Every medicine you give an animal has to be written down, and that entry has to survive on the farm for five years, ready to hand over the moment someone asks for it. Here is where the rule comes from, what it actually demands in practice, and why so many paper systems quietly fail it long before year five is up.
UK medicine treatment records must be kept for a minimum of five years from the date of treatment. This comes from the Veterinary Medicines Regulations, and Red Tractor's own assurance standards reinforce it. Keeping the record isn't enough on its own: it also has to be producible on inspection, meaning you can find it and hand it over on the day it's asked for, not dig it out later.
The retention requirement sits in the Veterinary Medicines Regulations 2013, the statutory framework that governs how animal medicines are supplied, stored and recorded across the UK. It sets out that anyone keeping food-producing animals must keep records of medicines administered, and keep them for at least five years from the date of treatment, even after the animal itself has left the farm.
Red Tractor doesn't invent a separate rule on top of this, but it does give the requirement teeth that the underlying regulation, on its own, mostly lacks day to day. An assessor checking your medicine book against the assurance standard isn't only checking that you keep records, they're checking that five years of them are actually there, actually complete, and actually yours to produce. A farm can be technically compliant with the regulations and still fail an assessment, if the records exist but can't be located when asked.
Five years is also longer than most people expect when they first hear the rule. It's not "keep last winter's book," it's keep every treatment record from a period that, on most farms, spans several staff changes, at least one change of vet practice, and a fair number of house moves, barn clear-outs and new filing habits. That length of time is exactly where the rule gets interesting, because it's long enough that a system which works fine in year one can fail quietly in year four.
This is the part of the rule that catches farms out, because it changes what "keeping a record" means. It isn't enough for an entry to exist somewhere on the holding, in a drawer, a shed, an old notebook in the dresser. It has to be producible, which means findable and handed over there and then, during the visit, for the specific thing the assessor or inspector asks for.
In practice that might be a request for every treatment given to a particular batch of lambs, every use of a specific antibiotic over the last two years, or the withdrawal record for an animal that went to slaughter last month. None of these are vague requests. An assessor names something specific and expects an answer within the visit, not a promise to post it on afterwards. If you can't locate the record in that window, it is treated the same as a missing record, whether or not you're confident it exists somewhere. What has been produced is what counts, not what you believe is true.
This is also, for most farms, the moment the retention rule stops being theoretical. Everyone agrees in principle that records should be kept. The real test is whether a specific record, from a specific date, three years back, can be found in under a minute in front of someone with a clipboard.
Most farms that get caught out on retention aren't being careless. They kept a perfectly good medicine book at the time. The problem is what happens to that book, and the ones before it, over five years of ordinary farm life.
None of this is really about diligence. A farmer who fills in every entry correctly on the day can still lose that entry eighteen months later to damp, a clear-out, or a book that quietly goes missing with someone who's moved on. The rule assumes the record survives the farm's ordinary churn. Paper often doesn't.
The fields that have to last the whole five years are the same ones that have to be in the entry on day one: what the animal was treated for, the product used, the dose given, the date, the withdrawal period, and which animal or group it applied to. None of that changes with age, an entry from four years ago is judged against exactly the same standard as one from last week.
The full detail on what belongs in each entry, and the common gaps assessors flag, is covered properly on the medicine book guide, so this page won't repeat it field by field. What matters for retention is simpler: whatever fields you're capturing now are the fields that need to still be readable, still attached to the right animal, and still findable, five years from now.
The failure modes above have one thing in common: they're all things that happen to paper over time, not mistakes made at the point of recording. That's the specific problem worth solving. A digital record doesn't have pages to lose, ink to fade, or a single person who's the only one who knows where it's kept.
FarmHQ logs each treatment from the same photo that creates the medicine-book entry, and keeps it on the farm's account indefinitely, well past the five-year minimum, rather than archiving or deleting it once the statutory period passes. Because it's searchable by animal, product or date, producing a record an assessor asks for is a matter of a few taps during the visit itself, not a search through boxes afterwards. If a farmer changes phones, brings on new staff, or hands the paperwork to an accountant, the records travel with the farm's account rather than with one person's memory of where the old books are kept. It isn't a dramatic change to what retention requires. It removes the specific ways paper tends to let a farm down over five years.
A minimum of five years from the date of treatment. This comes from the Veterinary Medicines Regulations 2013 and is reinforced by Red Tractor's own assurance standards, which expect the same records to be available at every assessment during that period.
It means the record has to be findable and handable-over on the day it's asked for, not merely stored somewhere on the farm. An assessor or inspector who asks for treatment records covering a specific animal, batch or date range expects them during the visit itself, not posted on afterwards.
It is treated as a non-compliance in the same way a missing record would be, even if you believe the entry exists somewhere in an old book or a drawer. An assessor works with what you can produce there and then, so a record that can't be located in time counts against you.
Yes. The five-year clock runs from the date of treatment, not from when you started using a particular system, so records from before a switch still need to be retained and producible. Export your data before closing an account, or move to a system that keeps history rather than resetting it.