A provocative finding from several double-blind clinical trials has been the association between greater adherence to placebo study medication and better health outcomes. We used data from the Studies of Left Ventricular Dysfunction (SOLVD) Treatment Trial (SOLVD-TT) and the SOLVD Prevention Trial (SOLVD-PT) to examine whether such associations could be validated and to examine several sources of bias and potential confounding.
Survival analytic methods were used to estimate the association between placebo adherence and several health outcomes, employing a number of modeling techniques to test for the existence of alternative explanations for the association. Higher adherence was defined as having taken ≥75% of prescribed study medication.
Higher placebo adherence was associated with improved overall survival in both SOLVD-TT and SOLVD-PT [hazard ratio (HR) = 0.52, 95% confidence interval (CI): 0.35 to 0.79 and HR = 0.52, 95%CI: 0.38 to 0.71, respectively]. Associations were similar for fatal or non-fatal cardiovascular or coronary heart disease events. Adjustment for both modifiable and non-modifiable cardiac risk factors (including age, gender, diabetes, blood pressure, smoking, weight, alcohol use, and levels of education) had minimal effect on the strength of the association. Little evidence of bias was found as an explanation for this relationship.
In these two trials, better adherence to placebo was associated with markedly superior health outcomes, including total in-study mortality and incident cardiovascular events. No important confounders were identified. These data suggest there may exist strong but unrecognized determinants of health outcomes for which placebo adherence is a marker.