[tyu2]

How on earth can they control for that in short term studies? Is it even possible to get any statistical significance on that? Smells like bullshit, there is no way there is causation on health from volunteering.

Think about it, if you are volunteering you are likely to not just have good health, but afford good health care, have plenty of free time for various activities, less stress in life, enjoy and seek social activities, outdoor activities, etc. All those things could impact health, but not volunteering itself.

[cpncrunch]

If you read the article, it gives more details. For example, they did a controlled trial on two groups of students: one group was put on a waitlist, the other group given tutoring tasks. The tutoring group had lower levels of inflammatory markers.

[tyu2]

Sure, but it has nothing to do with volunteering or health.

[cpncrunch]

They were volunteering as tutors, and the study measured inflammatory cytokines.

[nhumrich]

The same wat you control for anything. You assign a group to do a specific thing. If a certain group performs better, its due to the task not preconditions, because the tasks were assigned, not observed and the selection process is random.