Selected factors have been evaluated in order to determine their influences on the plasma lipoprotein proton NMR spectra of normal and cancer patients. The variables were donor's diet (fasting/non-fasting), temperature and time of sample storage, processing procedure, centrifugation speed, and water pre-saturation time. Plasma samples from fasting individuals that were placed immediately on ice, spun at 1,000 and 3,000 g for 15 minutes, and the proton NMR spectrum acquired with the Carr-Purcell Meiboom-Gill (CPMG) pulse sequence, using a two-second water pre-saturation time, consistently gave reproducible results. Resonances attributed to lactate were minimized under these processing conditions. Centrifugation speed and pre-saturation time did not affect the average line width; however, donor fasting state, processing temperature, and storage time did alter the line width. Most important, blood chemistry analysis revealed an inverse correlation between triglyceride levels and average methyl and methylene line widths. Thus, these factors alone caution against the indiscriminate use of proton NMR spectra to differentiate plasma from normal and cancer patients.