I said what I said. The reviewers chose to only look at clinical populations and counted a study that found significant effects on physical aspects of stress to be a "no effect" study because the experimental group was already low anxiety and tested low again after the subjects had been meditating for a while. Such studies shouldn't be counted at all, rather than listed as "no effect," because when you do a simple average, a study that started high and ended high then is averaged in with a study that started low and ended low.
This skews the average. It's like asking 9 white guys and one black guy if they have ever experienced discrimination against blacks and concluding that anti-black discrimination in this country is very rare because 9 out of 10 respondents said they had never experienced it. L.