Content uploaded by Glenn A Gaesser
Author content
All content in this area was uploaded by Glenn A Gaesser on Feb 04, 2024
Content may be subject to copyright.
Effects of Standing and Light-Intensity Walking
and Cycling on 24-h Glucose
NOE C. CRESPO, SARAH L. MULLANE, ZACHARY S. ZEIGLER, MATTHEW P. BUMAN, and GLENN A. GAESSER
School of Nutrition and Health Promotion, Arizona State University, Phoenix, AZ
ABSTRACT
CRESPO, N. C., S. L. MULLANE, Z. S. ZEIGLER, M. P. BUMAN, and G. A. GAESSER. Effects of Standing and Light-Intensity
Walking and Cycling on 24-h Glucose. Med. Sci. Sports Exerc., Vol. 48, No. 12, pp. 2503–2511, 2016. Purpose: This study aimed
to compare 24-h and postprandial glucose responses to incremental intervals of standing (STAND), walking (WALK), and cycling
(CYCLE) to a sit-only (SIT) condition. Methods: Nine overweight/obese (body mass index = 29 T3kgIm
j2
) adults (30 T15 yr)
participated in this randomized crossover full-factorial study, with each condition performed 1 wk apart. STAND, CYCLE, and WALK
intervals increased from 10 to 30 minIh
j1
(2.5 h total) during an 8-h workday. WALK (1.0 mph) and STAND were matched for upright
time, and WALK and CYCLE were matched for energy expenditure (~2 METs). Continuous interstitial glucose monitoring
was performed for 24 h to include the 8-h workday (LAB), after-work evening hours (EVE), and sleep (SLEEP). Three 2-h postpran-
dial periods were also ana lyz ed. Linear mixed mod els w ere used to test fo r condition differences. Results: Compared with SIT
(5.7 T1.0 mmolIL
-1
), mean 24-h glucose during STAND (5.4 T0.9 mmolIL
j1
) and WALK (5.3 T0.9 mmolIL
j1
) were lower, and
CYCLE (5.1 T1.0 mmolIL
j1
) was lower than all other conditions (all PG0.001). During LAB and EVE, mean glucose was lower for
STAND, WALK, and CYCLE compared with SIT (PG0.001). During SLEEP, the mean glucose for CYCLE was lower than all other
conditions (PG0.001). Compared with SIT, cumulative 6-h postprandial mean glucose was 5%–12% lower (PG0.001) during STAND,
WALK, and CYCLE, and 6-h postprandial glucose integrated area under the curve was 24% lower during WALK (PG0.05) and 44%
lower during CYCLE (PG0.001). Conclusions: Replacing sitting with regular intervals of standing or light-intensity activity during an
8-h workday reduces 24-h and postprandial glucose. These effects persist during evening hours, with CYCLE having the largest and most
sustained effect. Key Words: PROLONGED SITTING, PHYSICAL INACTIVITY, DIABETES, GLYCEMIA, POSTPRANDIAL
Working adults spend approximately one-half to
two-thirds of their working day sitting (27,37),
and this prolonged sitting is associated with in-
creased risk of weight gain and obesity (26,31), poor meta-
bolic health (12,19,20), and increased mortality (10,28,40).
Although these negative associations between sitting time
and adverse health outcomes are largely independent of physi-
cal activity (21,22,26), they are most evident in individuals
who are physically inactive (35,40). Acute (G7 d) experi-
mental trials have demonstrated that frequent breaks to sitting
may attenuate adverse glycemic (11,41,44) and insulinemic
(7,11,41,44) responses. Reducing prolonged periods of sitting
has therefore emerged as a new focus for reducing risk of
cardiometabolic diseases (18,38,39).
A recent review (3) highlighted the beneficial effects of
breaking up prolonged sitting with either standing or light-
intensity walking on metabolic outcomes, particularly post-
prandial glycemia. Few studies have directly compared the
acute effects of standing breaks and light-intensity activity
breaks on postprandial blood glucose. Bailey and Locke (2)
reported that interrupting sitting with 2 min of light-intensity
walking (2 mph) every 20 min for 5 h reduced postprandial
glucose concentrations but that 2-min bouts of standing every
20 min did not. Henson et al. (23) recently demonstrated that
breaking up prolonged sitting with 5-min bouts of standing or
self-paced, light-intensity walking (~1.9 mph) reduced post-
prandial glucose, with no differences between conditions.
