instruments include standardized ADR reporting forms and hospital medical records abstraction tools, supplemented by validated comorbidity and polypharmacy indices (Charlson Comorbidity Index, Medication Regimen Complexity Index). The qualitative component engages semi-structured interviews with 30 clinicians and 20 clinical pharmacists to elucidate barriers to ADR reporting, perceived age-related risk factors, and clinical decision-making processes. Validity and reliability are ensured through double data entry, inter-rater reliability checks (Cohen’s kappa >0.80 for causality and seriousness assessments), and triangulation between datasets. Quantitative analysis employs descriptive statistics to characterize ADR profiles by age group, followed by Poisson regression to compare incidence rates, adjusting for sex, polypharmacy, comorbidity burden, and healthcare setting. Multivariable logistic regression analyzes risk factors for severe/serious ADRs, with interaction terms for age group and polypharmacy, and time-to-event analysis (Cox proportional hazards model) evaluates resolution and outcome timing. Subgroup analyses stratify by therapeutic area (cardiovascular, central nervous system, anti-infectives) and by organ system involvement. The qualitative data are analyzed using thematic analysis, guided by the Theory of Reasoned Action to interpret clinicians’ reporting behaviors and the Health Belief Model to understand perceived ADR risks in older adults. A conceptual model synthesizes these findings, illustrating pathways from patient factors to reporting and clinical outcomes. Expected findings anticipate a higher ADR incidence rate in the elderly, with greater severity and poorer outcomes, driven by polypharmacy, comorbidity burden, and age-related pharmacokinetic/d pharmacodynamic changes. Certain drug classes (anticoagulants, antiplatelets, non-steroidal anti-inflammatory drugs, sedative-hypnotics, and antidiabetics) are expected to show disproportionate risk in older patients, while age-associated underreporting or delayed reporting is anticipated to be more pronounced in primary care settings. The study aims to quantify the extent to which polypharmacy mediates the relationship between age and ADR risk and to identify gaps in current pharmacovigilance systems that impede timely detection and intervention. The contribution to knowledge includes (1) robust quantitative evidence detailing age-stratified ADR burden and its determinants; (2) validation of predictive models for ADR risk incorporating age, polypharmacy, and comorbidity; (3) an evidence-based framework for targeted pharmacovigilance interventions and safer prescribing guidelines for older adults; and (4) practical insights into clinician reporting behavior to inform policy and training programs. The main conclusion is that elderly individuals experience a disproportionately higher burden of ADRs with greater severity, necessitating optimization of drug regimens, enhanced monitoring, and improved reporting mechanisms. Recommendations include implementing age-specific pharmacovigilance dashboards, routine reconciliation of medications at transitions of care, enhanced education on ADR recognition in geriatrics, and policy measures to incentivize high-quality ADR reporting from primary care and community settings.