In perioperative and intensive medicine, the metrological requirements for measurements (facts) and measuring instruments (methods) are poorly defined. One of the reasons could be the lack of consensus between physicians and scientific societies where minimum quality criteria apply. Full transparency is required when validating new measuring instruments [6]. Adopting the same understandings and definitions among doctors and scientists is obviously the first step. One solution would be to prohibit and describe these quality concepts (accuracy, accuracy and precision), for which no specified numerical value is given, with reference to « measurement error », « systematic measurement error » and « random measurement error ». Measurement accuracy is the proximity of the correspondence between a single measured value and a real or reference value of the measurement [1]. Accuracy is a quality and cannot be expressed as a numerical value, but is usually evaluated by a measurement error (Fig. 1) [1]. A measurement with a small measurement error is considered accurate [1]. A measurement error can therefore result in the result of a random measurement error (σ; Qualification of inaccuracy), systematic measurement error (bias, qualification of falsehood) or both [1].
Accuracy is the proximity of the agreement between the measured values obtained by repeated measurements on identical or similar quantities under certain and stable conditions [1, 3]. In other words, precision describes the variability of replicated measurements of a given quantity value without reference to a true or reference value (Fig. 1). Accuracy is a quality and should not be expressed as a numerical value, but is usually evaluated by random measurement error. The random measurement error can be expressed as a number by the standard deviation (σ) or variance (σ2) of the repeated measurements and assuming an average random error of zero (Fig. 1, 2). The coefficient of variation (2σ/mean) can also be expressed as variability in %. The specified conditions of the accuracy assessment can add variability of different types [3]. Repeatability is accuracy under conditions that involve the same measurement method, operators, measurement system, operating conditions, location, and replication measurements on identical or similar objects over a short period of time [1]. Reproducibility is accuracy in a range of conditions that include different locations, operators, measurement systems, and replica measurements on identical or similar objects [1]. Between repeatability and reproducibility, intermediate accuracy is the accuracy in a series of intermediate conditions of a measurement (Fig. 2) [1].
Each measure is always fraught with a certain degree of uncertainty. A good understanding of the different types of uncertainties, their name and definition is crucial for the proper use of measuring instruments. In perioperative medicine and intensive care, however, metrological requirements for measuring instruments are poorly defined and are often used inauthentically. The correct use of metrological terms is also crucial in validation studies. The European Union has published a new Medical Devices Directive, which states that for devices with a measuring function, the notified body is involved in all aspects related to the conformity of the product with metrological requirements. It is therefore up to scientific societies to set the standards in their field of expertise. Adopting the same understandings and definitions among clinicians and scientists is obviously the first step. In this metrological review (Part 1), we list and explain the main terms defined by the International Bureau of Weights and Measures in terms of sizes and units, measurement characteristics, measuring instruments, measuring instrument properties and measurement standards, with specific examples of perioperative medicine and intensive care. By analogy with measurement accuracy, instrumental accuracy is the proximity of the correspondence between the indications obtained by replica measurements on equal or similar quantities under certain and stable conditions [1, 3]. Although this is false, the quality of « instrumental precision » is often confused with its associated « quantity », the variability of indications. By analogy with the measurement bias, the instrumental bias is the average of the replica indications minus a reference quantity value [1].
It considers the systematic error of the measuring device. There is no quality associated with instrumental bias, such as « instrumental veracity » in VIM. Since accuracy qualifies a single measurement, this quality cannot be used to describe an instrument. However, the term « accuracy class » is used to describe measuring instruments that meet the specified metrological requirements. such as factories, schools, hospitals, temples, governments, etc. exist to provide products or services to people. What is essential with these products or services is that they are suitable for use and provide excellent service. These products must be well designed with functional perfection and, above all, manufactured directly. If a hospital is well equipped but does not meet your expectations, then the service is not good or the quality of the service is poor.
If a factory produces the cars that cause a lot of problems and do not work properly, then the car is not suitable for use in the true sense of the word or you say that the quality of the car is poor. Whether it is a product or a service, it must be suitable for use or to be adapted, so doctors must strictly share the same terms and definitions with other scientists. This is especially important in perioperative medicine and critical care, as clinical decision-making takes into account or even fully supports a variety of variables measured with medical devices, including advanced hemodynamic and respiratory monitoring. In the erroneous approach (traditional approach, see above), measurement error adds systematic and random errors, but no rules can be derived because they combine for a given measure. The fuzzy approach aims to characterize the dispersion (distribution scheme) of the values assigned to a measured value according to the information used [1].