This suggests that standing and walking may be equally ef-
fective for reducing postprandial blood glucose provided the
breaks from sitting are at least 5 min long.
Although interrupting sitting with short bouts of walking
at Q~2.0 mph (2,11,23,41) can improve glucose metabolism,
it is not known if slower walking speeds could have a similar
effect. This is relevant when using active workstations
because slower speeds (~1.0 mph) may be better for
maintaining productivity while simultaneously walking and
working (34,45). The workplace has been highlighted as an
opportune setting for health promotion (47), and desk-bound
employees are considered a key target group for sitting re-
duction strategies (44).
Address for correspondence: Glenn Gaesser, Ph.D., School of Nutrition and
Health Promotion, Arizona State University, 550 N 3rd Street; E-mail;
glenn.gaesser@asu.edu.
Submitted for publication December 2015.
Accepted for publication July 2016.
0195-9131/16/4812-2503/0
MEDICINE & SCIENCE IN SPORTS & EXERCISE
Ò
Copyright Ó2016 by the American College of Sports Medicine
DOI: 10.1249/MSS.0000000000001062
2503
APPLIED SCIENCES
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
The extent to which the improved glucose metabolism
during walking is due to postural changes alone, or to in-
creases in energy expenditure, is unclear. Cycling worksta-
tions can facilitate physical activity in office settings without
disrupting work performance or workflow and thus may
help desk-bound workers reduce risks associated with seden-
tary behavior (8,13). Recumbent cycling may offer a suitable
alternative to those who do not prefer walking or who might
have orthopedic limitations that preclude frequent and/or
prolonged periods of walking. Interrupting prolonged sitting
with 8 min of moderate-intensity cycling (52% of HR re-
serve) every hour for 8 h has been reported to reduce post-
prandial capillary C-peptide levels but had no effect on
postprandial glucose (1). We are unaware of any published
data on the effects of light-intensity cycling, equivalent in
energy expenditure to 1.0 mph walking, on glucose control.
The purpose of this study was to test the independent and
combined effects of changes in posture and energy expen-
diture on 24-h and postprandial interstitial glucose concen-
trations, assessed by continuous glucose monitoring, among
overweight/obese adults. A meta-analysis of prospective studies
indicated that elevated nonfasting blood glucose concentra-
tion is a risk factor for cardiovascular disease even in appar-
ently healthy adults without diabetes (32), and postprandial
hyperglycemia has been reported to better predict cardiovas-
cular disease than fasting blood glucose in normoglycemic
adults (9).
We hypothesized that intermittent changes in posture (i.e.,
standing), increasing energy expenditure (cycling), and the
combination of both (walking) would reduce 24-h and post-
prandial glucose concentrations when compared with uninter-
rupted sitting. It was further hypothesized that the reduction
in glucose would be greater for the increased energy expen-
diture conditions and that the combination of increased energy
expenditure and posture change (walking) would produce
the greatest reduction in 24-h and postprandial glucose.
METHODS
Screening. Overweight (body mass index (BMI),
Q25 kgIm
j2
)orclassIobese(BMI,30toG35 kgIm
j2
)men
andwomen,ages18to55yr,wererecruitedviae-mail
listserves and flyers posted throughout the university com-
munity. Because this study examined both blood pressure
(48) and interstitial glucose responses, participants had to
meet either prehypertension (systolic 9120 mm Hg or diastolic
985 mm Hg) or impaired fasting glucose criteria (5.6–
6.9 mmolIL
j1
). In this report, we present glucose results only.
