A pragmatic approach to sample acceptance and rejection

Sample rejection due to preanalytical errors is very common in many medical laboratories worldwide, though the decision when and how to refrain from analyzing such samples is handled very heterogeneously. As a rational, it is mostly stated that it is done to prevent the patient from being harmed by wrong medical decisions based on such values. But when thinking of the consequences of laboratory results instead of their quality being important per se, the rejection of preanalytically altered samples might be harming the patient by the need of re-collection and timely delay. The importance of the result is never the result in itself, but the consequences the result will have for the treatment or monitoring of the patient. Then again, these consequences are only foreseeable if the specific clinical situation or the general clinical setting for the respective parameter is known. Based on this mindset it can be necessary to modify general performance specifications into "personalized" performance specifications for each laboratory parameter, which can only be achieved in close interaction with clinicians. In this opinion paper we want to present a pragmatic approach to the development of such performance specifications by extending the recommendations of the 1st Strategic Conference of the EFLM on 'Defining analytical performance goals 15 years after the Stockholm Conference on Quality Specifications in Laboratory Medicine', depending on the availability of clinical information, data on biological variation or state-of-the-art recommendations, thereby redirecting laboratory medicine to a more patient focussed profession.