Journal Logo


Associations of Low- and High-Intensity Light Activity with Cardiometabolic Biomarkers


Author Information
Medicine & Science in Sports & Exercise: October 2015 - Volume 47 - Issue 10 - p 2093-2101
doi: 10.1249/MSS.0000000000000631


Changes in transportation, communications, workplace, and domestic entertainment technologies have significantly reduced daily demands to be active (13). Public health guidelines and research on physical activity and health have emphasized the time that adults spend doing moderate- to vigorous-intensity activity (MVPA): a minimum of 30 min of moderate activity 5 d·wk−1 or 20 min of vigorous activity 3 d·wk−1 is generally recommended (21). However, there has been a shift from focusing on this higher end of the physical activity spectrum to considering the potential benefits of light-intensity physical activity (LIPA) (5).

Despite LIPA’s large volume (some 40% of adults’ waking hours) and its substantial contribution to overall daily energy expenditure (11), its potential health benefits are not well understood. Although there are well-documented difficulties of capturing light activities by questionnaires, there is evidence for the potential health benefits of self-reported LIPA (2,27). LIPA, as measured objectively by accelerometers (in particular ActiGraph 7164 or GT1M, as used in the studies cited herein), has also shown beneficial associations with body mass index (BMI), waist circumference, cholesterol ratio, C-reactive protein (CRP), and 2-h plasma glucose (23,28,36). However, total LIPA is composed of a heterogeneous range of activities, with some activities regarded as low intensity, such as simply standing with little movement. As a consequence, the benefits that may arise from activities of all intensities grouped broadly as LIPA or from the more active end of the LIPA spectrum are not well documented. Understanding the potential benefits of both the lower end and the higher end of this spectrum is important in providing guidance on the promotion of LIPA (34), such as whether low LIPA (LLPA) is expected to be beneficial or whether high LIPA (HLPA) is needed to elicit benefits.

Various thresholds have been used to differentiate accelerometer-derived LIPA. Buman et al. (4) applied midpoint Copeland thresholds (8) and showed beneficial associations at the lower end (100–1040 counts per minute) and higher end (1041–1951 counts per minute) of the LIPA spectrum with self-rated indices of physical health and well-being in older adults. Hagstromer et al. (18) used MET values of activities to characterize low physical activity (100–759 counts per minute) and lifestyle physical activity (760–2019 counts per minute) in US and Swedish adults. The threshold of 760 counts per minute was proposed to be the point at which dynamic (ambulatory) light-intensity activities (such as sweeping, gardening, and playing golf) can be captured, whereas counts below 760 counts per minute are more likely to represent static (standing) light-intensity activities (such as cooking, ironing, or stretching) (29). This cutpoint was also used to examine associations of LLPA and HLPA with cardiometabolic biomarkers in adolescents (7). To date, no study has examined such associations in adults.

We examined the associations of time spent in objectively measured LIPA (categorized as LLPA or HLPA), moderate-intensity physical activity (MPA), and vigorous-intensity physical activity (VPA) with continuous measures of cardiometabolic risk biomarkers in US adults who participated in the National Health and Nutrition Examination Survey (NHANES).


Data were obtained from the 2003–2004 and 2005–2006 NHANES cycles (32). Of the 20,470 participants who participated in these data cycles, 10,020 were adults (≥20 yr). Exclusion for pregnancy, insulin therapy, and/or lack of accelerometer assessment left 7902 eligible participants (Figure, Supplemental Digital Content 1, Study inclusion flow chart, Of the eligible participants, 4625 had data for nonfasting outcome measures and covariates, valid accelerometer wear, plausible energy intake, and CRP values indicating no acute inflammation (CRP ≤10 mg·L−1). Of these, 2007 attended the medical examination center fasted (≥8.5 h), and 853 underwent oral glucose tolerance test (OGTT; in 2005–2006 only). The study was approved by the National Center for Health Statistics Ethics Review Board and was carried out in accordance with the Declaration of Helsinki. All participants provided a written informed consent form.

Physical activity.

