«

»

Sep 08

How certain can you be: the need for measurement uncertainty

Thousands of routine measurements are made each day. These measurements vary from clinicians making medical diagnoses, to providing evidence to protect our borders, to safeguarding the quality of our water. However, to have a real understanding of the value of any of these measurements you need to know both its quantity and its quality, i.e. how good the measurement really is and can you trust it?

This is demonstrated through the measurement uncertainty associated with each measurement.

Whenever a measurement is made there will always be some level of uncertainty or doubt about the result obtained. This is unavoidable and not due to mistakes in the application of the measurement method, but to the fact that all measurements are subject to variable factors which will contribute to the uncertainty in the result.

For example, when asked how long it takes to get home from work, you might say ‘about 45 minutes’. The ‘about’ indicates that you know the answer is not exactly 45 min – the true answer may lie anywhere between 40 and 50 minutes and depends on the traffic, the weather and a host of other factors. The range associated with the measurement (±5 minutes in this case) is the measurement uncertainty.

Measurement uncertainty allows individual measurement results to be meaningfully compared, for example to see whether a clinical limit has been exceeded or whether results produced before and after a drug intervention are genuinely different. In some cases, such as a quick screening test, a large measurement uncertainty may be acceptable but in others, for example monitoring the amount of a chemotherapy drug present in the body, a small measurement uncertainty is necessary. The impact on human health, the environment and the economy can be significant if measurement uncertainty is not sufficiently accounted for.

mu_thumbnail-of-flyer-1Recently clinical laboratories in the UK adopted the International Organisation for Standardization (ISO) standard 15189 (Medical laboratories – Requirements for quality and competence) which stipulates that measurement uncertainty is to be calculated for each clinical assay.

As the UK National Measurement Laboratory and Designated Institute for chemical and bio-measurement, we have considerable expertise in this field, and have been a leading exponent of calculating, and importantly reporting, measurement uncertainty for all quantitative measurement for many years. In this role we are providing support for clinical laboratories and informing the clinical community of the fundamentals of measurement uncertainty and will be running a specific training course in November to address this challenging topic.

 

If you have any further questions about measurement uncertainty or any of the courses, presentations or guides below please contact us.

 

Training courses

Measurement uncertainty for clinical testing laboratories, specific one-off training event for laboratory analysts and quality managers seeking accreditation to ISO 15189, 2 November 2016

Evaluating measurement uncertainty for chemical testing laboratories, scheduled training course on measurement uncertainty training in line with ISO principles, 12 October 2016

Find further information on LGC training courses on our website or contact us directly.

 

Upcoming presentations

BMSS Introduction to Mass Spectrometry Short Course, Eastbourne, UK, An introduction to small molecule quantitation, Chris Mussell, 13 September 2016 – this will contain a short section on measurement uncertainty

WADA-BIPM Symposium: Standards and Metrology for Anti-Doping Analysis, BIPM, Paris, France, “Bottom up” (GUM) approach to MU assignment for organic analytes, Chris Mussell, 28-29 September 2016

Waters UK Clinical Users Meeting, British Library, UK, Eschewing Obfuscation – Measurement Uncertainty & Mass Spectrometry, Chris Mussell, 19 October 2016

Advances in Clinical Analysis 2016, Chromatographic Society meeting, Burlington House, London, Measurement Uncertainty & Mass Spectrometry, Chris Mussell, 30 November 2016

 

Guides and webinars

Mass spectrometry and measurement uncertainty, Chris Mussell & Simon Cowen, webinar outlining basic concepts and approaches to measurement uncertainty estimation with examples

Evaluating measurement uncertainty in clinical chemistry, guide produced by LGC under the National Measurment System (NMS) that outlines ‘top-down’ approaches to uncertainty demonstrated by clinical analysis case studies

Guide to the expression of uncertainty in measurement, guide produced by the International Bureau for Weights and Measures (BIPM)

Leave a Reply

%d bloggers like this: