Predicate Rules and Hybrid Systems
DI-0301 continues: two threads about one question
Your Quality Analyst (QA) is back with the next piece of DI-0301, her data-integrity self-assessment. Two things are sitting on her desk, and they turn out to be the same question wearing two different coats.
Thread one picks up something module 1 left hanging. Her first test case, DED-0458 — the BECS (Blood Establishment Computer Software) donor-eligibility determination she ran through ALCOA (attributable, legible, contemporaneously recorded, original or a true copy, accurate) — passed “original or a true copy” today, because the live BECS record is what’s being examined. But she flagged a nuance: what happens the day someone decides it’s easier to just keep a printed report instead of going back into BECS? Is that printout still the real record, or a copy standing in for something it can’t actually replace? She never answered that. She’s back to answer it now.
Thread two is new. DI-0301 has reached Fairview, one of Lakeshore’s satellite collection sites, where donor-eligibility interviews are still conducted using a paper donor file — pen, clipboard, a physical folder per donor. For central reporting, certain fields from that paper file — donor ID, determination, date — get typed into BECS afterward, so Lakeshore’s central systems can see Fairview’s activity alongside every other site’s. Your QA asks the obvious question, and gets an answer that worries her more than a wrong answer would have: nobody at Fairview, and nobody at Quality, had ever actually decided which of the two — the paper file or the BECS entry — is the record of record. The standard operating procedure (SOP) covering donor eligibility at Fairview doesn’t say. It was just never written down.
No citations yet. Just two open questions, both circling the same thing: when a record exists in two forms, which form is actually the one that counts?
Why this matters — and the ITSM bridge
Here’s the failure both threads are circling. A record that exists in two places, with nobody having decided which one is authoritative, isn’t “extra safe because there are two copies.” It’s the opposite. If the two copies ever disagree — a re-keying error, a field skipped, a timestamp drift — nobody can say which one is wrong, because nobody decided in advance which one was ever supposed to be right. The ambiguity itself is the failure, long before any specific field turns out to be incorrect.
You already carry the instinct for this. In configuration management, you’ve almost certainly lived the moment where a CMDB (configuration management database) says one thing about an asset — its owner, its location, its patch level — and a spreadsheet someone built years ago for a specific project says something else. Nobody intended for the spreadsheet to become anyone’s reference, but it did, and now two “sources of truth” exist for the same fact. The fix in ordinary IT governance is always the same: designate the authoritative source, in writing, and treat the other one as a convenience view at best.
Here’s where the regulated world demands more. In ordinary IT, an authoritative-source disagreement between a CMDB and a spreadsheet is a process-hygiene problem — worth cleaning up, rarely urgent on its own. For a record a predicate rule already requires Lakeshore to keep, failing to document which format is authoritative isn’t just untidy. It means nobody can say, with evidence, whether the record satisfies ALCOA or Part 11 (21 CFR Part 11, Electronic Records; Electronic Signatures) at all — because ALCOA’s and Part 11’s protections (review, retention, signature controls, audit-trail review) only mean something once you know which artifact they’re supposed to be protecting. The ambiguity is the compliance gap. Not a future risk of one — the gap itself, today, as written.
The rule in plain English
Three ideas untangle both of DI-0301’s threads. Here they are, in order.
First: a “predicate rule” is just the regulation that already required the record to exist, before Part 11 ever entered the picture. Part 11 doesn’t invent new recordkeeping duties. It never tells Lakeshore it has to start keeping a record it wasn’t already required to keep. What it does is set conditions for when an electronic version of a record some other regulation already demands can count as trustworthy and legally equivalent to paper. For DED-0458, the predicate rule is 606.160 (21 CFR 606.160, Records) and 606.100 (21 CFR 606.100, Standard operating procedures) — the blood establishment current good manufacturing practice (CGMP) sections that already require a donor-eligibility determination to be documented, whether or not BECS exists. For the automated equipment behind it, 211.68 (21 CFR 211.68, Automatic, mechanical, and electronic equipment) plays the same role elsewhere in this curriculum. Part 11 rides on top of whichever predicate rule already required the record. It never stands alone.
