IMPROVING POWER WITH REPEATED-MEASURES - DIET AND SERUM-LIPIDS

The inability to detect associations between diet and serum cholestero l in cross-sectional population studies has been attributed to measure ment error in diet assessments and between-subject variability in lipi d concentrations. Current statistical methods can reduce the effects o f measurement error and allow within-subject comparisons when replicat e measures on individuals are available, even if the time between repl icates is as long as 4 y and replicate data are not available for all subjects. Data from 928 nondiabetic participants of the San Luis Valle y Diabetes Study with measures of 24-h dietary intake and fasting lipi d concentrations at baseline, at a 4-y follow-up visit, or both were a nalyzed in a random-effects model that allowed for an unbalanced desig n. Sex was included as a non-time-varying covariate and age, body mass index, and energy intake were included as time-varying covariates. Th e findings when LDL cholesterol (mmol/L) was regressed on saturated fa t intake (20 g/d) with all observations in a random-effects model (bet a = 0.14, P = 0.0016) were compared with results with observations res tricted to the first visit only (beta = 0.05, P = 0.52), a balanced de sign using averages across visits (beta = -0.12, P = 0.28), and a bala nced design with random effects obtained by excluding subjects without two observations (beta = 0.12, P = 0.0092). Study power was greatest in the random-effects model using all observations and time-varying co variates. These findings highlight the importance of even a single rep licate observation on a sub sample of subjects. We recommend analyzing all data rather than averaging measures across visits or omitting obs ervations to create a balanced design.