Physical activity was assessed objectively using a hip-worn uniaxial ActiGraph accelerometer (model 7164; ActiGraph, Ft. Walton Beach, FL), which collected data across waking hours in 60-s epochs for 7 d. Details of data collection and analysis have been described previously (25). Consistent with our earlier study (25), a valid day was defined as ≥10 h of wear time and no counts >20,000 counts per minute. Participants were included if they had at least four valid days (including at least one weekend day). A validated algorithm used previously with NHANES data (40) was employed to exclude nonwear periods (i.e., ≥60 consecutive minutes of 0 counts per minute, with allowance of ≤2 min of 1–49 counts per minute); the algorithm was modified so that nonwear periods could span midnight. Based on the cutpoint for sedentary (100 counts per minute) (22), the threshold (760 counts per minute) subdividing light activities (29), and Freedson cutpoints (16), counts were classified as follows: LLPA, 100–759 counts per minute; HLPA, 760–1951 counts per minute; MPA, 1952–5724 counts per minute; VPA, ≥5725 counts per minute. Total time at these intensities for each day was calculated, averaged across all valid days, and standardized for wear time using the residuals method.

Cardiometabolic biomarkers.

All anthropometric measures were collected by trained technicians according to standard procedures. Waist circumference was measured to the nearest 0.1 cm at the iliac crest; height was measured to the nearest 0.1 cm with a stadiometer; and weight was measured in pounds with a Toledo digital scale (Mettler-Toledo Inc., Columbus, Ohio) and converted into kilograms via an automated system, with the participant wearing underwear, a paper gown, and foam slippers. BMI was calculated as weight (kg)/height (m)2. Resting systolic blood pressure and diastolic blood pressure were measured with a mercury sphygmomanometer and reported as averages of three to four measurements, excluding the first reading and questionable values. Nonfasting biomarkers included HDL cholesterol (analyzed using the Roche/Boehringer-Mannheim Diagnostics direct HDL method) and CRP (analyzed via latex-enhanced nephelometry on a Behring nephelometer, with values >10 mg·L−1, indicating acute inflammation, excluded) (35).

Fasting biomarkers included the following: triglycerides (analyzed enzymatically using the Beckman Synchron LX20 analyzer in 2003–2004 and the Hitachi 717/912 analyzers in 2005–2006), plasma glucose (analyzed via the hexokinase method, using the Roche Cobas Mira analyzer in 2003–2004 and the Roche/Hitachi 911 analyzer in 2005–2006), and insulin (measured in the Tosoh AIA-PACK IRI immunoenzymometric assay in 2003–2004 and in the Merocodia Insulin ELISA immunoassay in 2005–2006). Correction equations were applied to account for differences in the analysis methods of glucose and insulin between sampling years. Corrected values within detectable ranges (n = 1673) for insulin (20–400 pmol·L−1) and glucose (3–25 mmol·L−1) were used in homeostatic model assessment (HOMA) to provide measures of β-cell function (HOMA-%β) and insulin sensitivity (HOMA-%S). Two-hour plasma glucose levels were obtained from fasting participants in the 2005–2006 survey only, via OGTT.


Interviewer-administered questionnaires obtained information on sociodemographic, behavioral, and medical attributes. Race/ethnicity, education, marital status, and family poverty income ratio were categorized as outlined in Table 1. Serum cotinine levels (examined continuously) were also used to classify smoking status (Table 1). A single 24-h diet recall, coupled with US Department of Agriculture food composition data, measured dietary variables: total energy, saturated fat, and alcohol intake (examined continuously); extreme total energy intakes (>4341 kcal) calculated from interquartile ranges were excluded. Dichotomous variables were generated from self-reported medical history of diabetes/cardiovascular disease/cancer, use of oral contraceptives, hormone replacement, and postmenopausal status, using the classification outlined by Lynch et al. (28). Current medication use was recorded and coded according to the Lexicon Plus (Cerner Multum Inc.) database (32).

Characteristics of the study population (NHANES 2003–2004 and 2005–2006).

Statistical analyses.

Analyses were conducted with STATA version 12.1 (StataCorp, College Station, TX). Consistent with the complex design, we used linearized variance estimates with weightings to achieve representativeness of the US population. Mobile examination center sample weights and fasting subsample weights were combined for the two data cycles (2003–2004 and 2005–2006). These weights and the 2005–2006 OGTT weights were then reweighted according to the method used by Troiano et al. (39) to account for any nonrandom absences in exposure, outcome, and confounding variables. Spearman’s correlation tests were used to assess the relationship between physical activity variables.

