Every pharmacovigilance system is built on one unit of information: the individual case safety report, or ICSR. Before a case can be assessed, coded or reported to an authority, someone has to decide whether it is valid.
That decision rests on four minimum criteria. In this article I explain each one, using situations of the kind I encounter as a community pharmacist in Switzerland.
The four minimum criteria
Internationally harmonised guidance (ICH E2D and EU GVP Module VI; European Medicines Agency [EMA], 2017; International Council for Harmonisation [ICH], 2003) defines a valid ICSR as one that contains at least:
- An identifiable patient.
- An identifiable reporter.
- At least one suspect medicinal product.
- At least one suspected adverse reaction or event.
If any of the four is missing, the case is not yet valid. It is not discarded, however. The safety team records it and tries to obtain the missing information through follow-up (EMA, 2017).
1. An identifiable patient
The patient does not need to be named. One identifier is usually enough: initials, age or age group, sex, date of birth or a patient number.
Valid: "A woman in her sixties developed ankle swelling two weeks after starting amlodipine." Age group and sex identify a real person.
Not yet valid: "Several customers have complained about headaches with this brand." No individual patient can be identified.
2. An identifiable reporter
The reporter must be a real, contactable person. That can be a pharmacist, doctor, nurse, the patient or a relative.
When a pharmacist submits a report, the reporter is clearly identifiable. In industry, the challenge often arises with indirect sources, such as social media posts or anonymous calls to a company hotline.
3. A suspect medicinal product
The report must name at least one product suspected of causing the reaction. A brand name, an active substance or a clear description is enough.
In practice, I always add concomitant medicines too, including over-the-counter products and herbal remedies. They may turn out to be the actual cause, or they may point to an interaction.
4. A suspected adverse reaction
The report must describe at least one event. "Felt unwell" is weak; "dizziness and fainting on standing" is far more useful.
It also helps to separate what the patient experienced from what was diagnosed. Both belong in the narrative.
Beyond validity: what reviewers assess next
A valid case moves on to further assessment. Four questions are central.
Is it serious? Regulatory seriousness has a specific meaning. A case is serious if it results in death, is life-threatening, requires or prolongs hospitalisation, causes persistent disability, leads to a congenital anomaly, or is otherwise medically important (ICH, 1994).
Is it expected? Expectedness, also called listedness, compares the reaction with the reference safety information, such as the Swiss product information for a marketed medicine (ICH, 2003).
Is it plausibly related? Causality is assessed with structured methods such as the WHO-UMC system (Uppsala Monitoring Centre, n.d.). Timing, dechallenge, rechallenge and alternative explanations all matter.
How quickly must it be reported? Serious, unexpected cases trigger expedited reporting. Internationally, the standard expedited timeline for these cases is 15 calendar days from the day the company first receives the minimum information (ICH, 2003), often called "day zero".
Coding and transmission
Once assessed, events are coded using MedDRA, the standard medical terminology for regulatory activities (Brown et al., 1999). Consistent coding is what allows similar cases to be found and compared across a large database.
Cases are then transmitted electronically in the ICH E2B format to authorities and, where relevant, to partner companies (EMA, 2017).
Why accuracy at the start matters most
During my MSc at the University of Basel, the Drug Safety & Database Research Seminar showed me how much safety analysis depends on the quality of individual cases. Missing dates, vague event descriptions or unclear medication lists cannot be corrected by later statistics.
That is why I treat the first conversation with a patient as the most important step. The four minimum criteria are easy to remember, but collecting them well under time pressure, in a busy pharmacy, is a practical skill.
Why this matters to employers
Safety teams in pharmaceutical companies process large volumes of incoming information from many sources: healthcare professionals, patients, literature, partners and patient support programmes. Recognising a valid case quickly, and knowing what to follow up, keeps the whole system compliant.
My experience gives me a reporter's view of the case, which helps me write follow-up questions that healthcare professionals can actually answer.
Key takeaways
- A valid ICSR needs an identifiable patient, an identifiable reporter, a suspect product and an event.
- Incomplete cases are followed up, not discarded.
- Seriousness, expectedness and causality determine how a case is processed and how fast it is reported.
References
Brown, E. G., Wood, L., & Wood, S. (1999). The Medical Dictionary for Regulatory Activities (MedDRA). Drug Safety, 20(2), 109–117. https://doi.org/10.2165/00002018-199920020-00002
European Medicines Agency. (2017). Guideline on good pharmacovigilance practices (GVP): Module VI – Collection, management and submission of reports of suspected adverse reactions to medicinal products (Rev 2). https://www.ema.europa.eu
International Council for Harmonisation. (1994). Clinical safety data management: Definitions and standards for expedited reporting E2A. https://www.ich.org
International Council for Harmonisation. (2003). Post-approval safety data management: Definitions and standards for expedited reporting E2D. https://www.ich.org
Uppsala Monitoring Centre. (n.d.). The use of the WHO-UMC system for standardised case causality assessment. https://who-umc.org
Member discussion