Which Records Must a Dental Laboratory Keep for Infection Control in Canada?

Which Records Must a Dental Laboratory Keep for Infection Control in Canada?

Ask a laboratory what records it keeps and the answer is usually “the sterilizer printouts”. That is one record out of eight, and it is the one that is easiest to keep because the machine produces it. The records that decide whether a laboratory survives an inspection are the ones no machine produces: what arrived, what was done to it, and where the case went when it left. A laboratory sits between two parties — the clinic that sent the case and, downstream, the patient who receives the prosthesis — and every record obligation it carries comes from one of those two directions.

The framework is not laboratory-specific, which is why it gets missed. The provincial reprocessing best practices require documentation and maintenance of records for each component of the cleaning, disinfection and sterilization process, based on the manufacturer’s recommendations and established guidelines, and they list the components: collection at point of use, containment and transport, disassembly, inspection, cleaning, disinfection or sterilization, rinsing, drying, reassembly and functional testing, clean and dry storage, clean transportation. They also require that the sterilization process be tested, monitored with results recorded and audited, and that all equipment and devices in each processed load be recorded to enable tracking in the event of a recall. Applied to a commercial laboratory, those sentences become the eight families in Table 1. Where the laboratory orders sterilization consumables, a Class 4 pouch, a Class 5 integrator and a 24-hour biological indicator each arrive with a lot number, and the lot number is the first column of the monitoring log rather than a detail for the purchasing file.

Why does a laboratory need more records than a dental office?

Because a laboratory has no patient record to hang the details on. In an office, much of the reprocessing trail is implied by the chart: this instrument set, this patient, this appointment, this date. The laboratory has a case number and a clinic account, and if the case number is not on the load record, then the load record describes a tray of metal rather than a device on its way to a patient.

The second reason is the handover. The college standard for Ontario dentists, in its laboratory section, requires the office and the commercial laboratory to coordinate so that appropriate cleaning and disinfection are performed on one side or the other, materials are not damaged by over-exposure, and disinfection is not unnecessarily duplicated. That sentence is impossible to evidence without a written record of who did what, which means the laboratory needs a disinfection record per case even when the office did the work, because the laboratory is the party that has to state what it received.

The third reason is that the recall obligation points outward. The provincial recall checklist says that where a recall follows a failed biological indicator, the recall has to include the devices in the failed load and all other devices processed in the sterilizer since the last successfully sterilized load. For a clinic, that is a tray of instruments. For a laboratory, it is a set of devices that may already be in a patient’s mouth, sitting in a clinic drawer, or on a courier van. The records are what make that recall more than an apology.

Schematic titled The record follows the case, showing five record panels for one case between two buildings: receiving record, disinfection record, load record, monitoring log and dispatch record, each listing the fields it has to carry, from the case number and condition on arrival to the load number the case was sterilized in. A caption notes that none of the five can be written honestly after the event. Footer reads clinieco.ca and MDEL 35334.
Figure 1. One case, five records. The receiving, disinfection, load, monitoring and dispatch entries are written at different moments by different people, which is why none of them can be reconstructed honestly afterwards.

What has to be on the case receiving record?

Six fields, and the two that get left out are the ones an inspector asks for. Case number and clinic account identify the case. Date and time received establish when the laboratory’s responsibility began. The number and type of items received is what makes a complete return possible. Whether impressions arrived bagged wet matters because it tells the next person whether the material can still be poured accurately. The condition on arrival — visibly contaminated, already disinfected, cracked model, blood-stained rim — is the field that protects the laboratory when something goes wrong downstream.

The sixth field is who opened the bag. It is a name, not an office stamp. A receiving record signed by someone who was not there is not a record of receipt, and in a dispute about whether an impression arrived clean the dated signature is the entire evidence base. A one-page receiving sheet on a clipboard at the bench, with the case number written before the bag is opened, is enough; a laboratory management system that nobody fills in is not.

How much detail does a disinfection record need?

Enough to repeat the process identically and enough to answer whether the material was damaged. That means the product, the dilution or concentration, the contact time actually elapsed rather than the contact time on the label, the method if the product allows more than one, who did it and when. If the office disinfected the impression before dispatch, the laboratory records that it verified the claim and notes anything visible, which is different from disinfecting it again.

The contact time field is where a laboratory gets caught by its own paperwork. A label that says five minutes and a log that says “disinfected” with no time is a record of an intention. Written as “product X at dilution Y, immersed 10 minutes, 14:20, initials”, the same line answers a question about efficacy and a question about material damage at once. The CliniEco guide to how impressions are disinfected on arrival at a Canadian dental laboratory covers the material side of that decision.

What has to be on a sterilization load record?

