The significance of subclinical thiamine deficiency in the elderly was determined by assessing response to thiamine supplementation in a randomized double-blind, placebo-controlled trial. Thirty-five of 222 people aged > or = 65 y had two concentrations of erythrocyte thiamine pyrophosphate (TPP) < 140 nmol/L 3 mo apart and 41 other people had the first, but not the second, TPP concentration below this value. Both groups were randomly assigned in a double-blind trial to oral thiamine (10 mg/d) or a placebo. All subjects randomly assigned to receive thiamine showed increases in TPP concentrations compared with control subjects. Only the subjects with persistently low TPP concentrations showed subjective benefits from treatment with improvements in quality of life (measured on a visual analogue scale; P = 0.02) and decreases in systolic blood pressure (P = 0.05) and weight (P < 0.01) when compared with subjects given placebo. There was a trend toward benefits in sleep and energy (P = 0.07). We conclude that a low TPP concentration on two occasions is a better predictor of response to treatment than an isolated measurement. Quality of life was enhanced by providing thiamine supplements. Blood pressure and weight were lower after thiamine supplementation.