When What You Found Is Not What Is in Force

Part of Regulatory Affairs and the Jurisdiction Map: Reading a Jurisdiction.

About 5 minutes. Free, in your language, no account needed.

Why this matters

You can find the right instrument and still have the wrong answer, because regulation moves and documents do not announce that they have been superseded.

The failure is quiet. A superseded regulation reads exactly like a current one. It has no banner, no strikethrough, and frequently still sits on the authority's own site because removing old versions destroys the historical record. A person who finds it, reads it correctly, and applies it faithfully will be wrong, and nothing in the document will have warned them.

The four ways a source goes stale

Superseded. A new instrument replaced the old one. India's New Drugs and Clinical Trials Rules, 2019 superseded Schedule Y, and a team working from Schedule Y would be applying a framework that no longer governs. The CRLN register records the supersession explicitly for this reason.

Amended. The instrument still governs, but a specific provision changed. This is harder to catch than supersession, because the document name and number are unchanged and only a section differs.

Transitional. Both an old and a new framework are in force simultaneously for different trials, usually determined by the date a trial was authorized. During a transition, the correct answer depends on which framework your particular trial sits under, and asking "what is the rule" without specifying the trial produces a confidently wrong answer.

Not yet in force. A published instrument with a future application date. ICH E6(R3) Annex 2 is a live example, with an in-force date of 15 January 2027. Applying it early is not compliance, and neither is ignoring the fact that it is coming.

How to tell, before relying on it

Four checks, in order, each of which takes minutes:

Check the authority's own current listing. If the authority publishes a list of instruments in force, that list is the answer. A document reachable by direct link but absent from the current listing is a strong signal.

Look for the commencement and application dates. An instrument states when it applies. A future date means it does not yet, and an absent date means you have not found the authoritative version.

Look for what it supersedes, and search for what supersedes it. Instruments usually say what they replace. Searching for the instrument's own name alongside words like "amendment" or "revision" is a crude method that works.

Record the date you checked. This is the check that makes the other three durable. A claim with a verification date can be aged out deliberately. A claim without one has to be redone from scratch, so in practice it is never redone at all.

The transition problem, in detail

Transitions are the hardest case and worth dwelling on, because they defeat the instinct that there is one right answer.

When a framework changes, trials already authorized under the old one usually continue under it for a defined period, while new trials go under the new one. For a multi-country study running for five years, this can mean the same sponsor operates under two frameworks in the same country at the same time, for different studies, legitimately.

The practical consequence is that a regulatory position must be recorded per trial, not per country. A country-level intelligence file that says "the framework is X" is insufficient during a transition, because the correct answer for a given study depends on its authorization date.

Teams get this wrong by updating the country file when the framework changes and then applying the new answer to every study, including the ones that legitimately remain under the old one.

In practice

A regulatory associate is asked to confirm the safety reporting requirements for an ongoing trial in a country that changed its framework eighteen months ago.

The country file has been updated and describes the new framework. The associate reports the new requirements.

The trial was authorized three years ago and sits, correctly, under the transitional provisions of the old framework, which has a different recipient list. Following the new requirements would mean sending reports to a body that does not expect them and omitting one that does.

What catches this is a single habit: before answering a question about an ongoing trial, establish the trial's authorization date, then ask which framework applies to a trial authorized on that date. Two questions instead of one, and the second is the one that matters.

What this looks like when it goes wrong

A sponsor's procedure cites an instrument by name and number, and that instrument was superseded four years ago. The procedure has been reviewed twice since and approved both times.

It survived review because reviewers checked that the procedure was internally consistent and matched the sponsor's practice. Nobody re-verified the external citation, because re-verification is not what a procedure review does.

The error surfaces during an inspection, where the inspector reads a citation to a superseded instrument as evidence that the sponsor is not tracking the regulatory environment. That inference is harsher than the underlying error, and it is difficult to argue against, because the citation has been sitting in an approved procedure for four years.

Key takeaways