Quality Control Records and Logs: Contents, Review, Retention
Quality control records are the documented evidence that a process was performed as specified, that its controls were checked, and that deviations were acted upon. A quality control log is the chronological subset of those records in which each control result, its date, its lot information and its performer are entered as the work occurs. This article examines what such records contain, how they are created, corrected and stored, what determines their retention period, how they support audits, and which errors most often render them unreliable. The evidence is drawn from U.S. federal guidance issued by the FDA, the CDC, the EPA and the USDA, supplemented by professional laboratory literature.
TL;DR
- Quality records (audits, management reviews, corrective actions) and technical records (data, forms, worksheets, control graphs, test reports) are distinct categories, and a complete system preserves both (FDA ORA Laboratory Manual).
- A defensible entry is legible, dated, attributed to its author and detailed enough that a competent person could repeat the work from the record alone.
- Retention is not uniform: it is set by the governing program or rule, for example six months for FSIS sanitation records under 9 CFR 416.16 (FSIS).
- A failed control is a trigger for troubleshooting, retesting and documented corrective action, not merely a number to be logged.
- The most frequent defects are illegibility, missing dates or initials, erasures, unexplained discarded data and unrestricted or unbacked electronic storage.
Table of Contents
- Definitions: Records, Logs and Their Purpose
- What a Quality Control Record Should Contain
- Creating Records Correctly
- Storage, Access and Control of Records
- Retention Periods and What Determines Them
- Using Logs to Detect Recurring Problems and Support Audits
- Common Defects That Make Records Unreliable
- Our Perspective
- Where Project Relay and Technical Review Fit
- FAQ
- Sources
Definitions: Records, Logs and Their Purpose
The terminology varies by sector, yet the FDA’s laboratory manual supplies a workable taxonomy. It separates quality records, such as internal-audit reports, management reviews and corrective or preventive actions, from technical records, such as raw data, forms, worksheets, control graphs and test reports (FDA ORA Laboratory Manual, revised May 15, 2019). A quality control log typically belongs to the second category: it is the running ledger of control results from which control graphs and summary statistics are later derived.
The purpose of the two categories differs. Technical records permit reconstruction of what was done to a particular sample, batch or event. Quality records demonstrate that the management system itself is monitored and corrected. An organization that retains only the former can show results but not governance; one that retains only the latter can show governance but not the evidence on which any result rests.
The settings in which these documents are required are diverse: clinical and public health testing, regulated laboratories, food establishments, environmental monitoring and manufacturing. The structure of the record, however, converges across them, which allows a common treatment.
What a Quality Control Record Should Contain
Core identification elements
The FDA manual states that records should be identifiable to the firm, product, person or event concerned, dated, attributed to the person who established them, and sufficiently detailed to preserve an audit trail (FDA). These four properties (identity, date, authorship, traceability) are the minimum on which every other element depends. An undated result cannot be sequenced against a maintenance event, and an unattributed result cannot be queried with the person who produced it.
Content of the laboratory record
The same manual enumerates what laboratory documentation should capture: the analysts performing testing, their training, calibration, equipment and its maintenance, the identities of media, reagents and reference materials, lot numbers and expiration dates, sample handling, preparation and receipt, results, reports and reviews (FDA). The breadth of this list reflects an evidentiary logic: each item removes one alternative explanation for an anomalous result.
Content of a control log entry
The CDC’s Quality Control Log Instructions specify that a QC procedure should state the controls used, how testing is performed, how often it is performed and what actions are required when results are unacceptable (CDC). The CDC’s example results logs capture facility and test identification, date, sample number or patient identifier, result, lot number, expiration date, reportable range where applicable and the performer’s initials. The following table summarizes the convergence of several federal and professional sources.
| Element | FDA lab manual | CDC QC log | Typical purpose |
|---|---|---|---|
| Date and performer | Required | Date and initials | Attribution and sequencing |
| Lot and expiration | Reagents, media, reference materials | Lot number, expiration date | Excludes material failure as a cause |
| Result | Results and reports | Result, reportable range | Primary evidence |
| Equipment, calibration | Required | Facility and test identification | Links results to instrument state |
| Review | Reviews documented | Corrective action documented | Demonstrates oversight |
Domain-specific variants
Sector rules add particular fields. USDA FSIS describes HACCP records as including monitoring of critical control points and critical limits, with actual times, temperatures or other quantifiable values rather than a bare indication of compliance (FSIS recordkeeping guide). For sanitation, FSIS requires documentation of the daily implementation and monitoring of the Sanitation SOPs and of corrective actions, authenticated by the responsible employee’s initials and date (FSIS).
