The evidence
Every number on this site comes from a published study. Each one is below with its sample size, its limits and a link you can follow.
No study has tested BetterBreaks. What follows is evidence about the behaviour the app reminds you to do, which is a different thing, and keeping the two apart is the only reason any of this is worth reading.
The strongest finding we have
higher all-cause mortality among adults who mostly sat at work, against those who mostly did not. Hazard ratio 1.16, 95% CI 1.11 to 1.20.
higher cardiovascular mortality in that same group. Hazard ratio 1.34, 95% CI 1.22 to 1.46.
hazard ratio for adults who alternated between sitting and not sitting. 95% CI 0.97 to 1.05, which contains 1, so no measurable excess risk.
481,688 adults in Taiwan, followed a mean 12.85 years, 26,257 deaths, adjusted for sex, age, education, smoking, drinking and body mass index. Among people who mostly sat and did little leisure activity, an extra 15 to 30 minutes of daily activity brought risk down to the level of inactive people who did not sit at work.
Observational cohort. It shows an association, not a cause. It tested the posture, not any software, and it says nothing about what BetterBreaks does to anyone's risk.
Study by study
Journal of Optometry · 2024
Pooled prevalence of computer vision syndrome was 69.0%, 95% CI 62.2 to 75.4.
66,577 participants · 103 cross-sectional studies
Individual estimates ranged from 12.1% to 97.3%, so the interval is the honest figure, not the midpoint. The pooled population includes students, not only office workers.
Read the studyInvestigative Ophthalmology & Visual Science · 2023
Blinking measured in situ with an eye-tracking headset fell to 10.7 blinks a minute while reading, against 32.4 in conversation and 31.3 while walking. The change appeared within one minute and held.
24 participants · randomised cross-over, eight tasks
Print, laptop, TV and phone suppressed blinking equally, so screens are not the culprit people assume. The study also looked for a link between blink rate and eye symptoms and did not find one.
Read the studyContact Lens and Anterior Eye · 2023
Two weeks of 20-20-20 reminders reduced digital eye strain and dry eye symptoms, p ≤ 0.045. One week after the reminders stopped, the improvement was gone.
29 participants · before and after, no control arm
Tear film and ocular surface signs did not change, p ≥ 0.089. Symptoms improved; nothing a clinician measures did.
Read the studyPLOS ONE · 2022
Across 22 samples, micro-breaks of ten minutes or less significantly increased vigor, d = 0.36, and reduced fatigue, d = 0.35, both p < .001. The effect on task performance was d = 0.16, 95% CI −.04 to .37, p = .117, which is not significant.
2,335 participants · meta-analysis of 22 samples
This is the finding most break apps leave out. Longer breaks did predict better performance, b = .07, p = .006, and gains showed up for creative work but not for cognitive tasks. We do not tell you breaks will make you more productive, because the best available meta-analysis could not show it.
Read the studyPreventive Medicine · 2023
Sedentary behaviour carried an odds ratio of 1.5 for neck pain, 95% CI 1.29 to 1.76. Computer use specifically was 1.3, 95% CI 1.12 to 1.53.
27 reports pooled · 46 reviewed
Three things cut against it. Sitting time on its own was not significant, 95% CI 0.86 to 2.07. The office-worker subgroup, which is our audience, was also not significant, 95% CI 0.98 to 1.89. And the authors found publication bias, p < 0.0001.
Read the studyWhere the evidence disagrees
Both looked at whether breaks prevent neck and shoulder problems. Both are systematic reviews of randomised trials, and both graded their own evidence as weak. Here is each one, in its own words.
Cochrane · 2018 · Hoe et al.
MD −0.25
Supplementary breaks reduced neck discomfort among data entry workers, 95% CI −0.40 to −0.11, from two studies. Shoulder and upper arm discomfort fell by 0.33, forearm and wrist by 0.18. The review labels all of it very low-quality evidence.
15 randomised trials · 2,165 office workers
Cochrane · 2019 · Luger et al.
SMD −0.08
Additional work breaks had no considerable effect on musculoskeletal pain, discomfort or fatigue, 95% CI −0.35 to 0.18. Higher-frequency breaks made no difference against lower-frequency ones, and active breaks made no difference against passive ones. All six trials were judged at high risk of bias.
6 randomised trials · 373 workers
The review designed specifically to test break schedules is the one that found nothing. So we do not say that breaks prevent neck or shoulder pain. We say that sedentary work is associated with neck pain, which is a different sentence, and we link both reviews so you can weigh them yourself.
The 2019 review also searched for any trial testing how long a break should be, and found none. Nobody has tested break duration, ours included.
The limits
Every item here is a place where someone selling a wellness app would usually round up.
Free on macOS and Windows, and nothing leaves your computer.
