Abstract
Sedentary behavior is linked to numerous poor health outcomes.
Purpose
To determine the effects of 7 days of increased sitting in free-living individuals on markers of cardiometabolic risk.
Methods
Ten, recreationally active participants (>150 min of moderate intensity physical activity per week, mean (SD) age; 25.2 y (5.7), BMI 24.9 m˙kg−2 (4.3)) completed a 7-day baseline period and a 7-day sedentary condition in their free-living environment. During baseline participants maintained normal activity. Following baseline, participants completed a 7-day sedentary condition. Participants were instructed to sit as much as possible, limit standing and walking and refrain from structured exercise and leisure time physical activity. The activPAL™ was used to assess sedentary behavior and physical activity. Fasting lipids, glucose and insulin were measured and an oral glucose tolerance test (OGTT) was performed following baseline and sedentary conditions.
Results
In comparison to baseline, total sedentary time (mean change (95% CI); 14.9% (10.2, 19.6)), and time in prolonged/uninterrupted sedentary bouts significantly increased, while the rate of breaks from sedentary time was significantly reduced (21.4% (6.9, 35.9)). For the OGTT, 2 h plasma insulin (mean change (95% CI); 38.8 uU˙ml−1 (10.9, 66.8)) and area under the insulin curve (3074.1 uU˙ml−1˙120 min−1 (526.0, 5622.3)) were significantly elevated after the sedentary condition. Lipid concentrations did not change. Change in 2 h insulin was negatively associated with change in light intensity activity (r=-0.62) and positively associated with change in time in sitting bouts longer than 30 (r=0.82) and 60 min (r=0.83).
Conclusion
Increased free-living sitting negatively impacts markers of cardiometabolic health and specific features of sedentary behavior (e.g. time in prolonged sitting bouts) may be particularly important.
Keywords: Sedentary, Metabolism, Cardiometabolic, Glucose, Insulin, Physical Activity, Inactivity
Introduction
Sedentary behaviors are behaviors performed in the seated/lying position and require low levels of energy expenditure (e.g. < 1.5 METS) (19). Americans spend an estimated 55 to 70% of waking hours sedentary (17) and even regular exercisers spend large portions of the day engaged in sedentary behaviors (2). Epidemiologic evidence indicates habitual sedentary behavior, and sedentary time independent of physical activity (PA), are associated with a host of poor health outcomes, including increased risk of obesity (9,10), metabolic syndrome (4,22), type 2 diabetes (9,10,12), cardiovascular disease (5,11), and premature mortality (24). Emerging evidence indicates that in addition to total time spent sedentary, other novel features such as the number of interruptions (or “breaks”) from sedentary behaviors, may influence the physiologic response to habitual sedentary behavior (3,7,8).
Because sedentary behaviors are ubiquitous and spontaneous (16), understanding their physiologic consequences has been challenging. Traditionally, research protocols have used bed-rest in humans and hind-limb immobilization in rodents to study how sedentary behaviors affect physiologic outcomes. These investigations indicate that insulin action (23) and lipid metabolism (1,27) negatively respond to sustained sedentary time and suggest changes to insulin signaling, glucose transport, and lipoprotein lipase (LPL) activity may govern these consequences (1,27). Although these data offer insight into the physiologic consequences of extreme sedentariness, their generalizability to more typical free-living settings is uncertain as even the most sedentary, but otherwise healthy individuals take breaks from sedentary behaviors to perform basic hygiene and activities of daily living.
Recent research has expanded on bed rest models by exposing participants to short-term experimental conditions more relevant to free-living sedentary pursuits (23). In a controlled laboratory study, Dunstan et al (3) reported that short (2-min) light and moderate intensity interruptions improve postprandial glucose and insulin responses compared to prolonged sedentary time. Similarly, Duvivier et al reported exercise does not fully negate the detrimental effects of sitting all day compared to day of low-intensity ambulatory activities with minimal sitting (6). These models are more representative of free-living sedentary behaviors than bed-rest studies; however the experimental conditions are still not typical of free-living behavior. Ultimately the goal of this research area is to determine if there is a causal relationship between sedentary behaviors and poor health outcomes and to establish public health recommendations to decrease sedentary time, if warranted. To do this, the interventions must reflect real-world sedentary behavior, where individuals are free to sit and break from sitting for any amount of time. It is also necessary to expand the intervention duration beyond 24-hours.
The primary objective of this study was to investigate the cardiometabolic response to seven days of increased sedentary behaviors in free-living individuals. A secondary objective was to investigate whether the cardiometabolic response could be linked to specific features of habitual sedentary behavior. We used the activPAL™ activity monitor to precisely measure change in total time sedentary, number, time and intensity of breaks from sedentary time, break-rate (15), step count and time in sedentary bouts longer than 20, 30 and 60 minutes, respectively.
Methods
Ethics Statement
The University of Massachusetts institutional review board (IRB) approved this study and all participants completed a health history questionnaire and an informed consent document approved by the University of Massachusetts IRB.
Recruitment and Eligibility
Men and women 18-35 y were recruited for this study. Eligible participants were in good physical health (no diagnosed cardiovascular, pulmonary, metabolic, joint, or chronic diseases) and currently participating in at least 150 minutes of moderate-intensity activity per week.
