Обратная связь
Step 8.1 – Implement a comprehensive quality assessment programme
Most modern laboratories have a comprehensive quality management programme that includes external quality assessment and internal QC. These programmes are well developed and accepted for IGRAs, but are often lacking for TB infection skin tests. However, ensuring quality, reproducibility and reliability is as necessary for TB infection skin tests as for any other diagnostic test. Essential elements of a quality assessment system include:
- SOPs, training and documented competency assessment (Area 4) – applicable to both IGRAs and TB infection skin tests;
- instrument verification and maintenance (Area 3) – not applicable to TB infection skin tests;
- method validation or verification (Area 2) – more applicable to IGRAs;
- lot-to-lot testing (Area 5) – not applicable to TB infection skin tests;
- internal QC – applicable for both IGRAs and TB infection skin tests;
- external quality assessment – applicable for both IGRAs and TB infection skin tests; and
- quality indicator monitoring and continuous quality improvement – applicable for both IGRAs and TB infection skin tests.
Step 8.2 – Establish and monitor QC
TB infection skin tests are highly dependent on human skills for proper administration and measurement of induration. Internal QC for TB infection skin tests has been widely neglected and may be a source of error that jeopardizes care in TB infection. The necessary internal QC for skin testing can be achieved using skilled and experienced monitors, but direct monitoring and supervision will require greater investments into human resources if tests are decentralized to peripheral facilities (which is feasible and improves access for individuals). In this situation, digital tools – for example, the mTST approach developed for external quality assessment or artificial intelligence based skin test reading applications – should be used as they become available in the future (see below) (45).
IGRA tests are sensitive to both pre-analytic and analytic errors. Pre-analytic errors include inadequate blood collection, transportation and mixture in tubes; exposure of blood samples to inappropriate temperatures; and excessive waiting time before processing. These variables primarily affect ELISA-based tests that use in-tube blood stimulation, whereas the additional normalization step required for the T-SPOT.TB test can mitigate the impact of preanalytical variables. Analytic errors include inadequate incubation (too short or too long), and errors in the detection and quantification of interferon-gamma. QC of analytic steps aims to detect errors due to test failure, environmental conditions or operator performance. QFT and other ELISA-based tests contain control tubes for negative (i.e. the nil tube) and for positive (i.e. the mitogen tube) results. Definitions of indeterminate IGRA results are detailed in Chapter 3.
For both ELISA- and ELISPOT-based tests, if excessive indeterminate, borderline or equivocal results emerge, the process should be reviewed and staff should be retrained in laboratory technical steps. Also, instruments should be checked and lot-to-lot testing of reagents is warranted in this situation. Pre-analytical steps (e.g. blood withdrawal and transportation) may need revision in case of high rates of indeterminate results or errors.
Step 8.3 – Develop an external quality assessment programme
Although not performed in laboratories, TB infection skin tests need continuous internal QC and external quality assessment. In many countries, TB infection skin test certification is required after complex training and control, but a follow-up quality assessment is not in place. mHealth technologies have been developed to allow quality assessment and QC of health care personnel performance even in remote areas (45). Instruction material is readily available.
For TST and TBST, there is a guide for reviewers performing the TB infection skin test quantitative assessment (46); there are also instructions for health care personnel on how to take mTST photos (47).
Both internal QC and external quality assessment will be particularly useful in countries where TBST will be adopted, but even for the classic tuberculin test, QC and quality assessment are needed to maximize the accuracy of TB infection testing.
IGRAs should also undergo close external quality assessment, in addition to internal QC. External quality assessment may include proficiency testing to determine the quality of the results generated at the testing site through comparison with a reference result. Re-testing of the same samples in different laboratories may also be used and is an approach that is well known by TB laboratory health care personnel. On-site supervisory visits are useful when implementing a new technology.
Step 8.4 – Monitor and analyse quality indicators
Quality improvement is a cyclic process in which data are obtained, recorded and analysed, corrective actions are undertaken and the assessment is repeated, using new data gathered after an intervention. Internal QC and external quality assessment are part of the quality improvement process and should be based on specific indicators. Quality improvement is a wellaccepted part of all laboratory processes and should be an integral part of the performance of IGRAs and CXR. Ensuring that health care personnel are performing the correct techniques for administration and reading of TB infection skin tests is critical for accurate test results, although quality improvement processes are often forgotten for these tests. Quality improvement can be achieved through in-person monitoring and supervision, or virtually.
Quality improvement in TB infection care does not refer only to TB infection testing. The country should establish target indicators for all steps of the TB infection cascade of care; for example:
- how many contacts are expected per index patient (highly dependent on demographical information);
- how many of the contacts should be tested for TB infection (i.e. how much loss is acceptable);
- how many of the contacts are expected to be positive (based also on local epidemiological data); and
- how many cases are lost in further steps of the cascade?
A published report using cascade of care analyses to identify, correct and evaluate problems in Brazil provides a useful example of this health system quality improvement approach (59).