To claim that, you would need to back it up with data, and that would be rather difficult because there is a percentage of people who seek therapy when have problems and those who don't, so it's hard to get the staring point re. correlation even before we proceed to causation. It's just like with alcohol and overall mortality - you can claim whatever you want about moderate drinking (teetotalers have worse outcomes because they don't drink or they have worse symptoms therefore they don't drink) and there will always be people in both camps.