Linear regression models examined associations of the four physical activity variables with continuously measured cardiometabolic biomarkers. Model A adjusted for sociodemographic, behavioral, and medical covariates retained in backward elimination (Table, Supplemental Digital Content 2, Covariates retained in backward elimination: P < 0.2 for retention, Model B included additional adjustment for waist circumference or an alternate version of model B using additional adjustment for BMI rather than waist circumference. To rescale results commensurately with the vastly differing volumes of each type of activity (volume decreases substantially as intensity increases; Table 1), we report the results partially standardized (i.e., representing the effect on mean biomarkers of increasing the mean values of each activity by 1 SD) (Table 2). With the exception of waist circumference, BMI, diastolic blood pressure, and LDL cholesterol, all biomarkers were log-transformed to improve normality. Models did not display problems from nonnormality, nonlinearity, or heteroscedasticity, except for fasting glucose. Those with known diabetes had both very high predicted values of fasting glucose and the most error in the prediction of their fasting glucose levels; accordingly, the results are not applicable for extreme glucose levels or for groups with a higher prevalence of diabetes.

Associations of physical activity (per SD) with cardiometabolic biomarkers in US adults age ≥20 yr from NHANES 2003–2006.

Interaction terms examined variations in associations with physical activity by gender, age (adults [<65 yr], older adults [≥65 yr]), race/ethnicity (non-Hispanic White, Mexican–American, non-Hispanic Black), and level of MVPA (active [≥30 min·d−1], inactive [<30 min·d−1]; all minutes were considered [i.e., total sum ≥1952 counts per minute] without weighting to avoid inflated standard errors). For race/ethnicity comparisons, the “other” group was excluded to avoid small sampling units (Figure, Supplemental Digital Content 1, Study inclusion flow chart, Significance was set at P < 0.05 for main effects and—to reduce Type I error—at P < 0.001 for interactions.


Population-weighted sociodemographic, behavioral, and medical characteristics are shown for the full sample and fasting subsamples in Table 1. Any biases from excluding participants with missing data were largely resolved by reweighting. The reweighted data for included participants (full sample, fasting subsample, and OGTT subsample) were similar in most characteristics (except smoking) to all eligible participants (data not shown). The mean ± SD age of the full sample was 47 ± 17 yr; 48% of the sample was male (Table 1). In the full sample, the strongest significant positive correlations were observed for LLPA and HLPA (r = 0.50), HLPA and MPA (r = 0.70), and MPA and VPA (r = 0.41). Weaker but significant correlations were observed for LLPA and MPA (r = 0.16), HLPA and VPA (r = 0.21), and LLPA and VPA (r = −0.04).

Associations of intensity of physical activity with cardiometabolic biomarkers.

Following adjustment for potential confounders, all intensities of physical activity were significantly beneficially associated with waist circumference, CRP, triglycerides, fasting insulin, HOMA-%β, and HOMA-%S (Table 2). Associations were consistently larger for MPA and VPA than for light activity, which in turn was consistently larger for HLPA than for LLPA, even when examined for every SD of exposure, which tends to favor lower-intensity activities. For example, each SD increment in VPA (4 min), MPA (22 min), HLPA (45 min), and LLPA (63 min) was significantly associated with an 8% (95% CI, 4%–12%), a 17% (95% CI, 11%–22%), a 14% (95% CI, 9%–19%), and an 8% (95% CI, 4%–12%) lower CRP, respectively (Table 1). For every minute of activity, the associations were strongest for VPA (≈2%), MPA (≈0.8%), HLPA (≈0.3%), and LLPA (≈0.1%). Each SD increment in LLPA was associated with a 1% (95% CI, 0.1%–1%) higher systolic blood pressure, independent of waist circumference. BMI, HDL cholesterol, fasting glucose, and 2-h plasma glucose were significantly associated with HLPA, MPA, and VPA only (Table 2). There were no significant associations between any intensity of physical activity and diastolic blood pressure or LDL cholesterol.

