The measurement gap

For as long as pharmacovigilance has existed, adverse events have been defined, coded, and reported primarily through the clinician's lens. A patient experiencing fatigue is coded as "fatigue" in MedDRA. A patient experiencing nausea is coded as "nausea." The coding captures the event. It does not capture the patient's experience of the event — its severity as they live it, its impact on daily function, its duration and trajectory, or the trade-offs they make between efficacy and tolerability.

This measurement gap has been quantified. Studies comparing clinician-reported adverse events with patient self-reports consistently show that clinicians underreport symptomatic adverse events by 40-50%. The discrepancy is greatest for subjective symptoms — fatigue, cognitive changes, gastrointestinal discomfort, mood alterations — precisely the events that most affect patient quality of life and treatment adherence.

Why this matters for pharmacovigilance

The consequences of this measurement gap are not abstract. If clinician-reported adverse events systematically underestimate the burden of symptomatic toxicity, then safety profiles based on clinical trial data underestimate real-world tolerability. Benefit-risk assessments built on these profiles may overweight efficacy relative to the adverse-event burden that patients actually experience.

The NCI's Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE) demonstrated in large-scale oncology trials that patient self-reporting captures symptomatic adverse events earlier, at higher rates, and with greater severity discrimination than investigator reporting.

The regulatory signal

The EMA's guideline on the use of patient-reported outcome measures in oncology marketing authorization applications represents the clearest regulatory signal that patient-reported safety data is moving from academic interest to regulatory expectation. The guideline does not yet mandate patient-reported adverse event collection in all trials, but it establishes the methodological framework for incorporating PRO data into benefit-risk assessment.

The challenge ahead is operational: pharmacovigilance workflows, MedDRA coding systems, and periodic safety report formats were designed for clinician-reported data. Integrating patient-reported outcomes requires rethinking data collection, coding, analysis, and presentation at every stage of the safety data lifecycle.