How Should Objectively Measured Physical Activity Data Be Used Analytically?

Abstract

Dear Editor-in-Chief: Accelerometers now allow researchers to monitor the intensity and duration of exercise and nonexercise activities with considerable precision. However, there is a need for agreed procedures to summarize the resulting mass of data. Pate et al. (5) have suggested distinguishing sedentary (1.0-1.5 metabolic equivalents (METs)) and/or light activities (1.6-2.9 METs) as well as moderate (3.0-6.0 METs) and/or vigorous activities (≥6.1 METs). They presented 13 h of accelerometry data for two persons with similar total energy expenditures (26 vs 24 METs·h−1) but markedly different activity patterns (sedentary, 25% vs 70%; light, 75% vs 23%; moderate to vigorous, 0% vs 7%). Such classification becomes important when testing associations between activity patterns and health outcomes. A study of seniors from Nakanojo (1-4, 6-8) has, since 2000, used specially adapted pedometer/accelerometers (modified Kenz Lifecorders) to establish detailed physical activity patterns 24 h·d−1 over an entire year. Findings have been summarized as daily step counts and daily durations of sleeping and sedentary, light, moderate, and vigorous activities. Our observations (1) demonstrate close correlations between the daily step count (x) and the daily duration of moderate-to-vigorous activity (y) (y = [1.96 · 10−7] x2 + [1.16 · 10−3] x; r2 = 0.93). Those with a higher daily step count are also likely to engage in more moderately vigorous activity. It seems that the little activity sedentary older adults take is usually of light intensity, whereas active individuals engage in significant periods of what for this age group is moderate activity. In an average individual taking 7000 steps·d−1, a quarter of the total effort (15 min·d−1) is moderate to vigorous; this percentage is lower (<20%, <5 min·d−1) in those taking <4000 steps·d−1 and higher (>30%, >30 min·d−1) in those taking >10,000 steps·d−1. Pate et al. (5) have recommended exploiting accelerometry to examine associations between activity levels and health outcomes. However, substantial data of this type already exist. Our studies (1) indicate that measures of physical and psychosocial health as diverse as depression (8), a poor health-related quality of life (7), arteriosclerosis, osteoporosis (3), sarcopenia, a lower level of physical fitness (2), and metabolic syndrome (4) are all associated with both the intensity and the total volume of habitual physical activity undertaken. The threshold amount of physical activity associated with better health seems greater for physical than for mental benefits: >8000 versus >4000 steps·d−1 and/or >20 versus >5 min·d−1 at a moderate-to-vigorous intensity. Physical health is better in seniors spending at least 20 min·d−1 in brisk walking and a further >50 min·d−1 in light activity, whereas mental health is associated with much smaller amounts of deliberate physical activity. Moreover, and perhaps contrary to an emphasis on sedentariness, physical and psychosocial health seems greater in individuals who undertake a larger proportion of their daily activity at a moderate-to-vigorous intensity. Yukitoshi Aoyagi, Ph.D. Exercise Sciences Research Group, Tokyo Metropolitan Institute of Gerontology, Itabashi, Tokyo, Japan Roy J. Shephard, M.D., Ph.D., LL.D., FACSM Faculty of Physical Education and Health, University of Toronto, Toronto, Ontario, Canada