Interested participants were invited to our research facil-
ities for screening on two separate occasions. Three fasting
blood glucose measurements using a ReliOn Prime Glucose
Monitor (Walmart, Bentonville, AR) and three resting blood
pressure measurements using the Oscar 2 ABP System
(SunTech Medical, Morrisville, NC) were taken 3 d apart
based on the protocol recommendations by the World Health
Organization (47). All measurements were taken within 2 wk
of the testing period at the research laboratory by trained
personnel. The same devices were used for each measurement,
and values were averaged to determine eligibility. The
intraclass correlation coefficient of our ReliOn blood glucose
data show a high degree of reliability (intraclass correlation
coefficient = 0.97, 95% confidence interval = 0.93–0.99).
In addition, participants had to be considered insufficiently
physically active (G150 minIwk
j1
of moderate-intensity
physical activity per week). Physical activity level was as-
sessed by the International Physical Activity Questionnaire
(42). A prescreening form was completed by interested par-
ticipants to exclude those who met any of the following ex-
clusion criteria: smoking, pregnancy, known coronary heart
disease, orthopedic limitations for performing physical activ-
ity, taking medications to control high blood glucose, special
dietary requirements, and being advised by a doctor to avoid
prolonged periods of sitting. This randomized crossover
full-factorial study was approved by the Arizona State Uni-
versity Institutional Review Board, and written informed con-
sent was obtained from participants before participation. The
study was registered as a clinical trial at ClinicalTrials.gov
(NCT02616809).
Protocol. Participants were required to complete the
following four conditions in random order: 1) sitting (SIT),
2) standing (STAND), 3) cycling (CYCLE), and 4) walking
(WALK) during an 8-h simulated workday. Each condition
was performed across four consecutive weeks, 7 d apart.
Each condition was designed to elicit a unique stimulus.
The SIT condition was considered the control condition (no
change in posture and no increase in energy expenditure).
The STAND condition elicited a change in posture only
(standing time), the CYCLE condition elicited an increase
in energy expenditure only (sitting plus cycling), and the
WALK condition elicited both an increase in energy ex-
penditure and change in posture (slow walking). Participants
were asked to refrain from exercise for 24 h before each
of the four testing visits and were provided with a stan-
dardized meal the evening before testing. All conditions
were performed in the same simulated office environment
in our research laboratory. All conditions required an 8-h
(0800–1600 h) intervention phase conducted in our research
facilities (LAB), a same-day evening postintervention assess-
ment phase from 1600 h until bedtime (EVE), and a sleep
phase from bedtime to wake time (SLEEP).
During SIT, participants were asked to remain seated
for the 8-h period while performing computer-related tasks
(similar to that of a typical office environment). Participants
were free to use the restroom before 0850 h, between 1000
and 1030 h, during lunch (1200–1230 h), and between 1400
and 1500 h, but no other physical activity was permitted.
Each break was recorded and replicated under each con-
dition. During STAND, participants were asked to stand
(using a height-adjustable TrekDesk Treadmill Desk, trekdesk.
com) for a predetermined time each hour as follows: 10 min
at 0850 and 0950 h, 15 min at 1045 and 1145 h, 20 min at
1240 and 1320 h, and 30 min at 1400 and 1530 h. This
http://www.acsm-msse.org2504 Official Journal of the American College of Sports Medicine
APPLIED SCIENCES
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
resulted in 2.5 h of accumulated standing time during the 8-h
workday and was based on our previous work on ambulatory
blood pressure (49), which was another outcome measure in
this study (48). Participants alternated between sitting and
standing on a nonmoving treadmill (Weslo Cadence G 5.9,
Logan, UT) that was placed underneath the TrekDesk Tread-
mill Desk. During WALK, participants worked at the Trek-
Desk Treadmill Desk while also walking at 1.0 mph. The
CYCLE condition consisted of eliciting movement while
seated through periodic bouts of slow pedaling on a Monark
cycle ergometer (894e) placed underneath the TrekDesk. The
work rate (approximately 20 W) and cadence (25–30 rpm)
were adjusted to match intensity and step rate during the
WALK condition. The frequency and duration of the walk-
ing and cycling bouts were identical with that during
STAND. The simulated office environment consisted of a
quiet observational room with a one-way mirror. This allowed
for continuous observation of participants to ensure protocol
adherence. Participants were permitted to continue computer-
based work uninterrupted during all conditions with brief
exceptions during transitions.
