The effect of the dosing vehicle (e.g., dough) on the ability of an in vitro gastrointestinal (IVG) method to predict relative bioavailable Pb associated with soil ingestion was evaluated. Bioaccessible Pb determined by the IVG method was compared with relative bioavailable Pb measured from dosing trials using juvenile swine for 18 contaminated soils ranging from 1270 to 14200 mg Pb kg(-1). Bioaccessible Pb was measured in the IVG gastric extraction (GE) and intestinal extraction (IE) solutions. Mean bioaccessible Pb values were 32.2% for GE without dough, 23.0% for GE with dough, 1.06% for IE without dough, and 0.56% for IE with dough. It is possible that phytic acid associated with the dough addition decreased bioaccessible Pb. In vivo relative bioavailable Pb ranges for different swine tissues were 1 to 87% for blood, 0 to 110% for liver, 1 to 124% for kidney, and 0.04 to 94% for bone. Strong linear relationships between IVG GE Pb with dough (r > 0.76, P < 0.0002), IVG IE Pb with dough (r > 0.56, P < 0.015), and IVG GE Pb without dough (r > 0.81, P < 0.0001) and in vivo bioavailable Pb as estimated with blood, kidney, liver, and bone were found. Inexpensive in vitro methods may be useful in providing an estimate of the variability in relative bioavailable Pb at a single study site. The IVG method can be used to estimate relative bioavailable Pb, As, and Cd in contaminated soil.