Field work is treated analogously. The EPA’s guidance on quality assurance project plans gives a field-record example comprising personnel, sample number, collection points, maps or diagrams, equipment or method, climatic conditions, unusual observations and dated, signed entries in a bound notebook (EPA QA/G-5, 1998). The same document cautions that the extent to which its elements are addressed depends on the project and may range from not applicable to extensive documentation. Proportionality is therefore a design principle, not an exemption.
The process history view
Professional laboratory literature adds a longitudinal perspective. Westgard QC describes a useful process history as including the method, initial performance data, the analyst, observed control values, decisions made about those values, problems, corrective actions and supervisory review. It further recommends retaining individual QC values, summary statistics, troubleshooting notes, corrective-action follow-up, proficiency-testing results, maintenance, calibration, reagent changes and instrument function checks (Westgard QC). The implication is that a log recording only numerical results is structurally incomplete, since the surrounding events are what make the numbers interpretable.
Creating Records Correctly
Legibility and reproducibility
The FDA manual requires that work be recorded legibly so that another competent person can repeat it from the record without additional explanation (FDA). This reproducibility criterion is a stricter standard than readability. It implies that method references, parameters and deviations appear in the record itself rather than residing in the memory of the analyst.
Authentication and review
Records should identify and date their author, authenticate the activity performed and provide for review of entries. Reviewer sign-off is not decorative: it is the mechanism by which an error in the first-line entry is intercepted before it propagates into a report. Where the review interval is long, errors accumulate unnoticed, and a recurring fault may span many entries before detection.
Correcting an entry
The FDA’s prescribed correction method for paper records is a single line through the erroneous entry, replacement information, and the correcting person’s initials and date. Erasure, overwriting and obliteration are prohibited because each destroys the original value on which the audit trail depends (FDA). Electronic systems should preserve the original value and record who changed it, when, and why.
Acting on a failed control
The CDC instructions direct that when quality control fails, the user troubleshoot likely causes such as expiration dates and storage conditions, retest the control and document the corrective action; patient results are reported only after acceptable QC (CDC). This model couples documentation to decision control: the log is both a record of the process and a gate on its output. A failed control with no documented response is, in an audit, indistinguishable from a failed control that was ignored.
Storage, Access and Control of Records
The FDA manual directs that originals and audit trails be preserved, that electronic records be backed up regularly with access restricted, and that records be stored securely in clean, labeled locations (FDA). Each requirement addresses a distinct failure mode: loss, unauthorized alteration and physical deterioration or misfiling.
A more formal example is the EPA Region 4 control-of-records procedure, which covers the identification, collection, indexing, access, filing, maintenance and disposal of quality and technical records generated under its laboratory quality-management system. It specifies that project files receive a project number before work begins, that associated technical records be uploaded to a controlled library, and that completed files be reviewed before transfer to the records repository (EPA Region 4). The sequence is instructive for organizations of any size: assign identity first, collect into a controlled location during the work, review at closure, then archive.
A related operational concern is version integrity, since a record that exists in several divergent copies cannot serve as evidence. Readers designing a numbering and approval scheme may find the discussion of document version control for teams relevant to this problem.
Retention Periods and What Determines Them
No single retention period applies to quality control records. The period is determined by the rule, program or contract under which the record is generated.
- EPA programs. Technical records are retained according to the EPA records schedule dictated by each program’s requirements; the schedules classify records as temporary or permanent and provide disposition instructions (EPA Region 4).
- FSIS sanitation records. Under 9 CFR 416.16, records are retained at least six months, kept at the establishment for 48 hours after completion and, if subsequently stored off-site, made available to FSIS within 24 hours (FSIS).
- Project-level planning. The EPA guidance holds that a quality assurance project plan should identify the required records, reporting formats, applicable disposition requirements, and the location and length of retention (EPA QA/G-5).
The practical consequence is that retention should be decided before the first record is created and written into the governing procedure. Where several rules apply to the same record, the longest governs. Organizations in sectors not covered by the sources above, such as medical device or pharmaceutical manufacturing, should consult the applicable regulation directly and, where contractual obligations exist, the contract.
Using Logs to Detect Recurring Problems and Support Audits
Records have a second function beyond proof: they are an analytic resource. Westgard QC characterizes records as the historical voices of a process, and notes that reviewing control results alongside maintenance, calibration, reagent and corrective-action history makes recurring causes easier to recognize (Westgard QC). A drift that coincides with each new reagent lot, or a failure that follows each instrument service, becomes visible only when those events are recorded in the same chronology as the results.
