Abstract The modifications induced in serum lipids in patients who were instructed only to modify their diet, smoking, and lifestyle were compared with those achieved when such nonpharmacological measures were complemented with small, once-daily doses of pravastatin in patients with primary hypercholesterolemia of mild-to-moderate degree who also had two or more additional coronary heart disease (CHD) risk factors. A total of 150 men and women, aged 21 to 69 years (mean age, 47.77 ± 9.81 years), with borderline and moderate hypercholesterolemia, who also had two or more CHD risk factors, were randomized to receive pravastatin or placebo after being instructed to reduce dietary intake of saturated fat and cholesterol, increase physical activity, and eliminate smoking for at least 4 weeks. The protocol required that patients have total cholesterol levels between 200 and 260 mg/dL, low-density lipoprotein (LDL) cholesterol ⩾ 130 mg/dL, and triglyceride levels <350 mg/dL, at the end of those 4 weeks. Participants were recruited from four clinical centers. Following the run-in period, patients who had the required total and LDL cholesterol levels were randomly assigned in double-blind fashion to a group that received pravastatin 10 mg once daily at bedtime (one tablet) or a group that received one matching placebo tablet once daily at bedtime. If the total cholesterol level after 10 to 12 weeks of treatment had fallen less than 15% or remained at 200 mg/dL or greater, the investigator had the option to decide in a double-blind fashion if the dose of the randomized study medication should be increased to two tablets at bedtime during the last 13 weeks of the study. Of the 150 patients randomized, 105 were eligible for the efficacy evaluation. Most of these received one tablet of the randomized study medication for the duration of the trial; only 5 patients treated with pravastatin and 7 treated with placebo had their medication dose doubled during weeks 14 to 26. Patients instructed to modify their diet, smoking, and lifestyle by itself showed no significant changes in mean plasma levels of total cholesterol, LDL cholesterol, triglycerides or high-density lipoprotein (HDL) cholesterol. The group treated with pravastatin showed significant reductions from baseline in total cholesterol and LDL cholesterol ( P < 0.001), which was apparent at week 6 and was maintained at weeks 13 and 26. A small increase in mean HDL cholesterol levels was recorded, but it failed to reach statistical significance. Triglycerides did not change. After 26 weeks of treatment with pravastatin, reductions from baseline in total and LDL cholesterol were 11.7% and 20.3%, respectively, and the increase in HDL cholesterol was 7.6%. At the end of the trial total cholesterol levels were reduced to <200 mg/dL in 47.8% of the patients treated with pravastatin but only in 18% of those given dietary advice alone; similarly 57.1% of patients taking pravastatin and 21.8% of those receiving placebo reached the target LDL cholesterol level of <130 mg/dL. During the trial 10 of 76 patients randomized to receive pravastatin (13%) and 15 of 74 of those receiving placebo (20%) reported side effects. These were mild and caused discontinuation in only one patient who developed a skin rash while on placebo. No serious adverse laboratory experiences were reported. A cholesterol-lowering diet, explained and monitored in the setting of a medical visit, was insufficient to produce the total and LDL cholesterol levels considered desirable for the hypercholesterolemic patient without CHD but with two or more CHD risk factors. The addition of a small dose of pravastatin once a day considerably increased the proportion of patients who reached the defined therapeutic goals for total and LDL cholesterol. Nevertheless, such an approach does not circumvent the need to improve current methods used by physicians to prescribe and supervise the required diet.