Twenty-Eight Grams of Prevention Is Worth a Half Kilogram of Cure

Abstract

Benjamin Franklin, early American inventor, entrepreneur, author, politician, and diplomat, was known for his pithy expressions. One of his most famous, oftquoted sayings is, "An ounce [28 g] of prevention is worth a pound [454 g] of cure." Less well known is that Franklin's sage advice about prevention was given in the context of his organizing the first fire-fighting company in Philadelphia (1 ). At that time, fires posed an ever-present risk for the residents of Philadelphia and other cities. Meticulous handling of candles, lanterns, embers, and ashes helped avoid property loss, financial ruin, injury, and death. The safety of the populace was inextricably tied to the fire-handling practices of every individual.Today, clinical laboratories are essential components of healthcare systems, contributing invaluable information for the diagnosis, treatment, and avoidance of disease. Decades of research and development have yielded a wide array of sophisticated assays and analyzers. Simultaneously, societal trends and improvements in disease management have fueled an increase in the prevalence of chronic disease, which is driving a clinical demand for long-term consistency in laboratory test results. Assay suppliers attempt to minimize lot-to-lot variation in reagents via manufacturing processes and by conducting lot-release testing to avoid distributing unsuitable lots. Clinical laboratories perform lot-to-lot validation testing to verify a manufacturer's performance claims and assure the ongoing reliability of testing. Yet, practicing laboratorians know that far too much time is spent dealing with inconsistent results.In this issue of Clinical Chemistry, AlgecirasSchimnich and colleagues report multiple failures of lot-to-lot validation procedures to detect significant between-lot differences in an insulin-like growth factor 1 (IGF-1) 3 assay over a 5-year period in 2 laboratories, the Mayo Clinic and the University of Virginia (2 ). During this period, the Mayo Clinic used 32 reagent lots, and the University of Virginia used 16 lots. With every lot change, each laboratory conducted lot-to-lot validation studies according to its standard operating procedures without identifying any reagent lot as significantly different from its predecessor. QC and proficiency testing also failed to identify lot-to-lot inconsistency. Clinicians in both institutions, however, began contacting the laboratories with increasing frequency about IGF-1 results that appeared spuriously high, yet the laboratories were unable to confirm the clinicians' suspicions in lot-to-lot validation studies, QC, or proficiency-testing results. Eventually, the results of an expanded investigation that included a retrospective, longitudinal analysis of lot-to-lot validation results and lot-specific patients' means, medians, and proportions of results exceeding reference interval limits clearly showed significant lot-to-lot inconsistency. Algeciras-Schimnich and colleagues are to be commended for their openness and transparency. Their relevant, compelling...