Baseline Visit
Participants reported to the Physical Activity and Health Laboratory following a 12-hour overnight fast. Using a standard floor stadiometer and physicians' scale (Detecto; Webb City, MO), height and weight were measured to the nearest 0.25 cm and 0.1 kg, respectively. Participants also completed a short survey asking about their current physical activity status (PAS). Participants were asked to choose a number which best described their activity in a normal week. Possible responses ranged from 0 to 7 with 0 corresponding to “avoided walking or exertion (e.g. always used the elevator, drove whenever possible instead of walking)”, and 7 corresponding to “ran more than 10 miles per week or spent over 3 hours per week in comparable physical activity”. To be eligible to continue with the study, participants must have reported 5 or greater (“Ran 1 to 5 miles per week or spent 30 to 60 minutes per week in comparable physical activity.”) on the PAS. The PAS was used as an initial screening tool and habitual activity levels were later verified using the accelerometer (described below).
Experimental Procedures
At baseline, participants' physical activity and sedentary behavior were measured for seven days. During this time, participants maintained their normal daily activity, including exercise. Accelerometer data from the baseline period were used to verify participants self-reported activity levels. If participants did not perform at least 150 minutes of moderate intensity activity, they were no longer considered eligible. No potential participants were deemed ineligible after the wearing the accelerometer. Within 24-hours of completing the baseline assessment, participants completed a seven day sedentary condition. Participants were instructed to increase their time in sedentary behaviors as much as possible, limit time standing and walking, and refrain from structured exercise and leisure time and occupational physical activity. Participants were instructed to accumulate no more that 5000 steps˙day−1 and wore an Omron pedometer to facilitate compliance. This device is valid for measuring steps per day (21) and has been used to provide referent goals for individuals to meet activity guidelines (26). Other studies have successfully used a prescription to decrease steps to study reduced activity (14,18). However we also used an accelerometer to precisely measure features of sedentary behavior. Data from the pedometer were not used in analyses, but were only used to provide participants real-time feedback on behavior.
Detailed Measurement of Active and Sedentary Behaviors
The activPAL™ (PALTechnologies: Glasgow, Scotland) activity monitor is a small (2.0 × 1.4 × 0.3 inches) and light (20.1 grams) device that uses accelerometer-derived information about thigh position to estimate time spent in different body positions (i.e., sitting/lying, standing and stepping). During each condition participants wore an activPAL™ on their right thigh for waking hours. The device was attached using a non-allergenic adhesive pad and positioned on the midline of the thigh, one-third of the way between the hip and knee. The time-stamped “event” data file from the activPAL™ software (version 5.8.5) was used to identify the duration of each sitting/lying, standing and stepping bout. The device also estimates intensity by assigning MET values of 1.25 and 1.4 to sitting and standing events, respectively and uses cadence to estimate METs for stepping events. The event file was converted to second-by-second data and a customized R (www.r-project.org) program was used to estimate the following 16 metrics:
MET-Hours – The sum of METs multiplied by time (hours).
Percent time sedentary – The sum of minutes spent in sitting/lying events divided by the total minutes the activPAL™ was worn.
Percent time light – The sum of minutes spent in light intensity divided by the total minutes the activPAL™ was worn. Note: Although the activPAL™ assigns a MET value of 1.4 to standing events, we considered all standing events as light intensity (e.g., in order to be considered sedentary, a seated or lying posture was required).
Percent time MVPA – The sum of minutes spent in MVPA divided by the total minutes the activPAL™ was worn.
Percent time active – The sum of minutes spent in either light intensity or MVPA divided by the total minutes the activPAL™ was worn.
Time sedentary (min) – The sum of minutes spent in sitting/lying events.
Time light (min) – The sum of minutes spent in light intensity (1.5-2.99 METs). Note: Although the activPAL™ assigns a MET value of 1.4 to standing events, we considered all standing events as light intensity (e.g., in order to be considered sedentary, a seated or lying posture was required).
Time MVPA (min) – The sum of minutes spent in MVPA (≥ 3METs). Note: any minute where intensity was ≥ 3METs was counted (i.e. activity did not have to be performed in bouts).
Time active (min) – The sum of minutes spent in light intensity (1.5-2.99 METs) or MVPA (≥ 3METs) (Total light time + total MVPA time).
Guideline minutes (min) – Sum of minutes spent in bouts of activity that qualify towards meeting the physical activity guidelines: activity of at least moderate intensity (≥ 3METs) and lasting at least ten consecutive minutes.
Breaks (count) – The number of times a sitting/lying event was followed by a standing or stepping event.
Break-rate (number breaks˙sed-hr−1) – The number of breaks per sedentary hour (15).
Step count (count) – The cumulative number of steps taken.
Time in sedentary bouts longer than 20 minutes (min) – Sum of minutes spent in sitting bouts that are at least 20 minutes in duration. Any time an individual interrupted a sitting event with a standing or stepping event of any length the duration of the sitting bout was ended.
Time in sedentary bouts longer than 30 minutes (min) – Sum of minutes spent in sitting bouts that are at least 30 minutes in duration. Any time an individual interrupted a sitting event with a standing or stepping event of any length, the duration of the sitting bout was ended.