From the inspector’s side, the FSIS inspection model treats record review as evidence that procedures were actually implemented, monitored at the required frequency and followed by appropriate corrective action (FSIS). Three questions therefore organize most audit-ready review:
- Was the control performed at the frequency the procedure specifies?
- Was every out-of-limit result followed by a documented response?
- Was the response itself reviewed and closed?
A periodic internal review that asks these questions of a sample of entries identifies gaps before an external party does. Where gaps are found, a structured register with named owners is more effective than informal reminders; the article on compliance gap analysis with a one-row register describes one such arrangement.
Common Defects That Make Records Unreliable
The FDA manual’s prohibitions double as a catalogue of failure modes: illegible work, missing dates or record authors, erasures or obliterations, unexplained discarded data, inadequate audit trails, and failure to back up records regularly or restrict access (FDA). Each can be traced to a process cause.
| Defect | Typical cause | Preventive control |
|---|---|---|
| Missing date or initials | Batch-completed entries at end of shift | Contemporaneous entry; review checklist |
| Overwritten values | No defined correction method | Single-line correction with initials and date |
| Unexplained discarded data | Retests without justification | Require reason for every repeat or rejection |
| Gaps in corrective action | Failure treated as a number | Mandatory action field linked to out-of-limit results |
| Uncontrolled copies | Local spreadsheets | Controlled library with restricted access |
The recurring theme is that unreliability usually originates in workflow design rather than individual carelessness. A form that does not demand the lot number will, over time, omit it.
Our Perspective
We observe, in the operational and security reviews we conduct, that records fail most often at handoffs rather than at the bench: the entry exists, but the link between the result, the contract or specification that governs it, and the person accountable for the decision is held in someone’s inbox. Our view is that a quality record should be designed backward from the question an auditor, a customer or a regulator will eventually ask, and that the cheapest moment to capture the answer is at the time of the work.
We also hold that proportionality is a leadership decision. Teams under pressure shorten documentation first, so the minimum fields should be few, unambiguous and enforced by the tool rather than by memory. This is the reasoning behind Project Relay, which we built to manage production records and contract information in one controlled place, and behind our technical systems and risk reviews, which test whether record-keeping controls hold up when conditions are not ideal.
— Jesse Hart
Where Project Relay and Technical Review Fit
Teams that keep production records and the contracts that govern them in separate files often discover the inconsistency only during an audit. Project Relay is our subscription tool for managing production records and contracts, priced at $49 per month. Organizations that first need an independent assessment of how their records, systems and controls actually behave may prefer a technical systems and risk review, which evaluates the surrounding infrastructure as well as the documents.
FAQ
What is the difference between a quality control record and a quality control log?
A quality control record is any documented evidence of quality activity, including audit reports, management reviews, corrective actions, data, worksheets and test reports. A log is the chronological ledger of control results within that body of records, typically carrying the date, test, lot, result and performer. Logs are therefore a subset of technical records, and they feed the control graphs and summaries used in review.
How should a mistake in a paper log be corrected?
The FDA laboratory manual prescribes a single line through the error, entry of the replacement information, and the initials and date of the person making the correction. Erasing, overwriting or obliterating the original value is prohibited because it destroys the audit trail. If the reason for the change is not obvious, it should be stated beside the correction.
How long must quality control records be kept?
The period depends on the governing rule or program. FSIS sanitation records under 9 CFR 416.16, for example, must be kept at least six months, while EPA technical records follow the retention schedule assigned to each program. The applicable regulation or contract should be identified before records are created and the period written into the procedure.
What should be done when a control result is out of range?
The CDC instructions direct the user to troubleshoot likely causes, such as expiration dates and storage conditions, retest the control and document the corrective action. Patient results are reported only after the control is acceptable. The log entry should therefore show the failed value, the cause investigated, the retest result and the person who reviewed the outcome.
Do small operations need the same level of documentation as regulated laboratories?
Not necessarily. EPA guidance states that the extent of documentation depends on the project and may range from not applicable to extensive. The identity, date, authorship and traceability of each entry remain advisable at any scale, with additional fields added as the applicable rules require.
Sources
- FDA ORA Laboratory Manual (revised May 15, 2019)
- USDA FSIS recordkeeping requirements
- USDA FSIS Sanitation Standard Operating Procedures
- CDC Quality Control Log Instructions
- EPA Guidance for Quality Assurance Project Plans, EPA QA/G-5
- EPA Region 4 control of records procedure
- Westgard QC: The Records
