The Regulatory Blueprint

Executive analysis on where medical device regulation is headed.

Published as the landscape shifts, covering the changes most relevant to how regulatory and quality programs operate.

21 CFR 820 → QMSR · ISO 13485:2016

FDA QMSR transition: aligning legacy 21 CFR 820 systems with ISO 13485:2016

Practical strategies for regulatory and quality leaders working through the harmonized framework.

FDA's Quality Management System Regulation (QMSR) replaced the standalone requirements of 21 CFR 820 with a framework built around ISO 13485:2016, incorporated by reference and supplemented by a small set of FDA-specific clauses. For manufacturers, that is not a paperwork update. It is a structural change in how a quality system has to be organized, documented, and defended under inspection.

The gap most teams underestimate is not the regulation text itself, but the internal documentation architecture built around the old 820 numbering. Procedures, training records, and audit checklists written against legacy section references need to be re-mapped to ISO 13485 clause structure, not rewritten from scratch, but deliberately cross-walked so nothing silently falls out of scope during the transition.

In practice, a defensible transition runs in three passes: a clause-by-clause gap analysis comparing your current QMS to ISO 13485:2016 plus the retained FDA-specific requirements; a prioritized remediation plan that closes high-risk gaps (management responsibility, CAPA, design controls) before administrative ones; and a controlled internal audit against the new structure before an external body finds the gap for you. Manufacturers who treat this as a compliance-mapping exercise, rather than a genuine systems review, tend to pass the paperwork check and fail the practical one.

The citation language hasn’t caught up yet. Warning letters issued after February 2, 2026 still cite 21 CFR 820, since the inspections behind them were conducted before QMSR took effect.1 FDA has been explicit that this is a lag, not a reprieve: corrective actions must be proposed and implemented under QMSR regardless of which regulation the letter itself names, so the re-mapping work is due now, not when the letters catch up.

1. Covington & Burling LLP, Quarterly Medical Device Warning Letters Update: January – March 2026 (May 15, 2026).

21 CFR 820.100 · ROOT CAUSE

High-stakes CAPA architecture: structuring root-cause investigations that survive scrutiny

How to build a corrective action record that holds up when a health authority reads it line by line.

Most CAPA findings do not come from a missing corrective action. They come from a root cause that will not survive a second reading. Investigators default to the first plausible explanation, close the CAPA, and move on, leaving a record that looks complete but collapses under a "why" asked three times instead of once.

A defensible CAPA is architected backward from the effectiveness check, not forward from the symptom. Before an investigation starts, define what evidence will prove the corrective action actually worked: a measurable, time-bound, independently verifiable criterion. Only then build the root-cause analysis that leads there. This reverses the usual order and eliminates the most common finding in FDA inspections: CAPAs closed on the basis of implementation, not effectiveness.

The second failure point is scope. A root cause stated at the level of "operator error" or "documentation gap" almost never survives scrutiny, because it stops at the first human in the chain instead of the system that allowed the error to occur. Structuring the investigation around systemic contributing factors (process design, training adequacy, detection controls) produces a corrective action that changes an outcome, not just a paper trail.

The data backs up why this is worth getting right. CAPA was tied for the most-cited quality system finding in FDA warning letters for 2025, and pulled ahead alone in the first quarter of 2026, cited in nine of seventeen letters.1 It is not a niche or declining risk area. It is the single most common reason a quality system fails under inspection right now.

1. Covington & Burling LLP, Quarterly Medical Device Warning Letters Update: January – March 2026 (May 15, 2026) and Quarterly Medical Device Warning Letters Update: October – December 2025 and 2025 Annual Summary (Feb. 19, 2026).

21 CFR 820.30(i) · CHANGE CONTROL

Design change governance: new submission or internal documentation update?

A framework for the decision that determines your entire regulatory timeline.

Every design or manufacturing change to a cleared or approved device forces the same question, and getting it wrong in either direction is expensive: treat a substantive change as routine documentation and risk a field action or warning letter; treat a routine change as submission-worthy and burn months of runway you didn't need to spend.

The determination should never be made by the engineer who authored the change. It belongs to a structured impact assessment that evaluates the change against intended use, technological characteristics, and performance specifications relative to the cleared or approved baseline: the same questions a reviewer will ask, asked internally first. Where a change touches risk controls identified in the ISO 14971 risk management file, that assessment needs to be reopened explicitly, not assumed to still apply.

Building a standing change control board (regulatory, quality, and engineering represented at every review, with threshold criteria documented in advance rather than debated case by case) converts this from a recurring point of friction into a repeatable, defensible process. The goal is not fewer changes; it is faster, more confident decisions about which changes need a submission and which do not.

Want this analysis before it’s published here?

Reach out directly and we’ll add you to the distribution for new Regulatory Blueprint analysis.

Get in Touch