Classic conditioning and placebo effects in crossover studies

OBJECTIVE: To find evidence of classically conditioned placebo effects in a placebo-controlled crossover drug study. Specifically, we tested a prediction of the conditioning model that the placebo response will be greater after drug exposure than before. METHODS: Twenty-four patients with mild to moderate essential hypertension and no contraindications to atenolol participated in the study. The study design required randomized assignment to one of three groups: placebo followed by 50 mg atenolol daily, followed by no treatment, each for 1 week; atenolol followed by placebo; and atenolol followed by nothing (to show residual drug effects). Twice-daily blood pressure measurements were made by patients at home; once-weekly measurements of blood pressure and heart rate were made by a research nurse. RESULTS: Before drug treatment, there were no differences in the antihypertensive responses of patients taking placebo and patients taking nothing (difference, 0.98 mm Hg; 95% confidence interval, -0.98 to 2.93). After atenolol treatment, placebo treatment produced a significantly greater antihypertensive response than no treatment (difference, -6.09 mm Hg; 95% confidence interval, -11.81 to -0.38). Thus the placebo response after atenolol administration was more than a residual drug effect. Similar patterns were observed for heart rate but not for blood pressure readings taken in the office. CONCLUSIONS: These observations are consistent with a conditioning model of placebo effects. These findings warrant further investigation in larger studies and in other disease models.