Time in sedentary bouts longer than 60 minutes (min) – Sum of minutes spent in sitting bouts that are at least 60 minutes in duration. Any time an individual interrupted a sitting event with a standing or stepping event of any length the duration of the sitting bout was ended.
Dietary Assessment
During the baseline period and the sedentary condition participants were asked to maintain their normal dietary habits. On the days prior to assessing cardiometabolic outcome variables, participants were asked to keep their diets as identical as possible. This was verified by having participants complete a 24 h dietary recall (National Cancer Institute: ASA24. http://riskfactor.cancer.gov/tools/instruments/asa24/). The ASA24, or automated self-administered recall, is a web-based tool that enables participant administered 24 h recalls and is freely available through NCI.
Markers of Cardiometabolic Health
Oral Glucose Tolerance Test: Following baseline assessment and sedentary condition participants reported to the laboratory following a 12-hour overnight fast. A catheter was inserted into a forearm vein, fasting blood samples were taken followed by a standard 2-hour oral glucose tolerance test (OGTT). Subjects ingested 75g of glucose (Sun Dex, Fisher Healthcare, Houston,TX) within 5 minutes, and blood samples were collected every 30 minutes (0, 30, 60, 90 and 120 minutes) for the next 2 hours. Samples were centrifuged immediately at (3,000 × g) for 15 minutes and plasma was aliquotted into polystyrene tubes and stored at -80°C until analysis. The following variables were used as markers of cardiometabolic health.
Fasting Lipids – Fasting triglyceride (TG) and cholesterol (total, HDL, LDL) samples were collected in sterile syringes and transferred to serum vacutainers for analysis. Plasma triglycerides were determined using an enzymatic colorimetric assay kit (Sigma Chemical, St. Louis, MO), and total cholesterol and HDL were analyzed using the cholesterol oxidase method (Analox Instruments, Lunenberg, MA). LDL was calculated from measured TG, total cholesterol and HDL levels (LDL = total cholesterol - (TG / 5 + HDL)).
Fasting and two-hour glucose concentration – Glucose concentrations were determined using the glucose oxidase method (GL5 Analox Analyzer [Analox Instruments, Lunenberg, MA]).
Fasting and two-hour insulin concentration – Insulin concentrations were determined using a radioimmunoassay kit (Millipore Corporation; Chicago, IL) specific for human insulin. Higher insulin concentrations suggest reduced peripheral insulin sensitivity, as more insulin is needed to dispose of the same concentration of glucose.
Area under the glucose curve (glucose-AUC) – Glucose concentrations from the 5 time points were used to calculate glucose-AUC using the trapezoidal method.
Area under the insulin curve (insulin-AUC) – Insulin concentrations from the 5 time points were used to calculate insulin-AUC using the trapezoidal method. Although data are limited, insulin-AUC has been associated with all-cause and cardiovascular disease mortality (21) and similar to fasting and two-hour insulin concentrations, a higher insulin-AUC in response to an OGTT suggests reduced peripheral insulin sensitivity as more insulin is needed to dispose of the same concentration of glucose.
Matsuda Index – Insulin action was estimated using the whole body insulin sensitivity index (10,000/square root of [fasting glucose × fasting insulin] × [mean glucose × mean insulin during OGTT]) established by Matsuda and DeFronza (composite-insulin sensitivity index (C-ISI)) (16). C-ISI represents a composite of hepatic and peripheral tissues and considers insulin sensitivity in the basal state after a carbohydrate load and is strongly correlated (r=0.73) with the direct measure of insulin sensitivity derived from the hyperinsulinemic-euglycemic clamp (16).
Data Cleaning and Reduction
ActivPAL™ data were downloaded and exported to csv files and all data cleaning and processing was performed using the statistics package and computing language R. Wear time was determined from detailed monitor logs that participants completed daily. Participants recorded the time the monitor was put on in the morning after waking and removed at night before bed, and anytime the monitors were removed during the day and the reason why the monitor was removed (e.g. shower). All non-wear time and sleep time were removed from the accelerometer file before analysis. At least ten hours of activPAL™ data were required for a day to be considered valid and at least four valid days (including one weekend day) were required for the condition to be considered valid (25).
Sample Size Calculations and Statistical Evaluation
Statistical analyses were performed using R-software programs and significance level was p<0.05. Sample size calculations were based on an expected 23.4% change in insulin-AUC following increased sedentary time, estimates of within person variability of 23.6% (17), and assuming a correlation coefficient of 0.6 between repeated measures of outcomes. With 10 participants we estimated approximately 90% power to detect the effect size between repeated measures. Repeated measures linear mixed models with likelihood ratio testing were used to evaluate the change in activPAL™ variables and markers of cardiometabolic health from the baseline to sedentary condition (primary objective). Linear regression models were fit to evaluate the relationship between change in cardiometabolic variables and activPAL™ variables (secondary objective).
Results
Ten participants (4 males, 6 females) completed the study. Participants were relatively young (mean (SD) age; 25.2 y (5.7)) and lean (BMI 24.9 kg˙m−2 (4.3)) (Table 1).