Second: whether a given record counts as a “Part 11 record” at all depends on what you actually rely on, not just what format exists. FDA’s (Food and Drug Administration’s) own guidance calls this the “narrow interpretation” of Part 11’s scope, and it works like a fork in the road. If Lakeshore chooses to keep a predicate-rule-required record electronically in place of paper, and relies on that electronic version to actually perform the regulated activity (accepting or deferring a donor, releasing a unit), Part 11 applies to it. But if Lakeshore generates a paper record, that paper record satisfies everything the predicate rule requires, and the people doing the work rely on the paper record — not an electronic copy — to do their regulated work, FDA generally doesn’t treat that as “using electronic records in lieu of paper records,” and Part 11 doesn’t reach the electronic copy the same way. Reliance decides the question. Format alone doesn’t.
Third: not every record that looks the same on paper and on screen actually preserves the same information — and that’s where FDA’s “static” and “dynamic” distinction comes in. A static record is fixed data: a paper form, a scanned image, a printout. A dynamic record is one whose format lets the user interact with the content — change a view, reprocess the underlying data, apply a different filter. A printout of a dynamic record isn’t automatically equivalent to the dynamic record itself. It’s a single frozen snapshot of something that was built to be examined more than one way. Whether a printout can stand in for the thing it was printed from depends on whether anything real was lost in the printing.
Put the three together and both of DI-0301’s threads answer the same way: figure out which format is actually relied on for the regulated activity, confirm whether that format is static or dynamic, and write the decision down. Don’t leave it to whoever happens to open the file next.
Citation decoder
| Citation | How to read it aloud | What it requires | In our scenario |
|---|---|---|---|
| FDA 2003 Part 11 Scope and Application guidance, Introduction | “the 2003 Part 11 scope guidance, introduction” | Guidance tier. Defines “predicate rules” as the underlying Act, PHS Act, and FDA regulations other than Part 11 itself. | Names what 606.160 and 606.100 actually are to DED-0458 and Fairview’s paper file: the rules that required the record before BECS or Part 11 entered the picture. |
| FDA 2003 Part 11 Scope and Application guidance, § III.B.1, “Narrow Interpretation of Scope” | “section three-B-one, narrow interpretation” | Guidance tier. Distinguishes choosing electronic format in place of paper (Part 11 applies) from relying on a paper record that independently satisfies the predicate rule (Part 11 doesn’t reach the electronic copy the same way). | The test DI-0301 applies to both DED-0458’s printout question and Fairview’s paper-versus-BECS question. |
| FDA 2003 Part 11 Scope and Application guidance, § III.B.2, “Definition of Part 11 Records” | “section three-B-two” | Guidance tier. Recommends documenting, in advance, whether you rely on the electronic or paper record for each predicate-rule-required record — e.g., in an SOP. | What Fairview’s SOP should have said and didn’t; what DI-0301 now recommends Lakeshore write down for DED-0458 too. |
| 21 CFR 11.2 | “eleven point two” | Regulation. Lets a person use electronic records/signatures in place of paper, for records kept internally (a) or submitted to FDA (b), provided Part 11’s own requirements are met. | The regulatory hook the “narrow interpretation” passage cites directly — the choice Part 11 actually governs. |
| FDA 2018 Data Integrity Q&A, Question 1d | “question one, part d” | Guidance tier. Defines “static” (fixed-data, like paper or an image) versus “dynamic” (format allows user interaction with the record’s content). | Whether a DED-0458 printout, or Fairview’s BECS re-entry, preserves what the live, dynamic BECS record actually is. |
| FDA 2018 Data Integrity Q&A, Question 10 | “question ten” | Guidance tier. A paper printout can satisfy retention if it is the original record (e.g., some instruments); it cannot if the underlying electronic record is genuinely dynamic and the printout loses that. | Plants the question module 4 answers for Lakeshore’s own legacy instrument — not resolved here. |
The actual text
Start with the predicate-rule definition itself, quoted verbatim from sources/fda-guidance/part-11-scope-and-application-2003.md:
The underlying requirements set forth in the Act, PHS Act, and FDA regulations (other than part 11) are referred to in this guidance document as predicate rules.