Participants wore an iPro2 continuous glucose monitor
(CGM) (Medtronic, Minneapolis, MN) for three consecutive
days for each condition, with CGM insertion on the day
before testing and removal on the day after testing. Inter-
stitial glucose concentration was recorded at 5-min intervals.
All CGM devices were calibrated using the standard finger
prick method and glucose meter, four times within each 24-h
period under each condition.
Four meals were provided per condition. These included
breakfast, lunch, and two dinners (one on the evening be-
fore, and one on the test day). Only one option was offered
for breakfast, and a selection of microwaveable meals with
a choice of snacks were offered for lunch and dinner. Initial
meal and snack selections were recorded for each participant
and repeated each subsequent week. The energy, macronu-
trient (in grams and as a percentage of total kilocalories), and
fiber contents for each meal were as follows: for breakfast,
479 T18 kcal, 84 T5 g carbohydrate (70% T5%), 10 T1g
fat (19% T1%), 14 T1 g protein (11% T1%), and 2 T0g
fiber; for lunch, 543 T10 kcal, 77 T3 g carbohydrate (57% T
2%), 16 T1 g fat (26% T1%), 23 T2 g protein (17% T1%),
and 8 T2 g fiber; and for dinner, 743 T16 kcal, 122 T6g
carbohydrate (66% T3%), 18 T2 g fat (22% T2%), 22 T4g
protein (12% T2%), and 12 T2 g fiber. Across the 24-h
period, macronutrient content as a percentage of the total
kcal consumption was 64% T3% carbohydrate, 22% T2%
fat, and 13% T2% protein. Breakfast and lunch were con-
sumed between 0815 and 0845 h and between 1200 and
1230 h, respectively. All meals consumed during the 8-h
workday (LAB) were brought directly to the participant
(allowing them to remain seated), and participants were ad-
vised to consume the dinner meal at the same time for each
condition. Participants were informed to avoid caffeine and
alcohol during the evening before each condition and during
the evening period of the day of each condition.
A Zephyr BioHarness
TM
(Annapolis, MD) was worn
during the LAB phase to measure HR during all conditions
(29). The Zephyr provides real-time HR monitoring, which
allowed the matching of HR between WALK and CYCLE
conditions via Bluetooth. This was removed at the end of the
day and not worn during the EVE and the SLEEP phase.
The activPAL
TM
triaxial physical activity monitor (PAL
Technologies Ltd, Glasgow, Scotland) was worn on the
right thigh during each condition for 24 h to record time
spent sitting, standing, and stepping both while inside and
outside the laboratory. This device was used to identify any
possible postural compensatory behavior in the evening of
the test day that may have occurred as a result of the con-
dition (17).
To measure possible spontaneous changes in physical
activity between conditions, each participant wore a GENEActiv
(Kimbolton, United Kingdom) accelerometer on his/her
nondominant wrist throughout the 5-wk study period (14).
Activity counts were accumulated for the 60-s epochs. Data
with at least 600 minId
j1
of wear time were included in
analyses. Nonwear time, defined as 60 min or more in which
the device did not pick up any activity, was excluded from
the analysis. Time spent in sedentary, light-, moderate-, and
vigorous-intensity physical activities was calculated using
published algorithms (15).
Data analyses. CGM data were used to calculate mean
interstitial glucose and total area under the curve (AUC) for
the entire 24-h period per condition and during each phase
(LAB, EVE, and SLEEP). Bedtime and wake time logs were
used to categorize data for EVE and SLEEP, which varied
per participant. To standardize the duration of EVE and SLEEP
phases across conditions for each participant, the average
bedtime and wake time was calculated for each participant
and used for each of the four conditions. For postprandial
analysis, 2-h postprandial incremental AUC (iAUC) was
calculated for each meal and for the cumulative 6-h post-
prandial period. The postprandial mean glucose and iAUC
were compared across conditions. AUC for 24-h glucose
and iAUC for postprandial glucose were calculated using the
trapezoidal method (46). For iAUC calculations, the three
preprandial glucose values obtained just before each meal
were averaged and used as a preprandial baseline. Glucose
values during the 2-h postprandial period that were below
the baseline were not included in the iAUC analysis (G5% of
all 2-h postprandial measurements). In addition, several in-
dices of glycemic variability were also assessed (24).
