RISK MANAGEMENT Challenge
Lack of a standard definition of a ‘lot’ and inconsistent labelling of material changes.
Variable analytical performance specifications (APS) between manufacturers and laboratories.
Commutability issues with artificial QC materials.
Insufficient sample numbers or range.
Underappreciation of cumulative drift.
Description
Absence of universally accepted criteria for what constitutes a new lot can result in confusion and errors when monitoring changes.
Difference in APS and lack of harmonisation or risk-based adaptation to clinical impact create inconsistency in verification processes.
Artificial QC materials may not behave like patient samples, making it harder to detect clinically relevant lot changes.
Limited patient sample availability may prevent detection of meaningful bias, especially near decision thresholds.
Small changes across several lots can accumulate, resulting in significant clinical impact that may go unnoticed.
Table 1. Challenges with reagent lot verification.
a new reagent or calibrator lot (ISO 15189:2022). ISO 22367:2020 frames this as a preventive control measure, where the verification process should be proportionate to the potential risk of patient harm (ISO 22367:2020). Lot changes can alter bias, precision, or calibration, and the impact may not be detected by IQC if control material and patient samples respond differently to the change.
In high-risk tests such as prothrombin time for international normalised ratio (INR) monitoring or activated partial thromboplastin time (APTT) for heparin therapy, lot verification is often detailed and methodical. This may involve reviewing and, if needed, recalculating the International Sensitivity Index (ISI) for INR, updating the mean normal prothrombin time (MNPT), reassessing therapeutic ranges, and confirming that results remain consistent across the changeover. These activities are supported by the ready availability of patient samples in high-volume testing, making side-by-side comparisons feasible.
Low-volume and one-at-a-time assays
present greater challenges. In these situations, obtaining enough patient material for parallel testing can be difficult, and reagent stability or cost may further limit the scope of verification. ISO 15189:2022 addresses this by allowing lot acceptance to be completed before results are reported, rather than before the lot is used, giving laboratories more flexibility while still protecting result quality. In some cases, laboratories may rely on a manufacturer’s certificate of analysis, but this should only be considered when the certificate provides sufficient analytical detail and the method has a well-documented record of stable performance.
The consequences of inadequate lot verification can be significant. Operationally, an unrecognised bias may
30
lead to an increase in reflex or repeat testing. Clinically, it may result in the inappropriate initiation or withholding of treatment when results cross a decision threshold. The risk is not uniform across all patients; those monitored frequently across multiple lots may be less affected than patients who receive all their testing within a single lot period. Thelen and Loh1,2
have highlighted this unequal
impact, stressing the need to consider patient testing patterns when evaluating the risk posed by lot changes.1 Recent literature has identified persistent weaknesses in current lot verification practices (Table 1). Despite manufacturer validation prior to release, clinically significant drift or shift can evade detection, sometimes first identified by clinicians rather than by routine QC or EQA processes.2,3 Proposed solutions include improved manufacturer transparency, harmonised acceptance criteria for high-impact assays, use of commutable patient- based materials, networked laboratory collaborations to pool samples and data, and integrating lot tracking into EQA schemes.2
– Koh et al.4
Statistical design also matters compared regression-based,
paired-difference, and mixed-model approaches, highlighting the need for method selection based on assay characteristics, available sample numbers, and decision-limit proximity. In some specialised areas such as
serology, the risk landscape differs markedly from chemistry and coagulation. Infectious disease serology assays are qualitative or semi-quantitative, with results based on binding intensity relative to a manufacturer-defined cut-off. As Dimech and colleagues emphasise, lot- to-lot variability is inherent and cannot be eliminated by recalibration.5
Small index
shifts near the cut-off can change result classification without any true change in patient status. Compounding this, regulatory restrictions often require the
use of manufacturer kit controls – which may themselves vary between lots – limiting their usefulness for long-term drift monitoring. In serology, third-party QC materials
offer a means to monitor performance over time, but they must be optimised for the specific assay, have minimal lot-to- lot variation, and be stable for extended periods.5
Best practice is to run such
controls regularly (eg daily before testing, after calibration, or post-maintenance) and to interpret them within acceptance limits that account for normal reagent variation. Historical peer-group data are particularly valuable here, as it helps distinguish normal lot changes from performance deviations that could impact clinical sensitivity or specificity.5 Risk-based lot verification should
therefore begin with identifying the potential harm if a change in bias occurred, estimating the likelihood of that harm, and designing verification steps proportionate to that risk. For high-risk tests, this may mean extensive patient sample comparisons and statistical analysis of bias and precision. For lower-risk tests, a reduced verification plan may be acceptable. For assays with unique performance constraints – such as serology – alternative strategies, including stored sample panels, peer- comparison programmes, and patient- based monitoring algorithms, can be effective where traditional verification is not feasible.2,5
Lot-to-lot verification is therefore more than a technical compliance requirement; it is a targeted control activity that, when applied within a risk-based framework, directly supports the laboratory’s obligation to minimise the likelihood of patient harm under ISO 22367:2020.
Delta checks: detecting undetected drift Delta checks compare a patient’s current result with one or more of their previous results to determine whether the observed change is plausible. They are a form of patient-based quality control (PBQC) and can detect analytical or pre- analytical issues that IQC alone may not reveal. ISO 22367:2020 recognises the value of such tools within a risk-based framework, as they use real patient data to identify patterns that may indicate risk to patient care.
When applied systematically, delta checks can highlight gradual analytical drift, sudden shifts due to calibration changes, or unexpected differences caused by specimen mislabelling or ‘wrong blood in tube’ (WBIT) events. They are particularly valuable where a change in result could lead to immediate
SEPTEMBER 2025
WWW.PATHOLOGYINPRACTICE.COM
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84