“The Act” is the Federal Food, Drug, and Cosmetic Act; “PHS Act” is the Public Health Service Act. Everything else in that sentence is the point: predicate rule isn’t a special category of regulation. It’s just FDA’s own shorthand for “whatever rule already required this record,” as distinct from Part 11 itself.
Now the narrow-interpretation passage, quoted in full:
Under the narrow interpretation of the scope of part 11, with respect to records required to be maintained under predicate rules or submitted to FDA, when persons choose to use records in electronic format in place of paper format, part 11 would apply. On the other hand, when persons use computers to generate paper printouts of electronic records, and those paper records meet all the requirements of the applicable predicate rules and persons rely on the paper records to perform their regulated activities, FDA would generally not consider persons to be “using electronic records in lieu of paper records” under §§ 11.2(a) and 11.2(b). In these instances, the use of computer systems in the generation of paper records would not trigger part 11.
Read that passage twice, because it’s the whole module. It doesn’t ask “does an electronic version of this record exist somewhere.” It asks what Lakeshore actually relies on to do the regulated work. The same guidance backs that logic with a direct recommendation, quoted here from its “Definition of Part 11 Records” section:
We recommend that you determine, based on the predicate rules, whether specific records are part 11 records. We recommend that you document such decisions.
And again, more specifically, a few lines later:
Accordingly, we recommend that, for each record required to be maintained under predicate rules, you determine in advance whether you plan to rely on the electronic record or paper record to perform regulated activities. We recommend that you document this decision (e.g., in a Standard Operating Procedure (SOP), or specification document).
Notice the word “in advance.” FDA isn’t recommending Lakeshore figure this out after an inspector asks. It’s recommending the decision get made and written down before the question ever comes up — exactly what didn’t happen at Fairview.
Here’s the regulation that passage cites directly, 21 CFR 11.2, quoted in full from sources/cfr/11.2.md:
For records required to be maintained but not submitted to the agency, persons may use electronic records in lieu of paper records or electronic signatures in lieu of traditional signatures, in whole or in part, provided that the requirements of this part are met.
For records submitted to the agency, persons may use electronic records in lieu of paper records or electronic signatures in lieu of traditional signatures, in whole or in part, provided that:
The requirements of this part are met; and
The document or parts of a document to be submitted have been identified in public docket No. 92S-0251 as being the type of submission the agency accepts in electronic form. This docket will identify specifically what types of documents or parts of documents are acceptable for submission in electronic form without paper records and the agency receiving unit(s) (e.g., specific center, office, division, branch) to which such submissions may be made. Documents to agency receiving unit(s) not specified in the public docket will not be considered as official if they are submitted in electronic form; paper forms of such documents will be considered as official and must accompany any electronic records. Persons are expected to consult with the intended agency receiving unit for details on how (e.g., method of transmission, media, file formats, and technical protocols) and whether to proceed with the electronic submission.
11.2(a) is the one that matters for DED-0458 and Fairview: neither record gets submitted to FDA directly, so the relevant permission is “persons may use electronic records in lieu of paper records… provided that the requirements of this part are met.” Part 11 is the permission slip for a choice — it doesn’t force the choice, and it doesn’t create the underlying duty to keep the record in the first place. That duty is the predicate rule’s.