Data were assessed for normality and variables with
skewed or kurtotic distributions were transformed to achieve
normality. Data are expressed as means TSD. Linear mixed
models (LMM) were used to test for differences between
treatment conditions for all glucose measurements. Age,
gender, BMI, and baseline glucose level were entered as
covariates in the model.
LMM were also used to compare HR differences between
conditions. Postural changes (activPALi) during work
hours and after work hours and time spent in sedentary,
STANDING, LIGHT-INTENSITY ACTIVITY, AND GLYCEMIA Medicine & Science in Sports & Exercise
d
2505
APPLIED SCIENCES
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
light-, moderate-, and vigorous-intensity physical activities
(GENEActiv) were analyzed with ANOVA.
Missing data. To avoid the underestimation of AUC
and iAUC due to missing data, we excluded data sets that
were less than 80% complete. Only incomplete data were
removed. Subject data were not excluded from all other
conditions as data were considered ‘‘missing at random.’’ A
minimum of six values per condition was required for each
iAUC or AUC comparison. This varied per condition (iAUC
and AUC) and phase (24-h, LAB, EVE, and SLEEP AUC
only). The mean completeness of AUC data (measured as a
percentage of the maximum number of values possible per
condition; n= 9) per phase (24-h, LAB, EVE, and SLEEP)
was Q75% for each condition: SIT (75% T11%), STAND
(84% T11%), WALK (84% T19%), and CYCLE (80% T
6%). Within each 2-h postprandial period, a maximum of
24 readings could be collected per subject per condition,
totaling a maximum of 216 readings per meal. At the subject
level, when the 2-h postprandial data were less than 80%
complete (less than 20/24 readings) it was removed from
the condition given the effect this would have on the cal-
culation of iAUC. Data completeness (calculated as a per-
centage of the maximum number of values possible per
condition, n= 9) for each postprandial period was Q78%
for each condition: SIT (89%), STAND (78%), WALK
(89%), and CYCLE (78%). Overall, missing data across
conditions were low. Within the 24-h period, a maximum of
288 readings could be collected per subject and 2592 read-
ings per condition. Data completeness (calculated as a per-
centage of the maximum possible readings) was Q84% for
each condition: SIT (84%), STAND (90%), WALK (91%),
and CYCLE (90%).
RESULTS
Ten participants were enrolled in the study. One partici-
pant withdrew because of health complications unrelated to
the study. Two participants met the criteria for impaired
fasting blood glucose; the remaining seven met the pre-
hypertensive criteria. Therefore, nine overweight/obese (BMI =
29 T3kgIm
j2
) adults (two males, seven females) with mean T
SD age of 30 T15 yr completed all four conditions.
Data from the Bioharness, activPAL, and GENEActiv
validated the study design (Table 1). HR for SIT and STAND
was significantly lower than WALK and CYCLE (PG0.01).
Minutes spent seated, standing, and walking during EVE
were not different between conditions (P90.05). As ex-
pected during the LAB phase, participants spent signifi-
cantly more time sitting during SIT (420 T54 min) and
CYCLE (342 T77 min) compared with STAND (252 T
99 min) and WALK (269 T27 min) (PG0.05). Significantly
more time spent walking was detected during WALK (146 T
18 min, PG0.01) and significantly more time spent stand-
ing was detected during STAND (174 T36 min, PG0.01).
The GENEActiv data revealed no significant differences
in physical activity or sedentary time between conditions
across the entire 5-wk study period. Moreover, the analysis
of GENEActiv data during the 24 h preceding each of the
four conditions also revealed no differences in sedentary
time or time spent in light-, moderate-, or vigorous-intensity
physical activities (data not shown). Finally, GENEActiv
data confirmed that participants were insufficiently active,
accumulating an average of only 15–19 minId
j1
of moderate-
intensity physical activity and 0–1 minId
j1
of vigorous-
intensity physical activity (Table 2).
