Is recruitment more difficult with a placebo arm in RCTs?
| Autor: | L. Forbes, S. Chinn, J. Figueroa-Munoz, P. Seed, A J Welton, M R Vickers, J A Cooper, T W Meade, T M Marteau |
|---|---|
| Rok vydání: | 1999 |
| Předmět: |
Random allocation
Appointment time medicine.medical_specialty business.industry General Engineering Alternative medicine General Medicine Placebo law.invention Randomized controlled trial Transgender hormone therapy law Intervention (counseling) Physical therapy General Earth and Planetary Sciences Medicine Alternate weeks business General Environmental Science |
| Zdroj: | BMJ. 319:853-853 |
| ISSN: | 1468-5833 0959-8138 |
| DOI: | 10.1136/bmj.319.7213.853b |
| Popis: | Editor—Welton et al found that the proportion of women who were willing to participate in a randomised controlled trial of hormone replacement therapy was higher if participation in a trial without a placebo arm was offered rather than participation in one with a placebo arm, although the difference was of borderline significance.1 Their conclusion was that inclusion of a placebo arm may reduce patients’ willingness to participate in a trial. We do not believe that this conclusion is justified. Allocation of participants to either of the comparison groups was by the woman’s choice of appointment time: the nurses offered participation in trials with and without placebo groups in alternate weeks. This is not random allocation, in which only chance can influence which group a participant is allocated to.2 In this study the nurses could predict in which weeks they would be offering participation in trials with or without placebo groups and so could have influenced the choice of appointment time. This kind of bias would have been avoided by randomising the weekly blocks, the intervention for that week being revealed to the nurses only after all appointments for that week were made. The nurses’ explanation of the reason for using a placebo could have influenced willingness to participate. We would have been reassured that nurses’ explanations were consistent and neutral by knowing that rates of willingness to participate across practices were similar. A lower rate of willingness to participate in some practices would imply differences between nurses in the way they described the reasons for inclusion of a placebo—that is, that it is not the placebo itself but the way in which it is explained that influences the decision to participate. This would make the study’s conclusion invalid. The study shows only that the use of placebos may influence rates of willingness to participate, not actual participation rates, in a trial of hormone replacement therapy. We are concerned that this paper may be used to justify not including a placebo in randomised controlled trials. It is important to include a placebo when the only treatments available are of unknown efficacy or safety.3 The real challenge is to develop ways of effectively explaining the reasons for using a placebo to potential participants when a placebo is ethically justified. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|