ISSUES IN ACCREDITATION
The process: assay-specific considerations Normality of data distribution All assays are assessed with regard to their population distribution. Often, this is influenced by the small number of IQC performed for each lot number due to unavoidable frequent lot number changes. Normality (or a tendency towards normality) is not achieved until n>30 and a Gaussian distribution is observed. Individual lot numbers are assessed. However, a combined result over many lot numbers may be used to increase the validity of the statistical analyses performed. This has a double benefit as the combination of consecutive lot numbers with different target values and performance criteria (acceptable SD) allows assessment of the assay across a range of target values rather than the single value as determined by the single level of control available.
Interpretation and statistical analysis Imprecision, bias, combined and expanded uncertainty of each assay are calculated, with the final reported value of MU being expressed as the expanded uncertainty. The quoted value takes into account a coverage factor of 2 (to equate to ±1.96 SD).
Imprecision uncertainty The random effects of the entire procedure over a defined period of time are termed the imprecision uncertainty. For Top-Down analysis, IQC controls are used at two levels across the clinically relevant range. The performance statistics for these QCs are then able to be used to calculate the imprecision uncertainty using the formulae in many of the references below.
Table 1. Examples of haematology assay categorisation and assays that fall into these categories for the purposes of MU assessment.
Assay
Fully automated assays on calibrated analysers Manual methods usually performed on calibrated
analysers where IQC is available Performed on calibrated analysers where IQC
is not available (numeric result) Methods usually performed on calibrated analysers Manual blood group
where IQC is not available (Pos/Neg result) Assays with pipetting step as the major
source of MU No commercial QC, in-house QC used
Assessment of error type, magnitude and bias Bias, if present, associated with each assay is distinct from MU and must be defined and assessed for its influence in further data analysis. Full descriptions of all types of error (of which bias is one) is outside the scope of this article, but is summarised in Table 2.
Assessment of EQA performance External quality assessment schemes regularly assess the performance of a site in comparison to others. The data from these schemes allow a result to be compared across multiple hospitals, and clinically significant differences or changes in this setting can be assessed. As a measure of proficiency testing, EQA in accordance with ISO/IEC Guide 17043 (Clause 3.7) is considered to include all EQA performed within the haematology department. Some assays are only assessed using EQA
18:03 Page 1
and have no associated IQC. In this scenario, EQA is used as the sole determinant of assay
www.cityassays.org.uk
Infliximab Therapeutic
Drug Monitoring
• Optimising Infliximab clinical use • Measurement of serum Infliximab trough levels
• Anti-Infliximab antibodies with reflex testing from Infliximab results <1µg/mL
For further information contact:
info@cityassays.org.uk or visit
www.cityassays.org.uk Tel: 0121 507 5348
Serum Infliximab Price: £25 Turn round: 3 working days
Anti-Infliximab antibodies Price: £35 Turn round: 7 working days
RBC enzyme assays
performance. Linear regression modelling using EQA performance data has been shown to be an effective means to assess assay bias between centres.10
Bias uncertainty Assay bias and the uncertainty of the reference value (when available) are combined to provide the bias uncertainty as below: Bias uncertainty = ([Reference bias2 [Assay bias/SEM2])1,2
] + where SEM = standard error of the mean.
This is not the same as the quantification of the bias in the form of systematic error, which should be minimised by recalibration (if appropriate). This value is concerned with the uncertainty associated with any bias that exists.
Combined uncertainty Combined uncertainty = (Bias uncertainty2 Imprecision Uncertainty2)1,2
+ Hb electrophoresis HIT screen Impedance aggregometry
Haematology example Haemostasis example Full blood count
Prothrombin time Manual prothrombin time Manual differential
THE BIOMEDICAL SCIENTIST AUGUST 2016
413
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