Batch Manufacturing Record (BMR): What It Is and What Goes In It
A Batch Manufacturing Record (BMR) is the complete record of how one specific batch of a drug product was made. It is filled in as the batch is produced, and it is the document an inspector reaches for when they want to know what actually happened rather than what was supposed to happen.
In US regulation the governing section is 21 CFR 211.188, Batch production and control records. You will also see it called a Batch Production Record (BPR) or, once it lives in software, an electronic batch record (EBR).
BMR vs master record: the distinction that matters
These two are constantly confused, and the difference is the whole point.
| Master Production Record | Batch Manufacturing Record | |
|---|---|---|
| What it is | The approved template. How this product is to be made. | The executed instance. How this batch was made. |
| Regulation | 21 CFR 211.186 | 21 CFR 211.188 |
| How many | One per product and batch size | One per batch |
| When written | Before production, approved in advance | During production, as it happens |
| Contains | Instructions, specifications, theoretical yield | Actual weights, times, results, yields, signatures |
The master record is prepared, dated and signed by one person and then independently checked, dated and signed by a second. The BMR begins as an accurate reproduction of that master record, checked for accuracy, and is then completed as the batch runs.
That is why "we updated the BMR" is usually the wrong sentence. If the instructions change, the master record changes under change control, and future batch records inherit it.
What 21 CFR 211.188 requires
The regulation asks for complete information on the production and control of each batch, including:
- An accurate reproduction of the master production or control record, checked for accuracy, dated and signed.
- Documentation that each significant step was accomplished — the dates, the identity of major equipment and lines used, the specific identification of each batch of component used, weights and measures of components, in-process and laboratory control results.
- Identification of the people performing and directly supervising or checking each significant step.
- A statement of actual yield, and of the percentage of theoretical yield, at appropriate phases of processing.
- Complete labeling control records, including specimens or copies of all labeling used.
- Description of the drug product containers and closures.
- Any sampling performed, and any investigation carried out under 211.192.
Where batch records actually fail
Almost none of the common findings are about the instructions being wrong. They are about the record not being believable.
Filled in afterwards. The single most damaging finding. A record completed from memory at the end of a shift breaks contemporaneous under ALCOA+, and once an investigator establishes that one entry was back-filled, every entry becomes suspect.
Uncontrolled corrections. A correction should leave the original legible, with the change initialled, dated and, where the reason is not obvious, explained. Overwriting or obliterating an entry converts a small error into a data integrity observation.
Blank fields. An empty box does not distinguish "not applicable" from "not done" from "not recorded". Fields that can legitimately be empty should be marked N/A rather than left open.
Yield outside range with no investigation. The percentage of theoretical yield exists so that a deviation is visible. Recording one outside limits and carrying on is a worse finding than the deviation itself.
Signed by someone who was not there. Supervisor signatures applied in a block at the end, for steps performed on a different shift, undermine the attributability the signature is supposed to establish.
Master record and batch record disagreeing. Where the executed record shows a step, a limit or an equipment ID that the approved master does not contain, one of the two is wrong and neither is defensible.
Electronic batch records
Moving to an EBR does not reduce the requirements; it changes where they are enforced. An electronic batch record is a record required by a predicate rule and kept electronically, which places it squarely under 21 CFR Part 11: validated system, secure computer-generated audit trail, limited access, and signatures that carry the signer's printed name, the date and time, and what the signature means.
The common failure in a hybrid setup is printing the electronic record and signing the paper. That does not take the electronic original out of scope; it just creates two records that have to agree.
Retention
Batch records must be retained for one year after the expiration date of the batch, or for certain OTC drug products not bearing an expiration date, three years after distribution. The retention clock runs from the batch, not from the year the record was written.
Frequently asked questions
What is a batch manufacturing record?
The complete record of how one specific batch of a drug product was manufactured, completed as the batch is produced. In US regulation it is governed by 21 CFR 211.188.
What is the difference between a BMR and an MPR?
The Master Production Record is the approved template describing how a product is to be made. The Batch Manufacturing Record is the executed instance for one batch, beginning as a checked reproduction of that master.
Who signs a batch manufacturing record?
The people who performed and who supervised or checked each significant step, identified individually. The master record it derives from is prepared and signed by one person and independently checked and signed by a second.
How long must batch records be kept?
One year past the batch's expiration date, or three years after distribution for certain OTC products with no expiration date.
What is an electronic batch record?
A batch record maintained in software rather than on paper. It is subject to the same GMP requirements plus 21 CFR Part 11, covering validation, audit trails, access control and electronic signatures.
Can a batch record be corrected?
Yes, provided the original entry stays legible and the correction is initialled and dated. Obliterating or overwriting an entry is a data integrity finding.