Table 1. Participant Characteristics (mean (SD)).
| N=10: 4 Males, 6 Females | |
|---|---|
| Age (yrs) | 25.2 (5.7) |
| Body Mass (kg) | 72.8 (20.2) |
| Height (cm) | 170.0 (11.9) |
| BMI (kg˙m−2) | 24.9 (4.3) |
| Waist Circumference (cm) | 78.3 (12.9) |
BMI = body mass index
Activity and Sedentary Behavior Variables
During the sedentary condition participants significantly decreased MET-hours (mean Δ (95% CI) = -3.5 MET-hrs (-4.9, -2.5), time in MVPA (-52.7 min (-65.8, -39.6)), time in light intensity (-87.6 min (-120.5, -54.8)) and guideline minutes (-41.7 min (-54.4, -29.1)). Total sedentary time (mean Δ (95% CI); 14.9% (10.2, 19.6)), and time in sedentary bouts longer than 20, 30 and 60 minutes significantly increased, while the rate of breaks from sedentary time was significantly reduced (Table 2). According to both the AP and the Omron pedometer, step count significantly decreased following the sedentary condition (AP: mean Δ (95% CI) = -6850.2 (-8578.3, -5122.1); Omron: -6522.9, (-8042.1, -5003.8). Although the decrease in Omron pedometer step count was slightly less the Omron estimate was not significantly different than the AP estimate. The purpose of the Omron was only to facilitate condition compliance by providing participants real-time feedback about behavior and thus all statistical analyses and step data presented in tables and discussed are from the AP. Activity and sedentary behavior variables (mean (95% CI)) for the baseline period and sedentary condition are reported in Table 2.
Table 2. Intervention variables (mean (95% CI)).
| Baseline Period | Sedentary Condition | |
|---|---|---|
| Activity and Sedentary Behavior Variables Measured by the activPAL | ||
| MET-Hours | 22.8 (21.3,24.2) | 19.2 (18.5, 20.0)* |
| Time Sedentary (%) | 61.3 (57.4, 65.3) | 76.2 (74.2, 78.3)* |
| Time Light (%) | 28.4 (25.3, 31.5) | 19.1 (17.5, 20.7)* |
| Time MVPA (%) | 10.2 (9.1, 11.4) | 4.4 (3.7, 5.0)* |
| Time Active (%) | 38.6 (34.7, 42.5) | 23.5 (21.4, 25.5)* |
| Time Sedentary (min) | 523.6 (492.1, 555.1) | 631.3 (592.9, 669.8)* |
| Time Light (min) | 245.1 (209.1, 281.1) | 157.4 (143.5, 171.4)* |
| Time MVPA (min) | 88.1 (76.3, 99.9) | 35.4 (30.8, 40.0)* |
| Time Active (min) | 333.2 (287.8, 378.6) | 192.9 (176.8, 208.9)* |
| Guideline Minutes (min) | 43.9 (30.9, 56.8) | 2.1 (0.4, 3.8)* |
| Breaks from Sedentary (count) | 49.1 (40.5, 57.8) | 46.4 (37.9, 54.9) |
| Break-Rate (number brks˙sed-hr−1) | 5.8 (4.7, 6.9) | 4.6 (3.8, 5.3)* |
| Step Count (count) | 10,221 (9,178, 11,264) | 4,308 (3,868, 4,749)* |
| Time in Sedentary Bouts ≥ 20 m (min) | 148.6 (136.1, 161.0) | 190.7 (168.4, 213.1)* |
| Time in Sedentary Bouts ≥ 30 m (min) | 136.6 (115.4, 157.9) | 183.6 (162.0, 205.3)* |
| Time in Sedentary Bouts ≥ 60 m (min) | 86.0 (66.5, 105.5) | 118.6 (95.3, 141.8)* |
| Cardiometabolic Markers | ||
| BMI (kg˙m−2) | 24.9 (22.2, 27.5) | 24.8 (21.6, 27.9) |
| Waist Circumference (cm) | 78.3 (70.3, 86.3) | 77.0 (68.2, 85.8) |
| Fasting Plasma Glucose (mg˙dL−1) | 93.9 (89.5, 98.2) | 94.4 (87.2, 101.5) |
| Fasting Plasma Insulin (uU˙ml−1) | 14.1 (10.2, 18.0) | 15.5 (11.2, 19.8) |
| 2 Hr. Plasma Glucose (mg˙dL−1) | 97.1 (80.0, 114.2) | 108.4 (94.0, 122.8) |
| 2 Hr. Plasma Insulin (mg˙dL−1) | 47.3 (26.8, 67.7) | 86.1 (61.3, 110.9)* |
| AUC-Glucose ((mg˙dL−1˙120min−1)) | 15,252 (13,284, 17,232) | 16,260 (14,220, 18,084) |
| AUC-Insulin ((uU˙dl−1˙120min−1)) | 9,756 (8,040, 11,484) | 12,840 (9,816, 15,852)* |
| C-ISI | 2.9 (2.3-3.5) | 2.4 (1.8-3.1)* |
| Total Cholesterol (mg˙dL−1) | 179.4 (171.6, 187.2) | 183.3 (175.5, 195.0) |
| LDL (mg˙dL−1) | 169.1 (159.8-178.3) | 171.4 (163.3, 179.5) |
| HDL (mg˙dL−1) | 58.5 (51.5, 65.5) | 58.5 (50.3, 66.3) |
| Triglycerides (mg˙dL−1) | 116.6 (82.8, 150.4) | 138.0 (96.1, 177.1) |
Activity and sedentary behavior variables are presented as mean per day. AUC = area under the curve. C-ISI = composite insulin sensitivity index.
