When change is not a good thing.

Abstract

Over the past 40 years, our understanding of the epidemiology and natural history of a number of common chronic diseases has been markedly enhanced by serial, nationally representative, cross-sectional surveys such as the National Health and Nutrition Examination Survey (NHANES)2 (1), longitudinal cohorts such as the Atherosclerosis Risk in Communities (ARIC) study (2), and more disease-specific cohorts including the Chronic Renal Insufficiency Cohort (CRIC) (3) and the Assessment, Serial Evaluation, and Subsequent Sequelae in Acute Kidney Injury (ASSESS-AKI) study (4). Many of these studies have banked biospecimens, which allow tracking of laboratory markers of disease over time. Such stored samples are invaluable resources that can lend insight not only on secular trends in population prevalence of disease but also on how disease appears and evolves over time within a study participant. Both of these require repeated measurements of the same analyte over an extended period of time (years to decades). This seemingly simple concept—repeated measurement of the same analyte over time—can be challenging to implement well. To begin with, the assay method itself may change. For instance, serum creatinine, the most common test to assess kidney function, can be measured using alkaline picrate (“Jaffe method”) or enzymatic assays (5). Even the Jaffe method has been refined over time to better handle interfering substances (6). In addition, there may be drift in the standards used to calibrate the assay, leading to systematic differences. In the case of serum creatinine, the National Kidney Disease Education Program has led efforts over the last decade to standardize testing, including adoption of isotope dilution mass spectrometry (IDMS)-traceable references (7). Thus, calibration of assays performed in different eras may be necessary. Laboratory recalibration via repeat measurement of stored specimens at a later (uniform) point in time assumes stability of the analyte and its concentration …