The guidance is specific about the load printout: it must be signed to verify that the required time, temperature and pressure were achieved. Everything else on the record exists to connect that printout to devices. Sterilizer identifier, load or cycle number, cycle type, contents traced to case numbers, operator, start and end time, and the chemical indicator results for the load.

Contents traced to case numbers is the field most laboratories do not have. It is also the field that a recall depends on: the recall scope is the failed load plus everything since the last successful one, and “everything” has to be nameable. Where a laboratory runs mixed loads — its own burs and wheels alongside a clinic’s returned prosthesis — the load record has to distinguish the two, because the reprocessing route differs for the two groups even when the cycle is the same.

Table 1. The eight record families a dental laboratory actually owns, the event that starts each one, and the fields without which the record cannot answer an audit question.
Record family What triggers it The fields that make it usable Who signs
1. Case receiving register Every case that arrives from a clinic Case number, clinic account, date and time received, carrier, number of items, whether impressions arrived bagged wet, condition on arrival, who received it. The person who opened the bag, not the office manager at month end.
2. Disinfection record Every impression, prosthesis or appliance the laboratory disinfects Case number, product and dilution, contact time actually elapsed, method (immersion or wipe), whether the office had already disinfected it, any material damage noted. The technician who did it.
3. Sterilization load record Every cycle the laboratory runs Sterilizer identifier, load or cycle number, cycle type, contents traced to case or tray, operator, start and end time, printed parameters attached, internal and external chemical indicator result. The operator who ran the cycle, on the printout.
4. Biological and chemical monitoring log Each day the sterilizer is used, for each cycle type used that day BI lot number and expiry, incubator identifier, placement in the load, result and time read, the control result, the Bowie-Dick result if the unit is dynamic air removal type. The person who read and interpreted the result.
5. Equipment and solution records Ultrasonic testing and solution changes, sterilizer maintenance, incubator checks Date, item, test method or service performed, result, solution concentration or batch, next due date, technician. The person who performed the check; service reports are filed rather than re-typed.
6. Staff training and competency records Induction and each periodic competency check Name, role, what was demonstrated, who assessed it, date, and the date the next assessment falls due. Reprocessing is not a task that can be learned once from a colleague. The assessor.
7. Dispatch record Every case that leaves the building Case number, clinic, date dispatched, carrier and tracking, which load the case was sterilized in, and the items the laboratory disinfected before packing. The person who packed it.
8. Event and recall records A positive BI, a failed cycle, a damaged package, a complaint, a returned case What happened, when it was noticed, the scope identified, who was notified and when, the risk assessment, the corrective action and the date the unit or process returned to service. The person designated in the recall procedure, by name and not by job title alone.

What monitoring records go with the load record?

Three, and they are different records rather than three columns of one. The chemical monitoring record holds the per-package internal indicator and the externally visible Class I indicator, plus the daily air-removal test with a Class II indicator where the sterilizer is a dynamic air removal type. The biological monitoring record holds a result for each day the sterilizer is used and for each type of cycle used that day, plus every load containing implantable devices, with the BI lot number, the incubator used, and the control. The physical monitoring record is the signed printout itself.

Two fields in the biological log earn their space. The lot number with its expiry date is what lets a laboratory telephone a supplier and say that every result from a batch reads unexpectedly; without it, a run of positives looks like a sterilizer problem when it is a test problem. The incubator identifier matters because a 24-hour read depends on the block holding temperature, and the dry-block incubator has its own verification record. A result read on an unverified incubator is a result with an asterisk. The companion guide to storage, shelf life and lot tracking of biological indicators covers what happens when a lot expires mid-month.

What do the equipment and training records have to show?

Equipment records prove that the cleaning and sterilization equipment was performing when the results were recorded. For an ultrasonic bath that means the test results and the solution changes, because a bath that has not been tested is a bath of unknown performance regardless of how the paperwork reads. For a sterilizer it means installation and operation according to the maker’s instructions, preventive maintenance with the actions documented, and the qualification runs that follow a repair. The provincial guidance treats these as part of the reprocessing record rather than as maintenance trivia: preventive maintenance requirements with documentation of actions sit in the same list as the cleaning and sterilization steps themselves.

Training records are the ones an inspector uses to test whether the other seven families are real. If the records show a competent process run by people with no documented training, the inspector will conclude the process works by accident; if they show trained staff and no records, the conclusion is that the records are decorative. What the record needs is modest: who was assessed, on what, by whom, when, and when the next assessment falls due. Orientation and continuing education for everyone involved in reprocessing is what the guidance asks to be documented.

How long does a laboratory have to keep these records?

There is no single national number, and a page that invents one would be doing a laboratory a disservice. The provincial guidance states that records must be retained according to the facility’s policy, and the college standard for dentists defers to the same kind of institutional policy. What that means in practice is that the retention period is a decision the laboratory has to write down and be able to defend, not a figure it can look up.