All recorded bedtimes were between 2200 h and midnight
(mean = 2318 h) and wake times between 0530 h and 0745 h
(mean = 0642 h). Mean within-subject differences in bed-
time (42 T20 min) and wake time (38 T25 min) were not
significantly different, and the earliest and the latest bedtime
or wake time differed by more than 60 min (75 min, for wake
time) for only one participant. For the calculation of glucose
AUC, EVE duration averaged 438 T35 min and SLEEP
duration averaged 444 T32 min.
TABLE 1. HR derived from Bioharness data, total time spent seated, standing, or walking during workday hours (LAB), after workday hours (EVE), and during the entire day until bedtime
(LAB + EVE) derived from the activPAL
TM
, and average daily time spent in different physical activity domains during each week for each condition derived from the GENEActiv.
Device Sit Stand Walk Cycle P
Bioharness LAB HR (bpm) 74 T674T680T6†83 T6†G0.001*
activPAL Seated (min) 420 T54|| 252 T99 269 T27 342 T77 0.001‡
LAB Standing (min) 33 T27 174 T36§26 T11 40 T47 G0.001‡
Walking (min) 10 T511T5146T18§68 T55 G0.001‡
Seated (min) 220 T86 243 T65 226 T41 224 T67 0.928
EVE Standing (min) 66 T21 55 T22 71 T39 43 T17 0.232
Walking (min) 29 T15 32 T23 53 T27 37 T21 0.275
Seated (min) 666 T74|| 522 T50 492 T56 560 T117 0.011‡
LAB + EVE Standing (min) 94 T39 227 T38§96 T44 70 T22 G0.001‡
Walking (min) 41 T16 46 T26 196 T19¶144 T59¶G0.001‡
GENEActiv Sedentary (min) 702 T176 702 T176 698 T164 703 T183 0.645
Light (min) 181 T91 181 T91 207 T85 185 T80 0.152
Moderate (min) 15 T21 15 T21 15 T20 19 T25 0.751
Vigorous (min) 1 T31T31T30T1 0.595
Data are presented as mean TSD (n= 9).
*PG0.05 (LMM).
†Differs from SIT and STAND only; PG0.05.
‡PG0.05 (ANOVA).
§Differs from all other conditions; PG0.05.
||Differs from STAND and WALK only; PG0.005.
¶Differs from SIT and STAND only; PG0.005.
http://www.acsm-msse.org2506 Official Journal of the American College of Sports Medicine
APPLIED SCIENCES
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Interstitial glucose. During the 24-h period, and dur-
ing LAB and EVE phases, mean glucose was 5%–12%
lower for STAND, WALK, and CYCLE compared with SIT
(Fig. 1 and Table 2; all PG0.001). The 24-h mean glucose
for WALK was also significantly lower than STAND (PG0.05),
and the 24-h mean glucose for CYCLE was significantly
lower (4%–12%) than all other conditions (PG0.001). Mean
glucose for CYCLE was also ~4% lower than STAND during
LAB (PG0.05) and ~5% lower than both STAND and
WALK during EVE (PG0.001). During SLEEP, only for
CYCLE did glucose remain significantly lower than SIT and
was also significantly lower than STAND and WALK (all
9%, PG0.001).
Cumulative 6-h postprandial mean glucose was 5%–12%
lower for STAND, WALK, and CYCLE compared with SIT
(Fig. 2 and Table 2; PG0.001). Cumulative 6-h postprandial
mean glucose was also lower for WALK compared with
STAND, and for CYCLE compared with both STAND and
WALK (all PG0.05). The differences in cumulative 6-h
postprandial mean glucose were entirely due to differences
observed after the breakfast and dinner meals, with CYCLE
having the greatest glucose-lowering effect. For the break-
fast 2-h postprandial period, mean glucose was lower for
STAND, WALK, and CYCLE compared with SIT (all PG
0.001) and for WALK and CYCLE compared with STAND
(both PG0.001). For the dinner 2-h postprandial period,
mean glucose was lower for STAND, WALK, and CYCLE
compared with SIT (all PG0.001) and for CYCLE com-
pared with STAND (PG0.001) and WALK (PG0.05).