Now the static/dynamic distinction, quoted verbatim from sources/fda-guidance/data-integrity-cgmp-qa-2018.md (Question 1d):
For the purposes of this guidance, static is used to indicate a fixed-data record such as a paper record or an electronic image, and dynamic means that the record format allows interaction between the user and the record content. For example, a dynamic chromatographic record may allow the user to change the baseline and reprocess chromatographic data so that the resulting peaks may appear smaller or larger. It also may allow the user to modify formulas or entries in a spreadsheet used to compute test results or other information such as calculated yield.
A chromatographic record is one lab technique’s output; Lakeshore doesn’t run that equipment, but the pattern transfers directly. BECS’s own donor-eligibility record is interactive in the same sense: someone can query it, re-sort a donor’s history, filter by a different date range, or pull up a different view of the same underlying data. A printed report is one fixed snapshot of one of those views, taken at one moment. That’s the entire difference between static and dynamic, made concrete.
Last, the guidance’s own answer on stand-alone instrument printouts, quoted in full (Question 10) — plant this one, don’t resolve it yet:
A paper printout or static record may satisfy retention requirements if it is the original record or a true copy of the original record (see §§ 211.68(b), 211.188, 211.194, and 212.60). During data acquisition, for example, pH meters and balances may create a paper printout or static record as the original record. In this case, the paper printout or static record, or a true copy, must be retained (§ 211.180).
However, electronic records from certain types of laboratory instruments—whether stand-alone or networked—are dynamic, and a printout or a static record does not preserve the dynamic record format that is part of the complete original record. For example, the spectral file created by FT-IR (Fourier transform infrared spectroscopy) is dynamic and can be reprocessed. However, a static record or printout is fixed and would not satisfy CGMP requirements to retain original records or true copies (§ 211.180(d)). Also, if the full spectrum is not displayed in the printout, contaminants may be excluded.
Hold that passage. It’s doing exactly one job in this module: showing you that “is a printout good enough” has two genuinely different right answers depending on the instrument, and that the difference turns on whether the instrument’s electronic record is static or dynamic by nature — not on habit, or on which answer is more convenient. Lakeshore’s own aging stand-alone laboratory instrument — the one DI-0301 hasn’t reached yet — raises exactly this question. Which case it looks like, pH-meter or FT-IR, is next module’s work, not this one’s.
Resolving DED-0458’s nuance, now
Module 1 left this open on purpose. Here’s the resolution, using this module’s own framework instead of guessing.
Apply the narrow interpretation first. Is DED-0458 a record a predicate rule requires? Yes — 606.160 requires Lakeshore to keep donor-eligibility determination records. Today, the people who perform the regulated activity — deciding whether a donor is accepted — rely on the live BECS record, not a printout, to do it. Under the narrow-interpretation test, that means DED-0458 is a Part 11 record: Lakeshore chose to keep this predicate-rule-required record electronically, in place of paper, and relies on the electronic version.
Now apply static versus dynamic. BECS’s donor-eligibility record is dynamic — it’s a row inside an interactive, queryable system, open to being re-examined with different views or filters, tied to its own audit trail and e-signature controls (the subject of last module). A printed report generated from it is static: one frozen view, taken once. If anyone later swaps the live record for that printout as “the” record of DED-0458, they’d be substituting a static snapshot for a dynamic original — the same substitution FDA’s Question 10 above says can’t be made safely for a genuinely dynamic record.
So the resolution: the live BECS record — not any printout generated from it — has to remain the thing Lakeshore treats as original and record-of-record for DED-0458. That’s not a new control to build. It’s a decision to make and write down, per the 2003 guidance’s own recommendation: document, in Lakeshore’s SOP or specification for DED-0458’s record type, that the electronic BECS record is the relied-upon original and any printout is a convenience copy only — not a substitute, and not itself acceptable as the retained record. The open ALCOA letter from module 1, “accurate,” is still unresolved and still someone else’s job. This one closes.
Fairview’s paper file, resolved the same way
Apply the same two tests to Fairview.