Dietary Assessment
According diet recalls total energy (Baseline: 1672.7 (1214.6, 2130.8), Sedentary: 1671.8 (1146.5, 2197.0)) and macronutrient content (Baseline: 45.9% (34.0, 57.9) CHO, 35.5% (22.5, 48.5) fat, 18.5% (10.2, 26.9) protein. Sedentary: 40.7% (27.3, 54.2) CHO, 39.2% (25.2, 53.3) fat, 20.0% (12.1, 27.9) protein) did not differ during the baseline period and sedentary condition.
Markers of Cardiometabolic Health
Glucose and Insulin Response: Following the sedentary condition, fasting glucose and insulin concentrations did not change from baseline. Glucose concentrations also remained stable in response to a glucose load (OGTT). Conversely, 2 hr. plasma insulin (mean Δ (95% CI); 38.8 uU˙ml−1 (10.9, 66.8)) and area under the insulin curve (3074.1 uU˙ml−1˙120 min−1 (526.0, 5622.3)) were significantly elevated in response to the glucose load (Figure 1), suggesting more insulin was needed to dispose of the same amount of glucose. This resulted in a significant 17.2% decrease in C-ISI. Cardiometabolic variables (mean (95% CI)) for baseline and the sedentary condition are reported in Table 2 and illustrated in Figure 1.
Figure 1.
OGTT glucose and insulin responses. Mean minute values for glucose (top left) and mean glucose AUC (top right) show an elevated, but not significant increase in glucose response following the sedentary condition. Mean minute values for insulin (bottom left) and mean insulin-AUC (bottom right) show a significant increase in 2 h plasma insulin and insulin AUC following the sedentary condition compared to baseline. AUC = area under the curve. * Significantly different than baseline (p<0.05).
Body mass, BMI, waist circumference and fasting lipids: There were no significant differences in any fasting lipid (TG, total cholesterol, HDL, LDL) values following the sedentary condition. Body mass, BMI and waist circumference did not change from baseline to post-sedentary condition (Table 2).
Secondary Analyses
Linear regression was used to evaluate the association between change in activity and sedentary behavior variables and change in insulin action. Because 2 hr. plasma insulin, insulin-AUC and C-ISI were the only cardiometabolic variables to significantly change following the sedentary condition, data are presented for these variables only. Change in 2 hr. plasma insulin was negatively associated with change in percent of time in light intensity activity (r = -0.62, p<0.05) and positively associated with change in time in sedentary bouts longer than 30 min (r = 0.82, p<0.01) and 60 min (r = 0.83, p<0.01). When change in time in MVPA was included in the models, the significant associations of time in sedentary bouts longer than 30 and 60 minutes persisted (p<0.05), while the association with percent of time in light intensity activity was slightly attenuated (p=0.09). Although not significant, change in total sedentary time (r = 0.57, p = 0.09) and break rate (-0.57, p = 0.09) were moderately correlated with change in 2 hr. plasma insulin. Change in insulin-AUC and C-ISI were not associated with any activity or sedentary behavior variable (Table 3).
Table 3. Association between change in activity and sedentary behavior variables and markers of insulin action.
| 2 Hr. Plasma Insulin | Insulin-AUC | C-ISI | |||||||
|---|---|---|---|---|---|---|---|---|---|
| β | r | p | β | r | p | β | R | p | |
| MET-Hours | 2.20 | 0.08 | 0.82 | -4.57 | -0.23 | 0.53 | 0.01 | -0.20 | 0.96 |
| Time Sedentary (%) | 2.91 | 0.49 | 0.15 | 5.72 | 0.13 | 0.73 | 1.59 | 0.20 | 0.58 |
| Time Light (%) | -4.89 | -0.62 | 0.05* | -9.74 | -0.16 | 0.65 | -1.28 | -0.12 | 0.74 |
| Time MVPA (%) | -2.49 | -0.13 | 0.73 | 1.78 | 0.01 | 0.97 | -7.56 | -0.28 | 0.43 |
| Time Active (%) | -3.18 | -0.52 | 0.12 | -0.55 | -0.12 | 0.75 | -0.02 | -0.18 | 0.61 |
| Time Sedentary (min) | 0.29 | 0.57 | 0.09 | 0.02 | 0.05 | 0.89 | 0.00 | 0.18 | 0.62 |
| Time Light (min) | -0.46 | -0.55 | 0.10 | -0.18 | -0.27 | 0.44 | 0.00 | -0.4 | 0.92 |
| Time MVPA (min) | -0.28 | -0.13 | 0.72 | -0.15 | -0.09 | 0.80 | 0.00 | -0.20 | 0.57 |
| Time Active (min) | -0.30 | -0.46 | 0.19 | -0.12 | -0.24 | 0.51 | -0.01 | -0.09 | 0.80 |
| Guideline Minutes (min) | -0.06 | -0.03 | 0.94 | -0.29 | -0.17 | 0.63 | 0.00 | -0.03 | 0.93 |
| Breaks from Sedentary (count) | -2.54 | -0.50 | 0.14 | -1.50 | -0.39 | 0.27 | 0.03 | 0.47 | 0.17 |
| Break-Rate (number brks˙sed-hr−1) | -18.75 | -0.57 | 0.09 | -3.02 | -0.12 | 0.74 | 0.01 | 0.03 | 0.94 |
| Step Count (count) | -0.01 | -0.17 | 0.63 | -0.01 | -0.13 | 0.72 | 0.00 | -0.21 | 0.56 |
| Time in Sedentary Bouts ≥ 20 m (min) | 0.56 | 0.46 | 0.18 | 0.22 | 0.24 | 0.50 | 0.00 | 0.21 | 0.56 |
| Time in Sedentary Bouts ≥ 30 m (min) | 0.93 | 0.82 | <0.01* | 0.34 | 0.39 | 0.26 | 0.00 | 0.04 | 0.91 |
| Time in Sedentary Bouts ≥ 60 m (min) | 0.91 | 0.83 | <0.01* | 0.39 | 0.48 | 0.16 | 0.00 | -0.07 | 0.87 |
AUC = area under the curve, C-ISI = composite insulin sensitivity index
Discussion
The primary objective of this study was to evaluate the cardiometabolic effects of an acute increase in free-living sedentary behavior (seven-day treatment). Consistent with more extreme models of sedentary behavior (i.e. bed rest), we found that when moderately active individuals increase sedentary time, insulin action is significantly reduced. We also identified that this decrease in insulin action was associated with a decrease in percent of time in light intensity activity and an increase in time in prolonged sedentary bouts (> 30 min and > 60 min). To our knowledge this is the first intervention performed in free-living individuals that measured specific features of sedentary behavior and their effects on cardiometabolic outcomes.
