Health-Authority Engagement
What an organisation says to a regulator becomes a commitment it will be held to, often years later and by different people. Engagement is therefore a controlled activity rather than a conversation: positions have to be consistent across functions, undertakings have to be captured at the moment they are made, and responses have to be defensible when the underlying issue recurs rather than merely sufficient to close the question in front of you.
What an explainer is not
A topic explainer is SPEQ’s synthesis of what a practice involves, cited to the standards that govern it. It does not reproduce their text, and it does not determine which of them apply to your product or process.
[ POSITION IN THE FRAMEWORK ]
7 DIMENSIONS · 20 LINKSA meeting with a regulator is an evidence event: what is said becomes part of the record, and the value comes from asking a question the organisation is genuinely prepared to hear an unwelcome answer to.
06 · QUALITY MATURITY — HEALTH-AUTHORITY ENGAGEMENT, REACTIVE TO ADAPTIVE
Contact happens when something is wrong or something is due. Each interaction starts from nothing.
Meetings are prepared for and minuted, but the record lives with whoever attended and advice from one programme never reaches another.
Questions are framed so the answer changes a decision, positions are agreed internally before the meeting, and outcomes are recorded where the next programme will find them.
Commitments made to authorities are tracked to completion like any other obligation, and divergence from advice is a documented decision rather than a drift.
The relationship is predictable in both directions: the organisation is known for filings that match what was discussed, which is what makes the next conversation cheaper.
SPEQ’s shared five-stage progression, labelled synthesis — not the FDA QMM rating scale. Where does your organization sit? Score your quality system →
07 · REGULATORY & EVIDENCE
GOVERNING STANDARDS · 4
Derived from the 4 standards SPEQ maps to this subject, across 3 regulatory bodies: ICH, FDA, EC.
RECORDS & OBJECTIVE EVIDENCE
- Meeting requests with the questions posed and the positions taken
- Briefing documents, and the internal alignment behind them
- Official minutes or the authority’s written response
- A register of commitments made, with their status
- Records of where advice was not followed, and the reasoning
COMMON INSPECTION FINDINGS
- Commitments made in a meeting with no tracking to completion
- Advice received and quietly not followed, with no documented decision
- Meeting outcomes held by individuals, so a later programme repeats the question
- Questions framed so that only a favourable answer is possible, which wastes the interaction
- Divergence between what was described to the authority and what was filed
The engagement instruments, and what each is for
Authorities provide structured routes for advice before a submission: formal meetings in the US across the development lifecycle, scientific advice and protocol assistance from EMA and national agencies in the EU, and equivalent mechanisms elsewhere. These exist to reduce the risk that a programme generates evidence the reviewer will not accept, and they are chronically under-used by organisations that treat them as a formality or fear exposing a weakness.
The instruments differ in what they bind. Advice is generally not binding on either party, but it is on the record — and a sponsor that received advice and departed from it without explanation will be asked why. A jurisdictional determination under a Request for Designation is binding. Knowing which category a given interaction falls into determines how carefully the briefing package has to be written.
A briefing package is an argument, not a data dump
The common failure is a package that presents everything the sponsor knows and asks the authority to react. What produces useful advice is the opposite: a specific set of questions, each with the sponsor’s own proposed answer and the reasoning behind it, supported by only the data needed to evaluate that reasoning. A regulator asked "is this acceptable?" will answer conservatively; a regulator asked "we propose X because Y — do you agree?" engages with the argument.
This also protects the organisation later. A question asked precisely produces an answer that can be relied on precisely. A vague question produces a vague answer that both sides will read differently when it matters.
Commitments are made faster than they are recorded
The most consequential failure in authority engagement is administrative. Undertakings are given verbally in meetings, embedded in response letters, or attached as conditions of approval, and they are not captured anywhere the organisation systematically tracks. Years later a commitment comes due, the person who gave it has moved on, and the evidence that it was met either exists somewhere or does not.
A commitment register is the control: every undertaking captured at the moment it is made, with its owner, its due date, and the evidence that will close it. It has to be populated from meeting minutes and from correspondence, not only from formal approval conditions — because the informal undertaking, given in a meeting to move past a difficult question, is exactly the one nobody logs.
One organisation, one position
Authorities compare. When two functions answer the same authority differently about the same product, or when a response to one authority contradicts a filing made to another, it is read as a quality-system failure rather than a coordination problem — because that is what it is. The underlying cause is usually that no single controlled record of position exists, so each response is drafted from whatever the author knows.
The control is a maintained record of positions per authority and product, and cross-functional review of responses before they are sent — including by the function that will have to deliver whatever is being promised. A regulatory affairs group that commits manufacturing to a timeline manufacturing has not seen has created a second problem while solving the first.
SPEQ interpretation — inspection is engagement too
Organisations tend to run authority engagement and inspection management as separate disciplines, with different owners and different records. From the authority’s side they are one relationship. A commitment given during an inspection closeout, a response to a Form 483 or an inspection finding, and an undertaking in a submission response all carry the same weight and all become the record against which the next interaction is judged.
SPEQ’s view is that the commitment register should span both. The register is the single place an organisation can answer the question a regulator will eventually ask — what did you tell us you would do, and did you do it — and splitting it by the channel the commitment arrived through guarantees that some of them are answered by nobody.
FREQUENTLY ASKED
Is regulatory scientific advice binding?
Generally no — advice from EMA, national agencies or FDA meetings is not binding on either party. But it is on the record. A sponsor that received advice and departed from it will be asked to explain the departure, so the practical effect is that advice constrains without formally binding. Jurisdictional determinations, such as an FDA Request for Designation outcome, are a different category and do bind.
What belongs in a commitment register?
Every undertaking the organisation gives an authority, whatever channel it arrived through: conditions of approval, post-approval commitments, responses to questions, inspection-closeout undertakings, and things said in meetings. Each entry needs an owner, a due date, and the evidence that will close it. The informal undertaking given to move past a difficult question is the one most often missing.
Who should approve a response to a health authority?
The functions that must deliver whatever it promises, not only the function drafting it. A response committing manufacturing to a timeline manufacturing has not reviewed creates a second problem while closing the first. Cross-functional review before dispatch is what keeps the organisation’s positions consistent.
Why does inconsistency between functions matter so much?
Because authorities compare what they are told, across functions and across submissions. Two different answers to the same question is not read as poor coordination; it is read as evidence that the quality system does not control what the organisation asserts, which is a far broader finding than the inconsistency itself.