Effects of small community walking intervention on physical activity, well-being, and social capital among older patients with cardiovascular disease in the maintenance phase: a randomized controlled trial.

Everyone had to be able to walk without severe pain and without cognitive decline, so nothing here speaks to patients who are more disabled or confused. It was 55 people in one region of Japan, which is small for an activity trial, and the authors themselves note other areas and countries may behave differently. Participants could not be blinded: people walking in a group of three knew they were in the special arm, and they also rated their own happiness, which can inflate that kind of score. The groups walked together only about once a week, and the authors say this may have been too little to add anything on top of the walking both groups were already doing. Nothing else about how people walked was controlled, so walking with a dog or listening to music was allowed and could have influenced the step counts. Only groups of three were tested, so no one knows whether pairs or larger groups would work the same way. Only people who agreed to a referral from their own doctor joined, so they may already have been more motivated about activity than the average patient. Walking was the only activity counted toward physical activity; cycling, swimming and similar exercise were not quantified. The paper never names a single primary outcome and tests several outcomes without adjusting for multiple comparisons, so the one significant difference in happiness could be a chance finding. One results sentence calls the group of three the "SNI+FI group", a label carried over from earlier work, so parts of the report are harder to follow than they should be.

The study was supported by a grant from the Japan Sports Agency. The authors declare no conflict of interest. The paper does not say whether the funder had any role in designing the study, analysing the data or writing the report.

Do not read this as evidence that walking in groups of three increases physical activity. Step counts rose by a similar amount whether people walked in threes or alone, and the difference between the groups was not significant. The only significant difference was in self-rated happiness, reported by people who knew they were walking in the group, and it was a modest change on a small sample.