Free-Living Model of Sedentary Behavior
It is well accepted that detraining and extreme sedentary behavior cause significant reductions in insulin action in both animal and human models (27). Stephens et al (23) reported that after one day of sustained sitting, insulin action was reduced 39%. In this study participants were confined to a wheel chair for more than 98% of the waking day. They were allowed to fidget and use their arms ad libitum but were not allowed to take breaks from sitting. Consistent with Stephens et al (23), we observed a 34.5% increase in insulin-AUC.
Importantly, our findings were observed using an ecologically valid design. In the present study participants were prohibited from exercise and encouraged to sit as much as possible for seven days, but were allowed to take breaks from sedentary behaviors and accumulate small amounts of light, moderate and vigorous intensity activity as dictated by their natural environment. This protocol resulted in participants spending (mean (95% CI)) 76.2% (74.2, 78.3) of the day sedentary and accumulating 4,308 (3,868-4,749) steps˙day−1 during the sedentary condition. These characteristics are consistent with population surveillance data (7-8, 17, 21-22) and support the notion that our model of sedentary behavior was similar to behavior typical of a habitually sedentary population. A longer intervention period may have resulted in larger and/or additional (e.g. negative changes in fasting lipids) metabolic consequences, but our results support emerging experimental evidence (3, 6, 23) that reduced insulin action may be an initial response to chronic sedentary behavior.
Detailed Measurement of Active and Sedentary Behaviors
Measuring free-living sedentary behaviors has traditionally been very difficult. Self-report (questionnaires, interviews, diaries etc.) consistently underestimates time spent sedentary and it has been shown repeatedly that waist worn activity monitors are biased and imprecise in characterizing features of sedentary behavior (13, 15). In the current study we used an activity monitor that was specifically designed to identify posture based on thigh position. With this device, we were able to capture details of sedentary behavior that are traditionally overlooked (e.g. breaks, duration of sedentary bouts). This led to novel findings indicating that less time in light intensity activity and more time in prolonged sitting bouts greater 30 and 60 minutes are directly related to increased 2 hr. plasma insulin. These results compliment those by Dunstan et al (3) who reported interrupting sitting time with structured bouts of light or moderate-intensity activity had positive effects on postprandial glucose and insulin responses. Although decreases in step count, time in MVPA and total time active (time in light intensity + time in MVPA) were not associated with decreased insulin action in the current study, more work is needed to distinguish the specific effects of time in prolonged sitting bouts vs. the type (e.g., stand vs. walk) and intensity (e.g., light vs. MVPA) of the activity performed during the interruption. Nonetheless, our results highlight the importance of simultaneously measuring and studying specific features of sedentary and active behavior in relation to health.
Two recent studies examined the effects of reduced steps˙day−1 on markers of cardiometabolic health in free-living people. After just 14 days and 3 days, significant reductions in insulin action were observed when healthy active volunteers reduced steps˙day−1 from 10,501 (SD: ± 808) to 1,344 (SD: ± 33) and 12,956 (SD: ± 769) to 4,319 (SD: ± 256), respectively (14, 18). These carefully designed studies advanced the field of sedentary behavior research by studying participants in free-living settings rather than using controlled laboratory models (e.g. bed rest, immobilization) and by using objective devices to measure the exposure. The current study extends this work by using the activPAL monitor to directly examine whether changes in detailed features of PA and sedentary behavior are associated with changes in health outcomes. Future studies that employ objective measurement approaches will likely reveal additional features of active and sedentary behavior important to health (e.g. stepping cadence, temporal features). Although beyond the scope of this paper, these approaches will also allow for careful examination of the association among PA and sedentary behavior variables and investigations into how individuals choose to re-allocate time when given a prescription to change sedentary behavior. Understanding how simultaneous changes in PA and sedentary behavior variables interact to impact health is a complex issue with significant implications for public health recommendations. Future studies that use direct, objective measurement techniques will continue to elucidate these relationships and will provide the evidence base for public health messages regarding sedentary behavior.