Following additional adjustment for waist circumference, associations of LLPA with CRP, triglycerides, insulin, HOMA-%β, and HOMA-%S were modestly attenuated (25%–38% relative). With the exception of HDL cholesterol (50% relative), effect sizes for all associations of HLPA were modestly attenuated (14%–31% relative) following additional adjustment for waist circumference (Table 2). Associations of MPA with CRP, HDL cholesterol, triglycerides, insulin, and 2-h glucose were partially attenuated (36%–50% relative), and associations with HOMA-%β and HOMA-%S were strongly attenuated (67%–75% relative) after additional adjustment for waist circumference. With the exception of triglycerides and 2-h glucose (43%–50% relative), associations of VPA with CRP, HDL cholesterol, insulin, HOMA-%β, and HOMA-%S were strongly attenuated (60%–75% relative) after additional adjustment for waist circumference. There was no notable difference in results when BMI, rather than waist circumference, was adjusted for as the measure of adiposity (Table, Supplemental Digital Content 3, Associations of physical activity… additionally adjusted for BMI…,

Following additional adjustment for MET minutes of MVPA, associations of LLPA with cardiometabolic biomarkers remained statistically significant with minimal attenuation, whereas associations of HLPA with cardiometabolic biomarkers either were no longer significant (BMI, waist circumference, HDL cholesterol, fasting glucose, and 2-h glucose) or were attenuated (CRP, insulin, HOMA-%β, and HOMA-%S) (Table, Supplemental Digital Content 4, Associations of light-intensity physical activity… additionally adjusted for MET minutes of MVPA…,

Significant interactions were observed: by gender, for associations of HLPA, MPA, and VPA with systolic blood pressure and for associations of VPA with BMI and triglycerides (Fig. 1A–E); by age, for associations of LLPA with CRP and for associations of HLPA with LDL cholesterol (Fig. 2A and B); and, by ethnicity, for associations of LLPA with waist circumference (Fig. 3). In summary, associations of HLPA, VPA, and MPA with systolic blood pressure tended to be significant and beneficial in women and tended to be nonsignificant or detrimental in men. Associations of VPA with BMI and triglycerides were either stronger in women or significant only in women, compared with men, respectively. Inverse associations of LLPA with CRP and of HLPA with LDL cholesterol were observed in older adults, but not overall. In terms of racial/ethnic differences, a beneficial linear association of LLPA with waist circumference was most evident within non-Hispanic Whites. For non-Hispanic Blacks, LLPA was adversely associated with waist circumference. No interactions, by sufficient MVPA, for associations of LLPA and HLPA with biomarkers were significant at P < 0.001.

Associations of physical activity with cardiometabolic biomarkers, by gender. Data are presented as marginal mean (95% CI) adjusted for sociodemographic, behavioral, and medical covariates retained through backward elimination for minutes per day of (A) VPA with BMI; (B) HLPA with systolic blood pressure; (C) MPA with systolic blood pressure; (D) VPA with systolic blood pressure; and (E) VPA with triglycerides. Only interactions P < 0.001 are shown. Shading and asterisk denote population mean.
Associations of physical activity with cardiometabolic biomarkers, by age. Data are presented as marginal mean (95% CI) adjusted for sociodemographic, behavioral, and medical covariates retained through backward elimination for minutes per day of (A) LLPA with CRP and (B) HLPA with LDL cholesterol. Only interactions P < 0.001 are shown. Shading and asterisk denote population mean.
Associations of LLPA with waist circumference, by ethnicity. Data are presented as marginal mean (95% CI) adjusted for sociodemographic, behavioral, and medical covariates retained through backward elimination for minutes per day of LLPA. Only interactions P < 0.001 are shown. Shading and asterisk denote population mean.


Despite LIPA’s high volume and substantial contribution to total energy expenditure, little is known about the associations of LIPA with cardiometabolic biomarkers or whether the associations vary with the intensity of LIPA. This study investigated, for the first time, the associations of LIPA (differentiated into LLPA and HLPA) with biomarkers of cardiometabolic risk in a large population-based sample of adults. Consistent with previous studies (4,5), all intensities of activity above sedentary (≥100 counts per minute) showed some significant beneficial associations with cardiometabolic biomarkers; adverse associations were only observed for systolic blood pressure. The overall pattern of results was consistent with higher intensity being associated with greater cardiometabolicbenefits. Many, but not all, of the beneficial associations of light activity were observed independently of MVPA. Collectively, these findings reinforce both long-standing recommendations for MVPA and more recent public health messages that participation in activity of any intensity (from LLPA to VPA) may be additionally beneficial for the cardiometabolic health of the overall adult population (1). This evidence for the benefits of LLPA and HLPA would also support innovative physical activity initiatives designed for those who have limited capacities or motivation for MVPA (34), namely, by suggesting that although the degree and nature of benefit may vary depending on which activities are increased and by how much, at least some benefit could be expected from increasing time spent in any activity.

