Sedentary Behavior, Physical Activity, and Markers of Health in Older Adults

Keith P. Gennuso; Ronald E. Gangnon; Charles E. Matthews; Keith M. Thraen-Borowski; Lisa H. Colbert


Med Sci Sports Exerc. 2013;45(8):1493-1500. 

In This Article

Abstract and Introduction


Introduction: The purpose of this study was to examine the association between sedentary behavior (SB), cardiometabolic risk factors, and self-reported physical function by level of moderate–vigorous physical activity (MVPA).

Methods: Cross-sectional analysis was completed on 1914 older adults age ≥65 yr from the 2003–2006 U.S. National Health and Nutrition Examination Survey. MVPA and SB were derived from ActiGraph accelerometers worn for 1 wk. MVPA was categorized as sufficient to meet the current U.S. guidelines (≥150 min·wk−1) or not; SB was split into quartiles. Various biomarkers were examined in laboratory analyses and physical exams, and the number of functional limitations was self-reported. Statistical interaction between SB and MVPA on the biomarker associations was the primary analysis, followed by an examination of their independent associations with relevant covariate adjustment.

Results: Average SB was 9.4 ± 2.3 h·d−1 (mean ± SD), and approximately 35% were classified as sufficiently active. Overall, no significant meaningful statistical interactions were found between SB and MVPA for any of the outcomes; however, strong independent positive associations were found between SB and weight (P < 0.01), body mass index (P < 0.01), waist circumference (P < 0.01), C-reactive protein (P < 0.01), plasma glucose (P = 0.04), and number of functional limitations (P < 0.01) after adjustment for MVPA. Similarly, MVPA was negatively associated with weight (P = 0.01), body mass index (P < 0.01), waist circumference (P < 0.01), diastolic blood pressure (P = 0.04), C-reactive protein (P < 0.01), and number of functional limitations (P < 0.01) after adjustment for SB.

Conclusions: The results suggest that sufficient MVPA did not ameliorate the negative associations between SB and cardiometabolic risk factors or functional limitations in the current sample and that there was independence on a multiplicative scale in their associations with the outcomes examined. Thus, older adults may benefit from the joint prescription to accumulate adequate MVPA and avoid prolonged sitting.


Sedentary behavior (SB), defined as participation in activities such as sitting and reclining during waking hours that do not increase energy expenditure substantially,[22] has been gaining increased focus in physical activity and health research. Traditionally, SB has been used to describe limited participation in moderate–vigorous physical activity (MVPA); however, research linking prolonged sitting to adverse health outcomes while controlling for MVPA has provided the evidence needed to identify SB as an independent behavior of interest. In particular, SB or proxy measures such as television viewing have been shown to be associated with an increased risk of obesity,[14] cancer,[12] metabolic syndrome,[9] type 2 diabetes,[14] and all-cause mortality[16] in various populations. There has been very limited research focusing on adults age ≥65 yr despite evidence to suggest older adults represent the most sedentary age group, spending approximately 60% to 70% of their waking hours in SB.[19]

Studies that have examined the association between SB and health outcomes in older adults have typically focused on metabolic syndrome and other cardiometabolic risk factors. For instance, increased risk of overweight/obesity, as measured by body mass index (BMI), was associated with a self-reported TV viewing time of >840 min·wk−1 in 1806 Japanese older adults.[15] Also, significant associations have been found for objectively measured SB time with lower limb,[6] central,[25] and total body adiposity[6,25] measured using dual energy x-ray absorptiometry. Data from a sample of 649 older adults who wore accelerometers in the Health Survey for England demonstrated direct associations for self-reported SB and TV time with a ratio of total to HDL cholesterol and for objectively measured SB with waist circumference.[24] In the AusDiab study, increased odds of having metabolic syndrome and several of its components were found in higher quartiles of self-reported SB and TV time.[11] Finally, data from a nationally representative U.S. sample of 1367 older adults from The National Health and Nutrition Examination Survey (NHANES) were used to examine the risk of metabolic syndrome associated with different parameters of accelerometer-derived SB.[1] An increased odds of metabolic syndrome in those with greater time spent in SB was found.

Most of the studies discussed provide evidence for an association of SB with metabolic syndrome and other cardiometabolic risk factors while controlling for participation in physical activity using multivariate regression techniques. We believe an examination of the association between SB and health outcomes stratified by level of MVPA, in addition to analyses statistically removing its effect, is worthwhile because it specifically answers the question of whether SB is associated with deleterious health effects in those that are sufficiently or insufficiently active. In addition, to our knowledge, just one study[21] has examined the association between physical function and objectively measured SB time in older adults, and this study did not examine the interaction between SB and MVPA. Therefore, the purpose of this study was to examine the association between SB, various cardiometabolic risk factors, and self-reported physical function by strata of MVPA participation. We hypothesized that there would be a statistical interaction between MVPA and SB on the various outcomes in a sample of older adults from the 2003–2006 NHANES. More specifically, we hypothesized there would be an association between SB and the outcomes in those with insufficient MVPA, but not in those who met the current physical activity guidelines.[27]