The concept of 'Average of normals' (AON) complements the function of conventional quality control (QC) in medical laboratories. With it, malfunction of instruments is detected early and during routine workflows by using the average of measured patient results as indicator. Using this method reduces the number of samples that have to be measured again and also the number of QC runs. Hence, patient safety is increased and costs are reduced.
Supplement
This function is added to an already implemented middleware. Existing QC functionalities are used as template to ensure easy adaption by the lab personnel. During the project, the statistical and scientific background is analyzed and the implementation concept designed. Since it is a larger extension of functionalities, changes and additions need to be implemented in the back-end as well as in the front-end.
Subject description
To set the required parameters, AON profiles are created for tests/instruments. Here, the used statistical method together with the test frequency, target values and limits for notifications are set. To create a profile that is sufficiently sensitive without producing too many false positive notifications, the patient results that are used can be filtered for external parameters that influence the test result. Filters are e.g. for gender, age, pregnancy, disease and medication of the patient.