Narrow interpretation first. What does the staff member at Fairview actually rely on, in the moment, to decide whether to accept or defer a donor? The paper file — that’s the record they consult, write on, and act on during the interview itself. The BECS entry happens afterward, typed in from the paper file, for central reporting. Under the narrow-interpretation test, Fairview hasn’t “chosen electronic format in place of paper” for the determination itself — it generates a paper record that satisfies the predicate rule (606.160, 606.100) and relies on that paper record to perform the regulated activity. By the guidance’s own logic, that means Part 11 doesn’t reach the BECS re-entry the same way it would reach a record Fairview genuinely replaced paper with.
Static versus dynamic, second. The paper file is static by nature — ink on a form. The question worth asking about the BECS re-entry is whether anyone at Fairview, or centrally, ever actually uses that entry interactively — re-queries it, cross-checks it against another site’s data, re-sorts it — or whether it only ever gets read back exactly as typed, as a static copy of the same facts the paper file already holds. DI-0301’s honest finding: nobody had actually thought about it in these terms. The BECS entry exists. Whether it’s functioning as a dynamic, independently reliable record or as a static mirror of the paper file was simply never decided.
The recommendation. As practiced today — paper relied on for the actual donor-eligibility decision, BECS entry made afterward purely for central visibility — the paper file is Fairview’s defensible record of record, and the BECS entry is a reporting copy. That’s a workable answer, but only if Lakeshore does what the 2003 guidance recommends and actually writes it down: an SOP statement naming the paper file as record-of-record for Fairview’s donor-eligibility determination, with the BECS entry documented explicitly as a derived reporting copy, not an independent original. Once that’s written down, the protections this course has already built — ALCOA (module 1), audit trail review and signature controls (module 2) — apply in full to the paper file: it’s the one that needs the retention, the controlled-document handling, the review attention appropriate to being the real record. The BECS copy still matters — a timing or accuracy gap in it could let a stale donor status reach another site before the cumulative record catches up, which is exactly the shape of problem this course picks up in full in module 6 — but it matters as a reporting feed, not as the thing ALCOA and Part 11 are protecting.
One honest flag, for the choice that was logged: this module uses Fairview as the site still running a paper donor file, by the drafter’s own choice, consistent with (and not contradicting) Fairview’s and Brookfield’s prior light appearances elsewhere in this curriculum.
What an inspector would actually look for
An FDA investigator or AABB (Association for the Advancement of Blood & Biotherapies) assessor wouldn’t ask whether Lakeshore has a predicate-rule policy. They’d ask for one specific thing: proof that Lakeshore knows, in writing, which format of each record it actually relies on.
For DED-0458 and records like it: a specification or SOP line naming the live BECS record as the relied-upon original, not an assumption that “it’s in BECS, so it’s fine.” A bad answer sounds like “we have both the system record and a printout, so we’re covered.” A good answer names one of them as authoritative and explains why the other is a convenience copy only.
For Fairview’s donor file: the same thing, in the other direction — an SOP statement that the paper file is the record of record, with the BECS entry named explicitly as a reporting copy. A bad answer sounds like “we keep both, so nothing’s really at risk.” A good answer can point to the line in the SOP that settles it, and show the paper file getting exactly the handling (retention, review, protection from alteration) a record of record requires.
For the choice itself: evidence the decision was made in advance, not reconstructed after DI-0301 asked the question. An inspector who senses the answer was invented on the spot, to satisfy the question being asked, will probe harder — and should.
Common traps
“If a record exists on paper, Part 11 never applies, full stop.” Not quite. Part 11’s reach depends on reliance, not on whether paper exists somewhere. If Lakeshore relied on an electronic record to perform the regulated activity and only generated a paper printout afterward for convenience, the narrow-interpretation passage above says the electronic record is still the one that counts — the paper copy existing doesn’t retroactively exempt it. DED-0458 is the proof: BECS is what’s relied on, so BECS is the Part 11 record, printout or no printout.
