What the research says
Plain-language reads on sitting, breaks and desk bodies. Every study is linked, and we keep the limits each one states. How we use sources
- How often should you get up from your desk? What the studies actually testedNo official guideline sets a break interval. Here's what studies of desk workers actually tested, from 2-minute walks every 20 minutes to microbreaks, and how to pick a rhythm you'll keep.
- Do short walking breaks lower blood sugar after lunch?Lab trials found that 2-minute walks during long sits lowered the rise in blood sugar after a meal. Here's what that research shows, what it doesn't, and how it fits a workday.
- Is sitting all day bad if you work out? What data from 1 million people showLarge studies found that about an hour a day of moderate activity appeared to offset the link between long sitting and early death. Here's what that means if you exercise and sit all day.
- Neck pain from computer work: stretch, strengthen, or both?What trials of office workers found about stretching and strengthening for neck pain, including the 2-minutes-a-day band study, and when to see a clinician instead.
- Micro-breaks for the afternoon slump: energy, fatigue and stressA meta-analysis found micro-breaks gave a small lift in energy and a small drop in fatigue, without hurting output. Here's what short breaks can and can't do for the afternoon slump.
- Does the 20-20-20 rule work? A small study put it to the testA small study found less reported eye strain during 20-20-20 reminders. Learn what it tested, what remains uncertain, and how the reminder works.
- Are standing desks worth it? What trials say about sitting less at workStanding-desk studies measured how much people sat. See what a trial of desks plus support found, and why sitting less does not prove a health benefit.
- Do break reminder apps work? Why prompts help and then fadeReviews found that apps and prompts helped people sit about 40 minutes a day less in the short term, fading over months. An honest look, from people who build one.