Behavior Change Techniques in Physical Activity Interventions for Multiple Sclerosis.

Coding the techniques depended on what each study wrote down, and most did not report them directly; the authors note that validation studies show only modest agreement when coding about 20% of the 93 techniques. The studies varied so widely in design and outcome that the review could not work out how large an effect any technique has, or whether it works. The finding that more techniques went with more activity came from an analysis run after the fact, and the authors say it could reflect something else, such as better-funded or better-powered studies rather than the techniques themselves. The review deliberately cast a wide net, and when adherence to the intervention was the main outcome rather than a measure of physical activity, the authors say results should be read with caution. Studies rarely reported why they chose their techniques, or how the techniques matched the theory behind the intervention, which the authors name as the main remaining problem.

The authors state they have no conflict of interest. The text supplied does not name a funder.

Do not read the finding that programs using 15 or more techniques were more likely to report increased activity as proof that adding techniques makes a program work. It came from a comparison run after the fact, and the authors themselves say the link could reflect better funding or larger samples rather than the techniques.