The strong associations observed with MPA and VPA are consistent with previous observations that have shown that reallocating time (using isotemporal substitution modeling) from other behaviors (sleep, sedentary, and LIPA) to MVPA was associated with the most favorable associations with cardiometabolic biomarkers (5). However, MVPA occupies only a small portion of the waking day, and it has been argued that significant shifts (increases) in MVPA are likely to be challenging at the population level given its already low volume. In this study, 3007 of the 4164 participants recorded no VPA, and over 60% had mean daily amounts of MVPA less than 30 min. In contrast, nearly 40% of an average participant’s time was made up of light activity in the form of LLPA (29%) and HLPA (10%). This is consistent with the findings observed by Hagstromer et al. (18) in Swedish and US adults.

From a public health perspective, the sizeable associations observed for both LLPA and HLPA suggest that increases in LIPA may have beneficial impacts on biomarkers of cardiometabolic health. These findings extend upon those observed by Buman et al. (4), in which LLPA and HLPA were beneficially related to self-reported physical health and well-being in older adults. However, the typically larger effect sizes for HLPA than for LLPA in this study and in that of Buman et al. (4) suggest that the degree of health benefits of LIPA may depend on the extent to which it is composed of lower-intensity activities (often static) or higher-intensity activities (often dynamic). Irrespective of intensity, substantial increases (>1 h) in LIPA have been shown to be both achievable and acceptable for older adults (17) and adults in a workplace-based intervention study (24). In the current study, a 1-SD shift in LLPA (∼1 h) or HLPA (∼45 min) was associated with a 1-cm lower waist circumference, 8% (LLPA) and 14% (HLPA) lower CRP, 4% lower triglycerides (LLPA and HLPA), and 7% (LLPA) and %3 (HLPA) lower insulin.

Additional adjustments for waist circumference had minimal impact on the strength of the associations for HLPA and had modest impact on the strength of the associations for LLPA with biomarkers. In contrast, associations of MPA and VPA with these cardiometabolic biomarkers were more likely to be substantially attenuated after additional adjustments for waist circumference. These findings are broadly concordant with experimental evidence indicating that the benefits of LIPA may not necessarily be driven entirely by energy expenditure and subsequent effects on central adiposity (15), just as the benefits of MVPA are not solely related or confined to weight loss. In the context of escalating rates of overweight and obesity (33), such benefits of LIPA may be of increasing preventive health relevance.

Observations from animal models (19) indicate that the reduced skeletal muscular contractions and reduced lipoprotein lipase activity that occur with inactivity (sitting induces no or little muscular contraction) (38) are offset by LIPA. In addition to overall muscular activity, understanding the recruitment and activation of specific human muscle fibers for each intensity of activity may assist in the interpretation of the results of the present study. MVPA predominantly activates mobilizer (fast glycolytic) muscle fibers, whereas LIPA utilizes stabilizer/postural (slow oxidative) muscle fibers (20). These physiological differences could potentially explain not only the beneficial associations observed for triglycerides (primary fuel source for oxidative fibers) and insulin but also our observed beneficial associations for glucose metabolism (fasting glucose and 2-h plasma glucose) with HLPA and MVPA, but not with LLPA.