The defensible way to set it is to work backwards from the longest thing that could require a record. A recall can reach devices that were dispatched to a clinic months earlier; a patient complaint can arrive after a prosthesis has been in use for a year; a device failure investigation can ask what the laboratory did on a particular date. A laboratory that sets its retention period in writing, states the reasoning in one sentence, and stores the records in a form it can search by case number has done what the guidance asks. A laboratory that keeps a box because nobody has thrown it out has not.

Which records belong to the clinic instead of the laboratory?

The clinical ones. The patient’s chart, the appointment record, the radiograph, the treatment plan and the consent all belong to the dental office, and a laboratory that finds itself holding them has usually been sent them by mistake. What the laboratory holds is the case file: what arrived, what was done to it, when it was sterilized, and where it went.

The boundary is worth writing down because it is where a recall gets confused. When a case has to be recalled, the laboratory can name the clinic, the case number, the dispatch date and the load — and cannot name the patient. The clinic turns a case number into a patient and makes the notification. A laboratory that has kept case numbers on every load record can therefore support a clinic’s recall in an afternoon. The CliniEco guide to how cases are packed and tracked between Canadian practices and laboratories describes the same handover from the dispatch end, and the CAD/CAM and resin-handling documentation requirements are set out separately in what documentation applies to CAD/CAM milling and resin handling in a Canadian dental laboratory.

Schematic titled A record, not a claim, showing two columns: what the entry must carry (written when the step was performed, a case or load number on the line, the name of the person who did it, tied to a specific cycle, retrievable by case number in minutes) and what makes it unusable (a week of entries in one pen, one tray of burs with no case number, initials four people could own, a result logged by date only, a box of loose printouts). Footer reads clinieco.ca and MDEL 35334.
Figure 2. The same entry, written two ways. The left column is a record; the right column is a claim, and the difference between them is a device, a person and a load number.

What makes an entry usable when an inspector reads it?

Five tests, in Table 2. Contemporaneous, identifies the device, identifies the person, ties to the load, and retrievable in minutes. Most laboratories pass the first three and fail the last two, because the first three are about writing and the last two are about filing. An inspector who asks for one case and one date and waits twenty minutes has learned something about the laboratory that no amount of correct entries can undo.

Table 2. Five tests, applied to a single line of a log. Most laboratories pass the first three and fail the fourth and fifth.
Test an entry has to pass What fails it Why an auditor cares
Contemporaneous A week of entries in the same handwriting and the same pen, written the Friday before an inspection. A record made at the time is evidence of what happened. A record reconstructed afterwards is evidence that a record was reconstructed.
Identifies the device “One tray of burs” with no case number and no pouch label. Without a device identifier, a recall cannot name what has to come back, and the recall requirement fails on the first step.
Identifies the person Initials that four people on the bench could plausibly own. The point of a signature is to reach a person who can describe what they did. An unreadable pair of initials is a placeholder.
Ties to the load A biological indicator result logged by date only, in a laboratory that runs six cycles a day. A result that cannot be attached to a specific load cannot release that load, and cannot recall it either.
Retrievable in minutes A box of loose printouts in the maintenance cupboard, unsorted. An inspector asks for one case and one date. A record that exists but cannot be found within a working session functions like a record that does not exist.

The fix for the fourth and fifth tests is structural rather than cultural. A load number that is written on the pouch label, on the case sheet and on the printout at the time of the cycle makes the tie automatic; a filing convention of one folder per month, sorted by load number, makes retrieval a matter of seconds. Neither costs money. Both fail on the day the laboratory is busy, which is exactly the day a batch of records gets filled in from memory.

What should a laboratory do about cases that have no records?

Not backfill them. A record created after the fact and dated as if it were contemporaneous is worse than a gap, because it converts a documentation weakness into a credibility problem, and an inspector who notices one reconstructed page will treat the whole log as unreliable. The defensible handling is to open a file, describe the gap in it, set out what is known, and begin the correct process from that date forward.

The honest version of that statement is short: “For cases received before [date] we have no receiving or load records; from that date the eight families in this file are kept, and retention is set out in our written policy.” A laboratory that says that has disclosed a gap. A laboratory that produces a tidy log for a period with no records is making a claim it cannot support. The same principle applies to a failed cycle: the event record is written when the event happens, and the companion guide to the six-step investigation protocol and its documentation duties set out what that record has to contain.

Two closing boundaries. The provincial reprocessing guidance sets the framework and leaves retention to facility policy, so a laboratory in a province with its own college or ministry requirement should treat that requirement as the floor rather than this page. And a record of a reprocessing step is not a statement that a device is sterile; the evidence for that is the monitoring chain, not the paperwork about it, which is why the auditors’ questions collected in the expert Q&A on what Canadian infection control auditors ask for almost always begin with a record and end with a printout.

