Change in Assay Methods Most Likely Responsible for Increased HDL Cholesterol values in NHANES 2003-2006 when compared to NHANES 1999-2002

Researchers are cautioned to interpret trends in HDL cholesterol for NHANES 1999-2006 in view of probable HDL cholesterol method effects.  The HDL cholesterol values showed an average increase of 3.0 mg/dL in NHANES 2003-2006 compared to NHANES 1999-2002. The HDL cholesterol was analyzed in 1999-2002 using two methods - heparin manganese precipitation and a direct HDL immunoassay depending on the participant age and amount of specimen. Most participants in 1999-2002 were measured by the precipitation method. Starting in 2003, all HDL cholesterol samples were analyzed using the direct HDL cholesterol immunoassay method. The heparin-manganese precipitation method and direct immunoassay method for 1999-2000, 2001-2002 and 2005-2006 showed an undesirable bias (>4%) when compared to the laboratory's HDL-cholesterol quality controls (Solomon Park Research Laboratories, Kirkland, WA) that were assigned values established by the Centers for Disease Control and Prevention. The CDC HDL cholesterol reference method uses heparin-manganese to precipitate HDL-cholesterol and the Abell-Kendall method to measure cholesterol. The HDL cholesterol for 1999-2000, 2001-2002 and 2005-2006 were adjusted using: Corrected HDL = [(Solomon Park assigned HDL value) x (Participant HDL)] /(Quality Control HDL value associated with participant sample). The bias for the HDL cholesterol method for 2003-2004 was acceptable (<4%) and the participant results were not corrected. In addition, there was a change in instrumentation in 2005-2006 and there were several modifications of the direct HDL cholesterol method. To control for these differences in methods and instrumentation, the HDL cholesterol was corrected using the Solomon Lab quality controls as described above. 

Despite this correction procedure, all age, gender, and race-ethnicity groups showed an increase in mean HDL cholesterol after 2003. It is most likely that the change from the precipitation method to the direct method in 2003 was responsible for the increase in HDL cholesterol values. Other covariates (body mass index, medications, physical exercise, smoking and alcohol consumption) may explain some of the HDL cholesterol increased values, but it is unlikely to account for the majority of the mean increase in HDL cholesterol. Further investigations will be done to attempt to explain the increased HDL cholesterol values and provide further guidance on the interpretation of HDL cholesterols for NHANES 1999-2006.