Can I help you? Physicians' expectations as predictor for treatment outcome
Supported by the following German Social Health Insurance Companies: Techniker Krankenkasse (TK), BKK Aktiv, Betriebskrankenkasse der Allianz Gesellschaften, Bertelsmann BKK, Bosch BKK, BKK BMW, Daimler-Chrysler BKK, BKK Deutsche Bank, Ford Betriebskrankenkasse, BKK Hoechst, HypoVereinsbank Betriebskrankenkasse, Siemens-Betriebskrankenkasse, Handelskrankenkasse, Innungskrankenkasse Hamburg. The trials were initiated due to a request from German health authorities (Federal Committee of physicians and Social Health Insurance Companies, German Federal Social Insurance Authority). The chair of Claudia Witt is endowed by the Carstens-Foundation.
Conflicts of interest
Patients' expectations of acupuncture treatment have widely been investigated; however, little focus has been on the physicians' expectations. We aimed to investigate (1) which patient characteristics lead to different expectations of physicians, and (2) whether physicians' expectations predict pain reduction and physical functioning in acupuncture and usual care treatment for chronic pain.
In four large multi-centre, randomized trials patients with chronic pain were randomized to receive usual care alone or 10 additional acupuncture treatments. Data were pooled. Baseline characteristics of the three expectation groups were compared, and the physicians' expectation and its interaction with the treatment group were included in two linear regression models predicting pain reduction and change in physical functioning. Other patient characteristics were included for adjustment.
9900 patients treated by 2781 physicians were analysed. Age, education and disease-related variables differed in the expectation groups. There was no interaction between treatment group and expectation. Patients, for whom the physicians had expected substantial improvement, showed more pain reduction (p < 0.001) and better physical functioning (p < 0.001) than patients for whom moderate improvement was expected. No significant differences were found between expected moderate and expected lack of success. However, the proportion of explained variance that was due to physicians' expectations was small considering total explained variance.
Physicians' high expectations at baseline predict better outcome, independent of the treatment. Since we adjusted for several patient variables including duration and severity of disease, this cannot be explained by prognostic factors only. Other explanations are discussed and recommended for future research.