Behavioral and Clinical Measures

Patients from all studies completed a general health questionnaire and a VAS (on a 0 to 10 scale) for their knee OA pain. Patients from study 1 and 2 also completed the WOMAC, the BDI, and the PCS. All questionnaires were administered on the day of brain scanning. Response categorization and brain regions of interest were identified using only the VAS measure and then tested for consistency using WOMAC. Thus, WOMAC provided an unbiased estimate of treatment response and of brain regional properties. Analgesic response was defined a priori on an individual basis as at least a 20% decrease in VAS pain from baseline to the end of treatment period; otherwise, subjects were classified as nonresponders. This threshold for analgesic response was chosen based on our earlier results [15] and also based on a recent meta-analysis estimate of the size of placebo analgesia [32]. In study 2, to partially compensate for regression to the mean effects, VAS was measured 3 times over a 2-wk period prior to the start of treatment and after cessation of medication use, averaged, and used as the indicator of pain at entry (designated as baseline in the figures).