The association between daily steps and health, and the mediating role of body composition: a pedometer-based, cross-sectional study in an employed South African population

BACKGROUND Walking is recognized as an easily accessible mode of physical activity and is therefore supported as a strategy to promote health and well-being. To complement walking, pedometers have been identified as a useful tool for monitoring ambulatory physical activity, typically measuring total steps/day. There is, however, little information concerning dose-response for health outcomes in relation to intensity or duration of sustained steps. We aimed to examine this relationship, along with factors that mediate it, among employed adults. METHODS A convenience sample, recruited from work-site health risk screening (N = 312, 37 ± 9 yrs), wore a pedometer for at least three consecutive days. Steps were classified as "aerobic" (≥100 steps/minute and ≥10 consecutive minutes) or "non-aerobic" (<100 steps/minute and/or <10 consecutive minutes). The data were sub-grouped according to intensity-based categories i.e. "no aerobic activity", "low aerobic activity" (1-20 minutes/day of aerobic activity) and "high aerobic activity" (≥21 minutes/day of aerobic activity), with the latter used as a proxy for current PA guidelines (150-minutes of moderate-intensity PA per week). Health outcomes included blood pressure, body mass index, percentage body fat, waist circumference, blood cholesterol and blood glucose. Analysis of covariance, adjusting for age, gender and total steps/day were used to compare groups according to volume and intensity-based steps categories. A further analysis compared the mediation effect of body fat estimates (percentage body fat, body mass index and waist circumference) on the association between steps and health outcomes, independently. RESULTS Average steps/day were 6,574 ± 3,541; total steps/day were inversely associated with most health outcomes in the expected direction (p < 0.05). The "no aerobic activity" group was significantly different from the "low aerobic activity" and "high aerobic activity" in percentage body fat and diastolic blood pressure only (P < 0.05). Percentage body fat emerged as the strongest mediator of the relationship between steps and outcomes, while body mass index showed the least mediation effect. CONCLUSION The study provides a presentation of cross-sectional pedometer data that relate to a combination of intensity and volume-based steps/day and its relationship to current guidelines. The integration of volume, intensity and duration of ambulatory physical activity in pedometer-based messages is of emerging relevance.