Previously, Hagstromer et al. (18) indicated that 760 counts per minute is likely to be the threshold at which activities involving ambulation are captured, speculating that 760 counts per minute could even be a better cutpoint for moderate-intensity activities. Therefore, intensities beyond HLPA are likely to require greater recruitment of glycolytic fibers, of which glucose is the primary fuel source (9). Findings from recent experimental studies are supportive of this assertion (3,14). Compared to uninterrupted sitting, postprandial (similar to OGTT) plasma glucose incremental area under the curve was similarly attenuated with the inclusion of both light-intensity (equivalent to HLPA) and moderate-intensity short frequent walking breaks (14), but not short frequent standing breaks (3). Equivocal findings exist for the effects of longer-standing breaks on postprandial glucose relative to sitting. Thorp et al. (37) reported an attenuation in postprandial glucose incremental area under the curve during the trial, whereas Miyashita et al. (31) reported no difference in postprandial glucose area under the curve the following day. To determine the relationship of standing with glucose metabolism, we need observational evidence that accurately captures posture and further experimental studies, particularly those with longer-term exposures.

The deleterious association of LLPA with systolic blood pressure is consistent with previous findings from NHANES (5) in which total LIPA was deleteriously associated with systolic blood pressure. Although this contrasts with the benefits seen for other markers, the effect size was small (1% higher systolic blood pressure per SD), and it could be reasonably assumed that this would likely not negate the other benefits of LLPA.

With a few exceptions, there was little evidence in our study indicating that the associations observed were present only in subgroups in terms of age, gender, race/ethnicity, or sufficient level of MVPA. Most associations were observed consistently across all subgroups. However, there were some statistically significant (P < 0.001) age, gender, and racial/ethnic variations in activity–biomarker associations, which warrant further investigation. These could be spurious findings related to multiple testing, genuine findings, or the result of differential misclassification, unmeasured confounders, and/or important but unmeasured biological differences (e.g., in fat deposition or physiological characteristics). For example, it has been shown that associations of pericardial fat and visceral fat differ according to intensity of physical activity (26,30) and ethnicity (6,10). More detailed outcome measures may be needed to better understand the potential pathways through which activities of different intensities may impact the cardiometabolic health of different groups, ideally in longitudinal or experimental studies.

A key strength of this study is the objective measurement of physical activity in a large population-representative sample. Although the use of accelerometers helps overcome limitations of self-report measurement (recall and reporting bias), they are also subject to measurement error and limited in their interpretability. Intensities of activity are classified by vertical movement between certain threshold ranges (29), with lower-intensity activities being the most poorly classified. While LLPA (100–759 counts per minute) is likely to comprise more standing and less ambulatory activities than is HLPA (760–1951 counts per minute), all standing falling within 100–760 counts per minute or all stepping being ≥760 counts per minute is unlikely to be the case (26). Multiple postures and intensities of movement can occur in each 1-min epoch, and the total intensity of all vertical hip accelerations within these epochs only partly correlates to true movement intensity, posture, and ambulation. To overcome these limitations, we need studies incorporating direct postural measures and finer-grained nonepoch data to provide direct evidence regarding the potential difference between standing and ambulatory LIPA (see Dowd et al. (12) for an example in adolescents). Elucidating mechanisms regarding muscular activity and energy expenditure would also be beneficial. The cross-sectional design precludes causal inferences and does not rule out the possibilities of reverse causation and residual confounding from unmeasured covariates.

These findings provide novel epidemiological evidence for the potential benefits of LLPA and HLPA. They also reinforce the importance of MVPA, which is the mainstay of public health recommendations. Furthermore, these findings highlight the need to further examine possible gender, age, and race/ethnicity differences so that population-specific recommendations can be made, if required. Future studies would benefit from the use of prospective and intervention trial designs (using measures that can delineate LIPA as standing versus ambulatory activity) and the examination of additional biomarkers to elucidate mechanisms.

The authors are most grateful to the US Centers for Disease Control and Prevention for allowing public access to NHANES data and to the participants for volunteering their time for the study.

This work was supported by the National Health and Medical Research Council Program (grant 566940 to N. Owen); National Health and Medical Research Council/National Heart Foundation Postgraduate Scholarship (grant 1056320 to B. Howard); Endeavour Research Fellowship (V. Carson); Australian Research Council Research Fellowship (FT100100918 to D.W. Dunstan) and Discovery Early Career Researcher Award (DE120101173 to N.D. Ridgers); National Health and Medical Research Council Principal Research Fellowship (grant 1026216 to J. Salmon), Training Fellowship (grant 569861 to G.N. Healy), and Senior Principal Research Fellowship (grant 1003960 to N. Owen); and Heart Foundation Postdoctoral Fellowship (grant PH 12B 7054 to G.N. Healy) and was supported in part by the Victorian Government’s Operational Infrastructure Support Program.