Primary sources. The record-keeping components, the monitoring requirements and the recall scope quoted above come from the provincial reprocessing guidance, the college standard for Ontario dentists and the provincial recall checklist; the retention period is a facility decision and the provincial documents say so rather than setting a number.

Where does a dental laboratory take this next?

Three of the tools here are usable rather than readable. The printable compliance log carries the fields a receiving check, a lot trace and a reprocessing record actually need, so the paperwork can start without inventing a format. The clinic supply planner turns a month of case volume into pouch, indicator and integrator quantities, so the bench is never restocked mid-cycle. The sterilization monitoring calculator works out how many indicators a year of cycles consumes, and the provincial self-check lists what an inspector asks for in each province. Multi-site or purchasing-group enquiries go through the wholesale account route. Reselling? become a distributor.

Questions that go past a template — a record format a particular inspector asked for, or a retention period your college is challenging — belong with the compliance consultation desk rather than a generic contact form.

If you would rather test the routine than read about it, the 24-hour biological indicator sample request is the simplest way to confirm that the incubator, the records and the staff habit all work together. The clinic consumables hub collects the rest of the guides for this setting.

Related Reading

Frequently Asked Questions

Does a dental laboratory have to keep patient records?

No. Patient records — chart, appointment history, radiographs, treatment plan, consent — belong to the dental office. A laboratory keeps a case file: what arrived and in what condition, what disinfection was performed, which load the case was sterilized in, and when and to whom it was dispatched. That is enough for the laboratory to support a recall; it is the clinic that turns a case number into a patient and makes the notification.

How long must a dental laboratory retain sterilization records in Canada?

The provincial reprocessing guidance does not set a number. It states that records must be retained according to the facility’s policy, so the laboratory has to choose a period, write it down with a one-sentence rationale, and be able to retrieve records by case number for that period. Working backwards from the longest event that could require a record — a recall of dispatched cases, a complaint about a prosthesis in use, a device investigation — is the defensible way to set the interval.

What has to be on a sterilization load record?

At minimum the sterilizer identifier, load or cycle number, cycle type, contents traced to case or tray, operator, start and end time, the printed physical parameters with a signature confirming that the required time, temperature and pressure were reached, and the chemical indicator results for the load. The contents field is the one that makes a recall possible, and it is the field most laboratories are missing.

Do biological indicator results have to be recorded per load or per day?

The testing requirement is per day and per cycle type: a biological indicator each day the sterilizer is used and for each type of cycle used that day, plus every load containing implantable devices. The record should still be tied to a specific load, because a result that cannot be attached to a load cannot release it or recall it. In a laboratory running several cycles a day, a date-only log does not meet that standard.

Should a laboratory write records for cases that arrived before it had a record system?

No. Reconstructing a contemporaneous-looking record after the fact turns a documentation gap into a credibility problem, and an inspector who finds one reconstructed page will doubt the whole log. The defensible approach is a written statement in the file describing the gap, what is known about that period, and the date from which the full record set is kept — with the retention policy attached.

What is the difference between a disinfection record and a sterilization load record?

A disinfection record covers items that cannot be sterilized — impressions, prostheses, articulators — and records the product, dilution, contact time and person. A sterilization load record covers a sterilizer cycle and records the sterilizer, cycle, contents, operator and the physical, chemical and biological results. A laboratory needs both, because the two processes answer different questions and neither one evidences the other.

Who signs the records in a dental laboratory?

The person who performed the step, not the office manager and not a job title. The receiving record is signed by whoever opened the bag, the load record by whoever ran the cycle, the biological log by whoever read the result. A signature that belongs to nobody in particular cannot be traced to someone who can describe what happened, which is the only reason to ask for one.

What records does a positive biological indicator create?

An event record, opened when the result is read rather than when the inspection is due: what was noticed and when, the scope identified — the failed load plus everything processed since the last successfully sterilized load — who was notified and when, the risk assessment, the corrective action, and the date the sterilizer returned to service with the requalification evidence attached. The records written before the event are what make the scope nameable.

CliniEco Medical holds Health Canada MDEL #35334. The record families and monitoring fields described here follow the provincial reprocessing guidance and the college standard that set them; retention is a facility decision and a specific laboratory’s obligation may be set higher by its province or college. This page is general information about reprocessing equipment and consumables, not clinical or legal advice. Follow the instrument and sterilizer manufacturers' validated instructions; where this page and a maker's instructions differ, the maker's instructions govern. Health Canada MDEL #35334 covers our establishment licence; it is not a statement about any individual product.

Already working with clinics? Become a regional representative.