The Newcastle exercise project - key messages are misleading and conclusions are not evidence based

Mutrie, N. and Woods, C. and Lowther, M. and Blamey, A. and Loughlan, C. (2000) The Newcastle exercise project - key messages are misleading and conclusions are not evidence based. British Medical Journal, 320 (7247). ISSN 1756-1833 (http://dx.doi.org/10.1136/bmj.320.7247.1470)

Full text not available in this repository.Request a copy

Abstract

Article responding to Newcastle exercise project by Harland et al. The authors believe that Harland et al have asked the wrong questions and therefore drawn the wrong conclusions. Harland et al wanted to know whether there was a difference between the various interventions and control in changes in physical activity score from 12 weeks to one year. Since there were no differences, the headline for This week in the BMJ stated that prescription of exercise is a waste of scarce resources. A better question would be to ask whether any group had increased their activity at one year compared with baseline. According to the data Harland et al present in table 2, the percentage of participants who had increased physical activity scores at one year compared with baseline ranged from 23% in the control group to 31% in intervention 3. If these are significant changes from baseline then the conclusion might have been that even the control condition can have a substantial impact in increasing physical activity over one year. Further economic analysis might then determine that the control (which seemed to include the basis of many intervention techniques such as assessment, feedback, and the provision of information) was the most cost effective intervention. The authors' conclusion that brief interventions are of questionable effectiveness is wrong since none of their interventions or even the control condition could be described as brief. In our own research we have shown that much briefer interventions (provision of an information booklet alone) can still increase physical activity up to six months.

CORE (COnnecting REpositories)