When the Auditor Asks for 90 Days of Field History, What Happens Next?

What Is a Field Inventory Audit Trail?

A field inventory audit trail is the complete, time-stamped record of every movement, restock, transfer, and return involving deployed medical device inventory, by lot, location, and custodian. Under FDA’s QMSR, that record needs to exist continuously, not be reconstructed after an investigator requests it.

The request sounds routine when it’s read out loud. Provide 90 days of movement history for consigned inventory at these five accounts, by lot number, including any transfers between reps. The investigator states it once, expects an answer, and moves to the next line item. For most field operations, that single sentence starts a scramble that has nothing to do with whether the inventory was actually handled correctly, and everything to do with whether anyone can prove it.

The Scramble

It usually starts with a phone call to the rep who covers those accounts, if that rep is still with the company. It continues with someone opening whatever spreadsheet is supposed to track consignment, discovering it was last updated three weeks ago, and cross-referencing it against text messages and emails to reconstruct what actually happened in between. Somewhere in that process, a transfer between two reps shows up that neither spreadsheet nor system ever recorded, because it happened as a phone call and a car trunk exchange in a hospital parking lot. Nobody wrote it down at the time because it felt like a formality, not a regulated event.

A week later, there is a document. It looks complete. It was built backward from memory, texts, and whatever partial system data existed, stitched together by someone who was not present for most of it and is now asserting, in effect, that this reconstruction is accurate.

Why the Reconstruction Is the Actual Risk

This is the part that gets missed. The original inventory handling might have been fine. The rotation might have happened correctly, the transfer might have been appropriate, the condition checks might have occurred. But a record built after the fact, under time pressure, from partial and informal sources, is not the same category of evidence as a record captured when the event occurred. An investigator does not have to prove the underlying process failed. A record that had to be reconstructed is, on its own, a signal that the system of record does not function continuously, which is exactly what QMSR’s process-effectiveness standard is built to catch.

A reconstructed record also invites a harder question than the original request did: if this took a week to assemble, how do you know it’s complete. That question does not have a good answer when the source material was a set of texts and a rep’s recollection.

What the Same-Day Version Requires

The alternative is not a better spreadsheet, it is a different capture model. Every movement, restock, transfer, and condition check has to be logged at the moment it happens, not reconstructed afterward. That means the sort order of a trunk case at restock is a logged event, not an assumption. It means a transfer between reps is captured at the point of exchange, not remembered later. It means a lot-specific pick, the exact box a rep was instructed to use for a given case, is recorded when it happens, so the 90-day history is a query, not a project.

The practical difference shows up in the response time. A same-day export of 90 days of movement history, by lot, by location, by rep, is possible only if that data was captured continuously in the first place. A multi-week reconstruction is what happens when it wasn’t, regardless of how good the underlying operational discipline actually was.

The Standard an Investigator Is Actually Applying

Under QMSR, the test is not whether a procedure exists describing how consignment movement should be tracked. It is whether the record produced on request demonstrates that tracking happened, continuously, without needing to be assembled after the question was asked. A same-day export answers that question directly. A reconstructed document answers a different question, one about how good the team is at recovering from a gap, which is not the same thing as showing the gap never existed.

Frequently Asked Questions

• What does FDA typically ask for during a field inventory review? A time-stamped movement history for consigned or field-deployed inventory by lot number, location, and custodian, often covering 60 to 90 days or more.
  
  

• Why is a reconstructed record considered risky? Because it signals the underlying system of record does not function continuously. A record assembled from memory, texts, and partial spreadsheets after a request is made is treated differently than a record captured when the event occurred.
  
  

• What’s the difference between a same-day export and a reconstructed report? A same-day export is a query against data captured continuously as it happened. A reconstructed report is compiled after the fact from incomplete or informal sources, and it can invite further scrutiny about completeness.
  
  

• How can manufacturers prepare for this kind of request before it happens? By capturing every restock, transfer, and condition check at the point it occurs, so the movement history already exists as a query-able record rather than a project that starts when the request arrives.
  
  

• Does this same 90-day exposure apply to consignment and loaner inventory, or only company-owned stock? Yes, and consignment is often the higher-risk category. FDA’s request doesn’t distinguish by ownership model, consigned inventory at a hospital and company-owned trunk stock are both field-deployed product subject to the same movement-history expectation. Consigned custody sits with the account between rep visits rather than with the rep directly, and that handoff outside the rep’s normal routine is one of the least documented movements in the entire chain.