Assessing Laboratory Method Validations for Informing Inference Across Survey Cycles in the National Health and...

Area of Science:

• Biostatistics
• Health Survey Methodology

Background:

• Laboratory tests on biological specimens are crucial for the National Health and Nutrition Examination Survey.
• Method validation studies ensure data consistency across survey cycles when laboratory methods change.

Purpose of the Study:

• To assess the reliability of statistical methods used for analytic method validation.
• To address concerns regarding the overproduction of adjustment equations in survey data.

Main Methods:

• Simulated 120 analytic method validation studies using 2017-2018 National Health and Nutrition Examination Survey data.
• Employed difference plots, descriptive statistics, t-tests, and Deming regressions for analysis.
• Utilized concordance estimates to evaluate assessment accuracy.

Main Results:

• Adjustment equations were appropriately recommended in 75.0% of studies.
• Over-recommendation occurred in 5.8% of studies, while under-recommendation occurred in 15.8%.
• Sensitivity ranged from 65.9% to 84.4%, and specificity from 74.7% to 97.5%.

Conclusions:

• The current methodology for analytic method validation performs moderately well.
• Underadjustment was more prevalent than overadjustment, indicating a conservative approach.
• Findings support the reliability of current statistical methods for ensuring data integrity.