Strengths and Limitations
Important strengths of this study include the within subject design, the use of a free-living setting and the detailed measurement of multiple features of SB with a validated device designed to differentiate sitting and standing behaviors. Controlled laboratory studies have revealed important consequences of sustained sedentary behaviors. The current study expands this evidence through a free-living intervention that allowed for the simultaneous evaluation of important activity and sedentary behavior variables. This type of design has only recently been made possible through improvements in the objective measurement of free-living behavior.
The major limitation of this study is our small, homogenous sample. Despite our small sample we were able to identify important relationships between distinct sedentary behavior variables and reduced insulin action. However, future work is needed to confirm the current results and to uncover additional associations in larger, more diverse groups. For example, it may initially seem surprising that an independent association of time in MVPA and insulin action was not observed, but this may be due to the lack of inter-individual differences in how time in MVPA changed from baseline to the sedentary condition. Participants were relatively young, healthy and active. Additional work is needed to evaluate the potential influences of age, sex, BMI, activity status and health status. A secondary limitation of our study is that we did not control energy intake. Although energy state impacts the effects of increased sitting (23), any energy imbalance in the current study was minimal as evidenced by no change in body weight and unlikely significantly impacted outcomes. Additionally, recall data indicated participant diets were very similar the days immediately prior to each OGTT and cardiometabolic assessment. Nonetheless, future mechanistic studies would benefit from controlling and measuring energy intake. Lastly, in this study participants were instructed to refrain from structured and leisure time physical activity during the sedentary condition. Future work should address how increasing sedentary time, while maintaining structured exercise impacts cardiometabolic outcomes.
Summary
This study provides further evidence that when active individuals replace active time with sedentary time, markers of insulin action are negatively affected. The primary contribution of this study is that these results were observed using a novel free-living model of sedentary behavior where participants performed intermittent bouts of ambulatory activity characteristic of typical sedentary behavior, and active and sedentary behaviors were precisely measured using an objective monitoring tool. Advances in objective monitoring tools enable precise measurement of characteristics of sedentary behavior in free-living experimental models. We anticipate that applying these tools will continue to expose characteristics of sedentary and active behavior important in disease initiation and development.
Acknowledgments
The authors would like to acknowledge the members of the Physical Activity and Health and Energy Metabolism Laboratories at The University of Massachusetts, Amherst, especially Amanda Libertine, Natalia Petruski and Richard Viskochil. The authors would also like to thank all of the participants that volunteered for this study. We also acknowledge our funding source NIH RC1HL099557.
Footnotes
Conflict of Interest: The results of the present study do not constitute endorsement by ACSM. The authors have no conflicts of interest to report.
References
- 1.Bey L, Hamilton MT. Suppression of skeletal muscle lipoprotein lipase activity during physical inactivity: a molecular reason to maintain daily low-intensity activity. J Physiol. 2003;551:673–6. doi: 10.1113/jphysiol.2003.045591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Craft LL, Zderic TW, Gapstur SM, Vaniterson EH, Thomas DM, et al. Evidence that women meeting physical activity guidelines do not sit less: an observational inclinometry study. Int J Behav Nutr Phys Act. 2012:9–122. doi: 10.1186/1479-5868-9-122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Dunstan DW, Kingwell BA, Larsen R, Healy GN, Cerin E, et al. Breaking Up Prolonged Sitting Reduces Postprandial Glucose and Insulin Responses. Diabetes Care. 2012;35(5):976–83. doi: 10.2337/dc11-1931. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Dunstan DW, Salmon J, Owen N, Armstrong T, Zimmet PZ, et al. Associations of TV viewing and physical activity with the metabolic syndrome in Australian adults. Diabetologia. 2005;48:2254–2261. doi: 10.1007/s00125-005-1963-4. [DOI] [PubMed] [Google Scholar]
- 5.Dunstan DW, Thorp AA, Healy GN. Prolonged sitting: is it a distinct coronary heart disease risk factor? Curr Opin Cardiol. 2011;26:412–419. doi: 10.1097/HCO.0b013e3283496605. [DOI] [PubMed] [Google Scholar]
