Journal of Medical Internet Research

Personalized prompts raised daily steps among Fitbit users

After two months, users receiving adaptive activity prompts increased their daily step count by 210 more steps than users receiving no prompts.

A large footprint sits among walking figures, with more small footprints near figures with speech bubbles and fewer gray footprints near those without.
Studied:
Randomized trial in 7,711 Fitbit users, 2 months, against users receiving no prompts
What changed:
Daily steps increased by 210 steps more with personalized prompts than with no prompts at two months
The catch:
Only 7711 of 13,463 enrolled users entered the primary analysis, risking selection bias

Editor of record: Boris Djordjevic · AgingResearch.News

Follow @agingbrief

Fitbit users received adaptive, fixed-rule or random activity prompts, or no prompts, in a four-arm randomized controlled trial enrolling 13,463 people. The adaptive algorithm personalized the prompts’ content and timing. The primary analysis included 7,711 participants, whose mean age was 42.1 years; 86.3% were female.

At two months, the adaptive group’s increase in average daily steps was 210 steps greater than the no-prompt group’s. Differences between the adaptive group and the fixed-rule and random groups were not statistically significant at that point. At one month, the adaptive group had larger step increases than all three comparison groups.

Why it matters

For efforts to keep people active as they age, the trial addressed whether digital prompts can personalize activity support at scale.

Caveats

The authors note that follow-up lasted only two months, leaving longer-term changes open. The self-selected, predominantly female Fitbit users may not represent men or people who do not use Fitbit.

Comments

Be the first to comment.