The funders of this study had no role in the analysis or interpretation of results.

The authors declare no conflicts of interest.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.


1. Australian Government. Australia’s Physical Activity and Sedentary Behaviour Guidelines. Canberra, ACT: Department of Health; 2014 [cited 2014 August 4]. Available from:
2. Autenrieth CS, Baumert J, Baumeister SE, et al. Association between domains of physical activity and all-cause, cardiovascular and cancer mortality. Eur J Epidemiol. 2011; 26 (2): 91–9.
3. Bailey DP, Locke CD. Breaking up prolonged sitting with light-intensity walking improves postprandial glycemia, but breaking up sitting with standing does not. J Sci Med Sport. [Internet]. 2014 [cited 2014 July 22]. Available from: doi: 10.1016/j.jsams.2014.03.008.
4. Buman MP, Hekler EB, Haskell WL, et al. Objective light-intensity physical activity associations with rated health in older adults. Am J Epidemiol. 2010; 172 (10): 1155–65.
5. Buman MP, Winkler EA, Kurka JM, et al. Reallocating time to sleep, sedentary behaviors, or active behaviors: associations with cardiovascular disease risk biomarkers, NHANES 2005–2006. Am J Epidemiol. 2014; 179 (3): 323–34.
6. Carroll JF, Chiapa AL, Rodriquez M, et al. Visceral fat, waist circumference, and BMI: impact of race/ethnicity. Obesity. 2008; 16 (3): 600–7.
7. Carson V, Ridgers ND, Howard BJ, et al. Light-intensity physical activity and cardiometabolic biomarkers in US adolescents. PLoS One. 2013; 8 (8): e71417.
8. Copeland JL, Esliger DW. Accelerometer assessment of physical activity in active, healthy older adults. J Aging Phys Act. 2009; 17 (1): 17–30.
9. De Feo P, Di Loreto C, Lucidi P, et al. Metabolic response to exercise. J Endocrinol Invest. 2003; 26 (9): 851–4.
10. Despres JP, Couillard C, Gagnon J, et al. Race, visceral adipose tissue, plasma lipids, and lipoprotein lipase activity in men and women: the Health, Risk Factors, Exercise Training, and Genetics (HERITAGE) family study. Arterioscler Thromb Vasc Biol. 2000; 20 (8): 1932–8.
11. Donahoo WT, Levine JA, Melanson EL. Variability in energy expenditure and its components. Curr Opin Clin Nutr Metab Care. 2004; 7 (6): 599–605.
12. Dowd KP, Harrington DM, Hannigan A, et al. Light intensity physical activity is associated with adiposity in adolescent females. Med Sci Sports Exerc. 2014; 46 (12): 2295–300.
13. Dunstan DW, Howard B, Healy GN, Owen N. Too much sitting—a health hazard. Diabetes Res Clin Pract. 2012; 97 (3): 368–76.
14. Dunstan DW, Kingwell BA, Larsen R, et al. Breaking up prolonged sitting reduces postprandial glucose and insulin responses. Diabetes Care. 2012; 35 (5): 976–83.
15. Duvivier BM, Schaper NC, Bremers MA, et al. Minimal intensity physical activity (standing and walking) of longer duration improves insulin action and plasma lipids more than shorter periods of moderate to vigorous exercise (cycling) in sedentary subjects when energy expenditure is comparable. PLoS One. 2013; 8 (2): e55542.
16. Freedson PS, Melanson E, Sirard J. Calibration of the Computer Science and Applications, Inc. accelerometer. Med Sci Sports Exerc. 1998; 30 (5): 777–81.
17. Gardiner PA, Eakin EG, Healy GN, Owen N. Feasibility of reducing older adults’ sedentary time. Am J Prev Med. 2011; 41 (2): 174–7.
18. Hagstromer M, Troiano RP, Sjostrom M, Berrigan D. Levels and patterns of objectively assessed physical activity—a comparison between Sweden and the United States. Am J Epidemiol. 2010; 171 (10): 1055–64.
19. Hamilton MT, Hamilton DG, Zderic TW. Exercise physiology versus inactivity physiology: an essential concept for understanding lipoprotein lipase regulation. Exerc Sport Sci Rev. 2004; 32 (4): 161–6.
