GLI GLI Quality Tool
GLI Quality Tool — Version 2.0

A TB lab competency rubric for DST reading and interpretation

Developing a TB lab competency evaluation rubric for DST reading and interpretation makes the expected standard visible, repeatable and fair. Drug-susceptibility testing (DST) results may come from culture-based phenotypic methods, molecular resistance assays, or both. A staff member can perform a test accurately yet still misclassify resistance, overlook a quality-control failure or communicate an uncertain result. The rubric therefore needs to judge the complete chain from reviewing the test output to making a defensible interpretation, not just whether the correct line or colour appears.

Australian laboratories should anchor the rubric in current WHO guidance, manufacturer instructions, local procedures and state or territory requirements. It can fit within the GLI Quality Tool’s four-phase roadmap and 12 Quality Systems Essentials. The approach works for a high-throughput service in Melbourne or Sydney, a regional laboratory in Perth, Adelaide or Darwin, and smaller networks sharing staff across urban and remote areas.

Define the competency and its limits

Build the rubric around a defined task, method and decision. A molecular assay reader needs criteria for resistance and indeterminate results; a culture-based reader must also assess growth, controls and plate quality. Record whether competence covers reading, interpretation, report wording or all three. Competence should be method-specific because experience with one rifampicin assay does not establish competence on every platform used by the laboratory.

List the evidence that will be accepted and the errors that require immediate escalation. Name current WHO technical guidance, the assay manufacturer’s instructions and reference laboratory criteria in the rubric. Where the laboratory holds NATA accreditation to ISO 15189, the standard must also align with validated local procedures. Critical errors can include reporting susceptible results with an invalid control, missing established resistance or changing an interpretation without documented reassessment.

The work area and resources must support the behaviour being assessed. Before staff competence is signed off, managers can use a biosafety cabinet certification schedule to confirm that containment equipment has current certification, defined maintenance intervals and documented corrective action. Imported media, controls or assay cartridges may require biosecurity permits and clearance through Australian Customs, so a delayed import should not be confused with a performance failure. The rubric should record whether required materials were available.

Map the DST workflow and evidence

Trace the rubric through the actual workflow: specimen accession and suitability, batch setup, quality-control review, reading, interpretation, authorisation and report communication. At each hand-off, identify what the staff member must do independently and what a senior reader must verify. This prevents a technically correct endpoint from concealing unsafe work, an undocumented assumption or a missed handover. The same standard should fit direct molecular tests and laboratory-confirmed phenotypic results.

Tailor the depth of assessment to the service. A large metropolitan laboratory may assess reading during routine batches, while a regional network may use remote supervision, fewer challenge samples and scheduled on-site review. Although a DST result is not itself the notification event, confirmed TB diagnoses must follow the relevant state or territory pathway and national reporting through the National Notifiable Diseases Surveillance System. Staff should know when to seek reference-laboratory confirmation rather than resolve an unusual pattern alone.

Collect several kinds of evidence instead of relying on a written test. Combine direct observation, a de-identified challenge panel, routine quality-control data, external quality assessment or proficiency results, and the final report. Keep assessor notes specific: record what was seen, the standard applied, any prompting provided and the action taken. Where two assessors review the same panel, capture their ratings separately before discussing differences.

Score observable performance consistently

Use a short scale with behaviour-based descriptions so assessors judge the same evidence in the same way. The rating should combine the technical decision, the process used to reach it and whether the staff member recognised when help was needed.

Evidence and rating anchors

A score must not conceal a serious weakness. Define whether a critical error overrides the total, which errors trigger notification to the supervisor and how much supervised practice is required after remediation. Use the same thresholds for new employees, established staff and staff returning after prolonged absence. A second list turns the ratings into a personnel decision rather than an average that can obscure a failed safety element.

Before approval, ask two trained assessors to score several sample cases independently. Resolve differences by referring to the written anchors rather than averaging first. Review the rubric whenever the method, reporting rules or reference-laboratory pathway changes, and record the reason for each revision.

Validate the rubric through observation

Pilot the rubric with both experienced staff and people new to DST. Include common susceptible results, known resistance, an invalid or indeterminate case and a situation requiring escalation. Cases should be de-identified or supplied as validated teaching material; routine patient specimens should not become an informal examination. The panel should reflect common workloads and include enough resistant results to test judgement without revealing the source laboratory.

Although smear microscopy and DST require different technical criteria, the discipline of a peer observation program can be adapted for DST reading. A second qualified reader can independently examine selected outputs, blinded to the first reader’s call. Disagreements become a discussion about criteria and evidence, followed by a documented decision when the interpretive standard is unclear. This is particularly useful in smaller Australian networks where internal peer review may otherwise be infrequent.

Run periodic re-checks after initial competence and use blinded panels as part of continuing quality assurance. Trend missed resistance calls, invalid-result acceptance, inappropriate report wording and agreement between readers. Investigate recurring patterns as a training issue, method issue or resource issue. Keep feedback specific and linked to the rubric anchor while protecting patient confidentiality and using quality-assurance processes rather than disciplinary shortcuts.

Build safe and sustainable controls

Decontamination is part of competence because shortcuts can contaminate work areas and compromise both staff safety and future assessments. The autoclave waste decontamination protocol provides a practical basis for validating waste segregation, cycle parameters, loading practices and routine verification, subject to local biosafety, environmental and equipment requirements. Staff should demonstrate how they respond to a failed cycle or damaged container, not merely recall the procedure.

Include safety-critical behaviour in direct observation. Depending on the method, this may cover safe handling of sputum specimens, correct use of personal protective equipment, prevention of aerosol exposure, segregation of sharps and waste, clean-up after spills and response to a laboratory exposure. The rubric should distinguish a correct answer given with unsafe practice from unsafe practice that was not observed but may be inferred from records. The latter should be assessed through another method rather than marked as proven competence.

Resource reliability is also an Australian consideration. Procurement through national suppliers or local distributors may provide strong cold-chain support to metropolitan hospitals, while regional and remote services can face weekly courier schedules, holiday interruptions and temperature excursions. For shipments crossing borders, teams should check biosecurity permits and customs requirements before dispatch. For services receiving specimens from First Nations communities, transport, culturally safe communication and reliable return of results need explicit ownership, especially when follow-up depends on timely clinical information.

Implement, review and improve

Introduce the rubric as a controlled document linked to relevant GLI Quality Tool phase checklists and downloadable materials. Map its evidence to the Quality Systems Essentials for personnel, equipment, documentation, assessment, safety and continual improvement. Assign a rubric owner, train assessors, pilot the ratings and obtain approval under the laboratory’s quality-management arrangements. Version the form, evidence sources, decision rules and review date so each competency decision can be reconstructed.

Monitor whether the rubric changes behaviour and quality over time. Useful measures include assessment completion, critical errors, agreement between readers, time to remediation, proficiency-test performance and recurring report queries. Review results at management meetings and use corrective and preventive action when weaknesses cross several domains. The rubric should remain a learning and assurance tool, with feedback that identifies the exact gap and a practical route back to competence.

Appoint a rubric owner, select representative challenge cases, confirm assessor training and schedule the first blinded assessment within 30 days. Place the completed form, panel responses and remediation evidence with the staff record, then use the findings to strengthen the laboratory’s wider quality system.