Case Classification vs Report Type in Pharmacovigilance: Key Differences Explained

This blog features:

1. Key differences: Case classification vs Report type
2. How does it improve the regulatory submission

Case Classification and Report Type in Pharmacovigilance

“Two Different Sorting Systems for one ultimate reason”

Every ICSR that lands in a safety database gets sorted along several axes at once: serious or non-serious, expected or unexpected, spontaneous or solicited, initial or follow-up, product liability or not, pregnancy or not.

It’s easy to treat all of this as one big intake checklist. It isn’t. Two distinct classification systems are running in parallel — report type and case classification — and conflating them is one of the more common blind spots in case processing.

Report Type vs. Case Classification: Why Both Exist and What Each One Actually Answers

If you’ve processed cases in a safety database, you’ve seen “Report Type” and “Case Classification” sitting as separate sections on the same case form. They look similar.

Report type

Report type is straightforward: it identifies the source the case came from. Common report types include:

• Spontaneous
• Clinical trial
• Legal (product liability / litigation-derived)
• Post-marketing / non-interventional study
• Literature

Report type answers where did this case come from, and that answer determines the reporting pathway from the start — spontaneous reports carry an implied causality, clinical trial reports follow protocol-driven unblinding and expedited reporting rules under the trial’s own safety plan, legal-sourced reports trigger specific handling for product liability tracking, and literature reports follow their own periodic screening and duplicate-detection process.

Case classification

Case classification is a different question: how should this case be categorized for understanding within the company’s own processing framework — and critically, this is not a single fixed regulatory taxonomy. Beyond the ICH-defined seriousness criteria, companies classify cases by internal convention, within the boundaries of applicable guidance, in whatever categories serve their signal management and regulatory tracking needs. Typical internal case classification categories include:

• Product liability
• Prospective vs. retrospective (was the outcome known before or after enrollment/exposure)
• Pregnancy reports
• Serious vs. non-serious (by adverse event criteria)
• Special situations (overdose, misuse, abuse, medication error, off-label use)

So a company can classify the same case as, for example, “serious, pregnancy-related, retrospective, spontaneous-sourced” — four labels doing four different jobs, only one of which (report type = spontaneous) tells you where the case came from. The other three are case classification decisions made against internal convention layered on top of ICH/regional requirements.

Why maintain both systems rather than one combined taxonomy?

Because they answer different operational questions that all feed into the same regulatory decision:

1. To identify the type of case — report type tells processors immediately what intake rules, source-verification steps, and duplicate-check logic apply.
2. To understand the importance of the case — case classification (seriousness, special situation flags) tells the team how much scrutiny and urgency the case warrants.
3. To understand the expectedness of the case being processed — classification against the current RSI/CCSI determines whether the case is even a candidate for expedited handling.
4. To support triaging — a combined view of report type and classification is what actually drives queue prioritization, not either field alone.
5. To follow country-specific and event-specific conventions — many national regulators layer their own classification or reporting requirements on top of ICH baseline (additional local seriousness criteria, specific special-situation reporting like pregnancy exposure registries), and both report type and case classification feed into which country-specific rule set applies.

Every area of information has its uniqueness of values it would provide.

Conclusion

Case classification in pharmacovigilance behaves like a set of independent checkboxes but functions as a coupled system: seriousness gates the timeline, expectedness (gated by RSI currency) gates whether that timeline is expedited, causality gates whether solicited/spontaneous reports are even in scope, and report format determines whether structured data survives the trip from intake to regulatory submission intact.

The organizations that get this right treat classification as a governed, auditable decision process — with defined triggers for “unassessable,” explicit RSI version control, and structured-data-first case building — rather than a set of dropdown fields a case processor fills in once and moves past.