20. Hamilton MT, Owen N. Sedentary behavior and inactivity physiology. In: Bouchard C, Blair SN, Haskell WL, editors. Physical Activity and Health. 2nd ed. Champaign (IL): Human Kinetics; 2012. pp. 59–61.
21. Haskell WL, Lee IM, Pate RR, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation. 2007; 116 (9): 1081–93.
22. Healy GN, Clark BK, Winkler EA, Gardiner PA, Brown WJ, Matthews CE. Measurement of adults’ sedentary time in population-based studies. Am J Prev Med. 2011; 41 (2): 216–27.
23. Healy GN, Dunstan DW, Salmon J, et al. Objectively measured light-intensity physical activity is independently associated with 2-h plasma glucose. Diabetes Care. 2007; 30 (6): 1384–9.
24. Healy GN, Eakin EG, Lamontagne AD, et al. Reducing sitting time in office workers: short-term efficacy of a multicomponent intervention. Prev Med. 2013; 57 (1): 43–8.
25. Healy GN, Matthews CE, Dunstan DW, Winkler EA, Owen N. Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003–06. Eur Heart J. 2011; 32 (5): 590–7.
26. Larsen BA, Allison MA, Kang E, et al. Associations of physical activity and sedentary behavior with regional fat deposition. Med Sci Sports Exerc. 2013; 46 (3): 520–8.
27. Lee IM, Paffenbarger RS Jr. Associations of light, moderate, and vigorous intensity physical activity with longevity. The Harvard Alumni Health Study. Am J Epidemiol. 2000; 151 (3): 293–9.
28. Lynch BM, Friedenreich CM, Winkler EA, et al. Associations of objectively assessed physical activity and sedentary time with biomarkers of breast cancer risk in postmenopausal women: findings from NHANES (2003–2006). Breast Cancer Res Treat. 2011; 130 (1): 183–94.
29. Matthews CE. Calibration of accelerometer output for adults. Med Sci Sports Exerc. 2005; 37 (11 Suppl): S512–22.
30. McGuire KA, Ross R. Incidental physical activity and sedentary behavior are not associated with abdominal adipose tissue in inactive adults. Obesity. 2012; 20 (3): 576–82.
31. Miyashita M, Park JH, Takahashi M, Suzuki K, Stensel D, Nakamura Y. Postprandial lipaemia: effects of sitting, standing and walking in healthy normolipidaemic humans. Int J Sports Med. 2013; 34 (1): 21–7.
32. National Health and Nutrition Examination Survey (NHANES) Web site [Internet]. Atlanta (GA): Centers for Disease Control and Prevention (CDC); [cited 2014]. Available from: Accessed February 17, 2015.
33. Ng M, Fleming T, Robinson M, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. [Internet]. 2014 [cited 2014 July 28]. Available from: doi: 10.1016/S0140-6736(14)60460-8.
34. Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation. 2003; 107 (3): 499–511.
35. Sparling PB, Howard BJ, Dunstan DW, Owen N. Recommendations for physical activity in older adults. Br Med J. 2015; 350: h100.
36. Stamatakis E, Davis M, Stathi A, Hamer M. Associations between multiple indicators of objectively-measured and self-reported sedentary behaviour and cardiometabolic risk in older adults. Prev Med. 2012; 54 (1): 82–7.
37. Thorp AA, Kingwell BA, Sethi P, Hammond L, Owen N, Dunstan DW. Alternating bouts of sitting and standing attenuates postprandial glucose responses. Med Sci Sports Exerc. 2014 [cited 2014 July 8]. Available from: doi: 10.1249/MSS.0000000000000337.
38. Tikkanen O, Haakana P, Pesola AJ, et al. Muscle activity and inactivity periods during normal daily life. PLoS One. 2013; 8 (1): e52228.
39. Troiano RP, Berrigan D, Dodd KW, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008; 40 (1): 181–8.
40. Winkler EA, Gardiner PA, Clark BK, Matthews CE, Owen N, Healy GN. Identifying sedentary time using automated estimates of accelerometer wear time. Br J Sports Med. 2012; 46 (6): 436–42.


Supplemental Digital Content

© 2015 American College of Sports Medicine