- 6.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. doi: 10.1371/journal.pone.0055542. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Healy GN, Dunstan DW, Salmon J, Cerin E, Shaw JE, et al. Breaks in Sedentary Time. Diabetes Care. 2008;31:661–666. doi: 10.2337/dc07-2046. [DOI] [PubMed] [Google Scholar]
- 8.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;32:590–597. doi: 10.1093/eurheartj/ehq451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hu FB. Sedentary lifestyle and risk of obesity and type 2 diabetes. Lipids. 2003;38:103–108. doi: 10.1007/s11745-003-1038-4. [DOI] [PubMed] [Google Scholar]
- 10.Hu FB, Li TY, Colditz GA, Willett WC, Manson JE. Television watching and other sedentary behaviors in relation to risk of obesity and type 2 diabetes mellitus in women. JAMA. 2003;289:1785–1791. doi: 10.1001/jama.289.14.1785. [DOI] [PubMed] [Google Scholar]
- 11.Hu FB, Stampfer MJ. Is type 2 diabetes mellitus a vascular condition? Arterioscler Thromb Vasc Biol. 2003;23:1715–1716. doi: 10.1161/01.ATV.0000094360.38911.71. [DOI] [PubMed] [Google Scholar]
- 12.Hu FB, Stampfer MJ, Colditz GA, Willet WC, Rimm EB. Physical activity and television watching in relation to risk for type 2 diabetes mellitus in men. Arch Intern Med. 2001;161:1542–1548. doi: 10.1001/archinte.161.12.1542. [DOI] [PubMed] [Google Scholar]
- 13.Kozey-Keadle S, Libertine A, Lyden K, Staudenmayer J, Freedson PS. Validation of wearable monitors for assessing sedentary behavior. Med Sci Sports Exerc. 2011;43:1561–1567. doi: 10.1249/MSS.0b013e31820ce174. [DOI] [PubMed] [Google Scholar]
- 14.Krogh-Madsen R, Thyfault JP, Broholm C, Mortensen OH, Olsen RH, et al. A 2-wk reduction of ambulatory activity attenuates peripheral insulin sensitivity. J Appl Physiol. 2010;108:1034–1040. doi: 10.1152/japplphysiol.00977.2009. [DOI] [PubMed] [Google Scholar]
- 15.Lyden K, Kozey Keadle SL, Staudenmayer JW, Freedson PS. Validity of two wearable monitors to estimate breaks from sedentary time. Med Sci Sports Exerc. 2012;44:2243–2252. doi: 10.1249/MSS.0b013e318260c477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Matsuda M, Defronzo R. Insulin sensitivity indices obtained from oral glucose tolerance testing. Diabetes Care. 1999;22(9):1462–70. doi: 10.2337/diacare.22.9.1462. [DOI] [PubMed] [Google Scholar]
- 17.Matthews CE, Chen KY, Freedson PS, Buchowski MS, Beech BM, et al. Amount of time spent in sedentary behaviors in the United States, 2003-2004. Am J Epidemiol. 2008;167:875–881. doi: 10.1093/aje/kwm390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mikus CROD, Libla JL, Taylor AM, Booth FW, Thyfault JP. Lowering physical activity impairs glycemic control in healthy volunteers. Med Sci Sports Exerc. 2012;44:225–231. doi: 10.1249/MSS.0b013e31822ac0c0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Owen N, Healy GN, Matthew CE, Dunstan DW. Too Much Sitting: The Population Health Science of Sedentary Behavior. Exercise and Sport Sciences Review. 2010;38:105–113. doi: 10.1097/JES.0b013e3181e373a2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pyorala M, Miettinen H, Laakso M, Pyorala K. Plasma insulin and all-cause, cardiovascular, and noncardiovascular mortality: The 22-year follow-up results of the Helsinki Policemen Study. Diabetes Care. 2000;23:1097–1102. doi: 10.2337/diacare.23.8.1097. [DOI] [PubMed] [Google Scholar]
- 21.Schneider PCS, Basset D. Pedometer measures of free-living physical activity: comparison of 13 models. Med Sci Sports Exerc. 2004;36:331–335. doi: 10.1249/01.MSS.0000113486.60548.E9. [DOI] [PubMed] [Google Scholar]
- 22.Sisson SB, Camhi SM, Church TS, Martin CK, Tudor-Locke C, et al. Leisure time sedentary behavior, occupational/domestic physical activity, and metabolic syndrome in U.S. men and women. Metab Syndr Relat Disord. 2009;7:529–536. doi: 10.1089/met.2009.0023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Stephens BR, Granados K, Zderic TW, Hamilton MT, Braun B. Effects of 1 day of inactivity on insulin action in healthy men and women: interaction with energy intake. Metabolism. 2011;60:941–949. doi: 10.1016/j.metabol.2010.08.014. [DOI] [PubMed] [Google Scholar]
- 24.Thorp AA, Owen N, Neuhaus M, Dunstan DW. Sedentary behaviors and subsequent health outcomes in adults a systematic review of longitudinal studies, 1996-2011. Am J Prev Med. 2011;41:207–215. doi: 10.1016/j.amepre.2011.05.004. [DOI] [PubMed] [Google Scholar]
- 25.Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008;40:181–188. doi: 10.1249/mss.0b013e31815a51b3. [DOI] [PubMed] [Google Scholar]
- 26.Tudor-Locke CE, Myers AM. Challenges and opportunities for measuring physical activity in sedentary adults. Sports Med. 2001;31:91–9100. doi: 10.2165/00007256-200131020-00002. [DOI] [PubMed] [Google Scholar]
- 27.Zderic TW, Hamilton MT. Physical Inacivity Amplifies the Sensitivity of Skeletal Muscle to the Lipid-Induced Downregulation of Lipoprotein Lipase Activity. J Appl Physiol. 2006;100:249–257. doi: 10.1152/japplphysiol.00925.2005. [DOI] [PubMed] [Google Scholar]