Blinding effectiveness and association of pretreatment expectations with pain improvement in a double-blind randomized controlled trial Judith A. Turner a,b, * , Mark P. Jensen b , Catherine A. Warms b , Diana D. Cardenas b a Department of Psychiatry and Behavioral Sciences, Box 356560, University of Washington School of Medicine, Seattle, WA 98195-6560, USA b Department of Rehabilitation Medicine, Box 356490, University of Washington School of Medicine, Seattle, WA 98195-6490, USA Received 11 December 2001; accepted 27 February 2002 Abstract Patient, provider, and clinical investigator expectations concerning treatments are believed to play important roles in patient response. This study examined the association of patient and research nurse/physician pretreatment expectations of pain relief with actual pain relief, the accuracy of patient and research nurse guesses about patient medication assignment, and changes in research nurse and patient pain relief expectations over the course of a randomized double-blind trial of amitriptyline versus an active placebo for patients with chronic pain and spinal cord injuries (SCI). Patient expectations of pain relief with amitriptyline were associated significantly with actual pain decrease for patients in the amitriptyline, but not placebo, condition. Research nurse/physician expectations did not predict patient pain relief. Both patients and the research nurse were able to guess patient medication assignment at a rate significantly greater than chance. The research nurse’s, but not the patients’, expectations of pain relief with amitriptyline decreased significantly over the course of the study. These findings have implications for future randomized controlled trials. Fully double-blind conditions are very difficult to achieve, and it is informative to assess patient and research clinician expectations and guesses regarding medication assignment. q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Expectations; Blinding; Randomized controlled trial; Nonspecific effects; Pain 1. Introduction The role of nonspecific treatment effects (those attributa- ble to factors other than specific active components of the treatment; also referred to as placebo effects) in the outcomes of patients receiving interventions for pain has been the subject of much interest and controversy. Patient and health care provider expectations concerning treatments are believed to be particularly important (Turner et al., 1994; Turner, 2001). Studies conducted more than two decades ago found that patients’ expectations concerning the effects of a medication were significantly associated with their responses to that medication (Luparello et al., 1970; Macdonald et al., 1980). More recently, clients’ posi- tive expectancies of treatment have been shown to be asso- ciated with improvement in psychotherapy studies (Safren et al., 1997). Provider expectations related to study medica- tions have also been found to be associated with patients’ responses to treatment (Shapiro et al., 1954; Gracely et al., 1985). Surprisingly, however, little research has prospectively examined whether patient and clinician expectations of degree of pain relief with the study treatment predict actual pain relief. One of the few such studies found that physician, but not patient, expectations of amount of pain amelioration immediately following a pain-relieving procedure were associated significantly with patients’ actual pain relief (Galer et al., 1997). A recent study of patients with low back pain randomly assigned to receive acupuncture or massage therapy found that patients with higher pretreat- ment expectations for the treatment they received were more likely to have improved function after treatment (Kalauokalani et al., 2001). Further examination of the asso- ciations of patient and provider pretreatment expectations of pain relief with patient outcomes in different patient popu- lations and with different treatments is needed to increase scientific understanding of the role of such expectations in pain treatment response. It is also important to study expec- tations of other study personnel, such as research nurses, who have contact with patients in clinical trials. Concern has been expressed that research nurses in such trials might influence patients’ responses by behaviors reflecting expec- tations based on their experience of other patients’ reactions Pain 99 (2002) 91–99 0304-3959/02/$20.00 q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. PII: S0304-3959(02)00060-X www.elsevier.com/locate/pain * Corresponding author. Tel.: 11-206-543-3997; fax: 11-206-685-1139. E-mail address: jturner@u.washington.edu (J.A. Turner).