“Keeping a record in two formats is safer than keeping it in one.” DI-0301’s Fairview finding refutes this directly. Two formats with no documented decision about which one is authoritative isn’t redundancy — it’s ambiguity waiting for the two copies to disagree. The fix isn’t picking a format at random. It’s documenting, per the 2003 guidance’s own recommendation, which one Lakeshore actually relies on — and then making sure that one, specifically, gets the protection a real record of record requires.
What this means for your job
- Ask, for any record your team touches, “which format do we actually rely on to do the regulated work?” Not which format exists — which one people would actually act on if the two ever disagreed. That answer is the Part 11 record, regardless of what else happens to exist alongside it.
- Treat “we keep it in both formats” as an incomplete answer, not a reassuring one, until someone can show you the SOP line naming which one is authoritative. Fairview’s gap wasn’t that two formats existed — it’s that nobody had ever written down which one mattered.
- When a system is interactive — queryable, filterable, re-sortable — ask whether anyone has ever swapped a printout in for it as “the” record. That substitution is exactly where a dynamic record quietly turns into a static one without anyone deciding it should.
- Push your QA to document the reliance decision in advance, not after an inspector asks. FDA’s own guidance recommends this explicitly. A decision made under inspection pressure looks like what it is.
- Remember predicate rules are what actually reaches a record — Part 11 is the layer on top. If someone tells you “Part 11 doesn’t apply here, so we’re fine,” ask what predicate rule still governs that record regardless. There’s almost always one underneath.
Check your understanding
No — and this is a genuine application of the narrow-interpretation test, not a simple yes/no recall. The predicate rule (21 CFR 600.15) requires the temperature record to exist; the paper log is what satisfies that requirement and what shipping staff actually rely on to release product. The spreadsheet is a separate, derivative document — never relied on to perform the regulated activity, and not itself required to be maintained under the predicate rule. Under the guidance’s own test, Part 11 reaches records required under predicate rules that are relied on in electronic format in place of paper. This spreadsheet is neither the required record nor the one relied on. It’s a convenience copy, and FDA’s narrow interpretation says the use of a computer to generate it doesn’t trigger Part 11.
No, and this is worth correcting precisely. A predicate rule isn’t a special or stricter category of anything. It’s FDA’s own shorthand — defined in its 2003 Part 11 guidance — for whatever regulation (211.68, 606.160, 606.100, or any other section of the Act, the PHS Act, or FDA’s regulations) already required a record to exist, as distinct from Part 11 itself. 606.160 is a predicate rule the moment Part 11 is the thing being discussed; it’s an entirely ordinary CGMP regulation the rest of the time. Calling something “a predicate rule” doesn’t change what it requires — it just names its relationship to Part 11 in a given conversation.
Not exactly — and the distinction matters. Nothing in Part 11 or its predicate rules was necessarily being violated in substance: the paper file, as actually used, likely does satisfy 606.160’s and 606.100’s requirements. The real finding is a documentation gap against FDA’s own recommendation (guidance tier, “should” language) to decide and document, in advance, which format is relied on for each predicate-rule-required record. An undocumented decision isn’t automatically a regulation violated — it’s a recommendation not yet followed, which is exactly the kind of gap a self-assessment should catch and fix before it hardens into something worse, like the two copies actually disagreeing with nobody positioned to say which one is right.
Where this goes next
DI-0301 now has a real answer for both of its open threads: DED-0458’s live BECS record stays the documented original, and Fairview’s paper file is the documented record of record, with its BECS entry named as a reporting copy. One question stays open on purpose — whether Lakeshore’s own aging stand-alone laboratory instrument behaves more like FDA’s pH-meter example or its FT-IR example. That’s legacy-systems-and-part-11-gaps’s whole job next: apply FDA’s own four-part legacy-system test to that specific instrument, and answer, for Lakeshore’s own equipment, the question this module only planted.