Titanium dioxide is a common food additive of increasing interest in dietary intake studies and dietary exclusion studies. Food labelling for titanium dioxide is imprecise so a method was developed for its rapid determination in foods using acid digestion and inductively coupled plasma optical emission spectrometry (ICPOES). Twenty-five foods thought to contain titanium dioxide were obtained. Based on preliminary digestion studies, samples (500 mg) were digested in 18 mol l-1 H2SO4 at 250 degrees C for 1 h and then diluted to 5.9 mol l-1 H2SO4 before determination of titanium by ICPOES at 336.121 nm. Emission intensity was suppressed by H2SO4 so standards were matched for acid concentration. Titanium dioxide embedded in gelatine was used to assess accuracy. A standard reference material of known titanium concentration and six foods of known titanium dioxide content were used as external reference materials. Limits of detection were 2-7.5 ppb, depending on spectral integration times, and the signal was linear up to 5 ppm. Results for all control samples were in good agreement with the expected values. Twelve of the foods contained detectable titanium, ranging from 0.001 to 0.782% by weight, but only eight indicated this on their labels, four being exempt under food labelling regulations. Based on food portion sizes, an individual's daily intake of titanium dioxide could exceed 200 mg from just one of these products. This method may facilitate future studies on titanium dioxide intake, given the present limitations of food labelling.