Purpose: To investigate the presence of reactivity, if any, to wearing sealed and unsealed pedometers, with and without step count recording.
Methods: On the first visit to the laboratory, 63 participants (41 females and 22 males; age = 23.6 ± 9.6 yr, body mass index = 22.7 ± 3.0 kg·m−2), blinded to the study aim, were provided with a sealed pedometer (New Lifestyles NL-1000) and informed that it was a "body posture monitor" (covert monitoring). Participants wore the pedometer throughout waking hours for 1 wk. On return to the laboratory, stored step counts were downloaded, and participants were informed that the device was a pedometer. Participants wore the pedometer under three more conditions-sealed, unsealed, and unsealed plus logging daily steps in an activity diary-each having a duration of 1 wk. The order of participation in each condition (sealed/unsealed/diary) was balanced across participants. Mean daily step counts recorded during the four conditions were compared using a repeated-measures ANOVA.
Results: There was a significant overall effect of condition (P < 0.001; covert monitoring = 8362 ± 2600 steps per day; sealed condition = 8832 ± 2845 steps per day; unsealed condition = 9176 ± 3299 steps per day; diary condition = 9635 ± 2709 steps per day), with post hoc analyses revealing that mean step counts were significantly higher in the diary condition than those reported during both the covert and sealed conditions (both P < 0.003). No significant gender effects were observed (P = 0.33).
Conclusion: The greatest increase in step counts occurred in the diary condition, suggesting that reactivity to pedometers is greatest when participants are requested to wear an unsealed pedometer and record their step counts. This has validity implications for short-term pedometer studies investigating habitual free-living activity that require participants to provide a daily log of their step counts.