The Effectiveness of Self-Guided Digital Interventions to Improve Physical Activity and Exercise Outcomes for People With Chronic Conditions: A Systematic Review and Meta-Analysis.

No included study was judged to be at low risk of bias. Most raised some concerns and five were at high risk in at least one area, largely because many had no published protocol and the outcomes were self-reported by people who knew which group they were in. The device-measured activity result is unreliable. Its confidence interval crossed the no-effect line and the four studies differed widely from one another (77.7% heterogeneity). The trials cover eleven different conditions with widely differing programmes, so a single pooled number hides a lot of variation. Most participants were under 60, so the results may not carry across to older or more disabled patients, and people with disabilities were almost absent from the included studies. The 16 trials reported their data incompletely or inconsistently, some analysed by intention to treat and some did not, and the reviewers sometimes had to compute or read values off figures to pool anything at all. The requirement that interaction be automatically generated excluded many digital programmes, and there is no standard definition of minimal contact, so the reviewers used their own.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The text supplied names no funder. The protocol was registered with PROSPERO (CRD42019132464).

Do not take this as proof that a self-guided app works as well as supervised exercise. That comparison comes from the authors' own conclusion, not from a pooled result: only one small trial used supervised exercise as its comparator and it was not included in the meta-analysis. The pooled benefit was small, and activity measured by devices showed no statistically significant improvement.