The efficacy and safety of pravastatin and bezafibrate (in retard form) were compared in a randomised double-blind trial comprising 96 patients (48 men, 48 women; mean age 52.5 [20-68] years) with primary hypercholesterolaemia types IIa and IIb. After four weeks' treatment 6 out of 38 patients (400 mg/d bezafibrate) and 27 out of 58 patients (20 mg/d pravastatin) reached a LDL cholesterol level of 190 mg/dl or less. In the other 31 patients of the pravastatin group the dose was raised to 40 mg/d. During the twelve-week course of pravastatin total cholesterol concentration fell from a mean of 364 +/- 75 mg/dl (initial value) to 281 +/- 61 mg/dl (P less than 0.01), while LDL-cholesterol fell from 288 +/- 81 mg/dl to 206 +/- 64 mg/dl (P less than 0.01) and triglyceride concentration from 168 +/- 83 mg/dl to 148 +/- 80 mg/dl (P less than 0.05). During the twelve-week course of treatment with 400 mg bezafibrate total cholesterol concentration fell from a mean of 363 +/- 91 mg/dl to 325 +/- 73 mg/dl (P less than 0.01), LDL-cholesterol level fell from 284 +/- 88 mg/dl to 242 +/- 70 mg/dl (P less than 0.01) and the triglyceride concentration from 173 +/- 91 mg/dl to 121 +/- 83 mg/dl (P less than 0.01). HDL cholesterol concentration rose by 9% in the bezafibrate group and by 8.4% in the pravastatin group (P less than 0.05). Except in the case of HDL-cholesterol, the falls were significantly different in the two treatment groups: pravastatin was superior to bezafibrate in terms of the reductions in both total and LDL-cholesterol (P less than 0.01 for each). However, bezafibrate produced a greater fall in serum triglycerides (P less than 0.05). No serious side effects were associated with either drug.