For the cumulative 6-h postprandial iAUC, CYCLE was
44% lower compared with SIT (PG0.001), and WALK
was 24% lower compared with SIT (PG0.05). CYCLE 6-h
postprandial iAUC was also 39% lower than STAND
(Table 2; PG0.01).
No measures of glycemic variability (24) were signifi-
cantly different across conditions (P90.05).
DISCUSSION
In support of our hypothesis, intermittent changes in
posture and/or increasing energy expenditure via standing,
walking, or cycling at a light-intensity (~2 METs) through-
out an 8-h simulated workday reduced 24-h and postprandial
glucose concentrations compared with uninterrupted sitting.
Compared with SIT, 24-h mean glucose was reduced by
approximately 5%–11% after STAND, WALK, and CYCLE,
TABLE 2. Mean interstitial glucose concentration and AUC for SIT, STAND, WALK, and CYCLE during workday hours (LAB), after workday hours (EVE), and SLEEP across the 24-h period
and during the postprandial periods (breakfast, lunch, and dinner).
Phase Measure Sit Stand Walk Cycle
24 h Glucose (mmolIL
j1
) 5.7 T1.0 5.4 T0.9* 5.3 T0.9*§5.1 T1.0¶
AUC (mmolIL
j1
per 1440 min) 7954 T1186 7849 T1186 7585 T1281 7492 T1186
LAB Glucose (mmolIL
j1
) 5.6 T1.0 5.3 T0.9* 5.2 T0.9* 5.1 T0.9*§
AUC (mmolIL
j1
per 480 min) 2671 T524 2578 T571 2482 T485 2476 T571
EVE Glucose (mmolIL
j1
) 5.9 T0.9 5.5 T1.0* 5.5 T0.9* 5.2 T0.9¶
AUC (mmolIL
j1
per 438 min)# 2590 T619 2475 T619 2422 T535 2294 T574
SLEEP Glucose (mmolIL
j1
) 5.5 T1.1 5.5 T0.8 5.5 T1.0 5.0 T0.9¶
AUC (mmolIL
j1
per 444 min)# 2478 T737 2453 T577 2484 T737 2343 T621
6-h postprandial Glucose (mmolIL
j1
) 6.0 T1.1 5.7 T1.1* 5.5 T1.0*§5.3 T0.9*§||
iAUC (mmolIL
j1
per 360 min) 234 T67 216 T75 177 T37†132 T38*‡
Breakfast Glucose (mmolIL
j1
) 6.2 T0.9 6.0 T0.9* 5.4 T0.9*€5.4 T0.9*€
iAUC (mmolIL
j1
per 120 min) 85 T46 76 T36 66 T35 44 T20
Lunch Glucose (mmolIL
j1
) 5.4 T0.7 5.4 T0.7 5.2 T0.7 5.2 T0.7
iAUC (mmolIL
j1
per 120 min) 68 T23 63 T22 46 T29 42 T16
Dinner Glucose (mmolIL
j1
) 6.2 T0.8 5.7 T0.8* 5.6 T0.8* 5.3 T0.8*€||
iAUC (mmolIL
j1
per 120 min) 81 T39 80 T31 65 T19 46 T16
Data are expressed as mean TSD (n= 9); all LMM (Bonferroni test).
*Differs from SIT PG0.001.
†Differs from SIT PG0.05.
€Differs from STAND PG0.001.
‡Differs from STAND PG0.01.
§Differs from STAND PG0.05.
||Differs from WALK PG0.05.
¶Differs from SIT, STAND, and WALK PG0.001.
#Based on bedtimes and wake times, mean duration of EVE = 438 min and mean duration of SLEEP = 444 min.
FIGURE 1—Mean 24-h interstitial glucose profiles for SIT, STAND,
WALK, and CYCLE. The intervals for standing and light-intensity
physical activity during the LAB phase are illustrated by shaded areas.
Mean bedtime was 2318 h and mean wake time was 0642 h, n= 9. See
text and Table 2 for details and statistical comparisons.
STANDING, LIGHT-INTENSITY ACTIVITY, AND GLYCEMIA Medicine & Science in Sports & Exercise
d
2507
APPLIED SCIENCES
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
and the cumulative 6-h postprandial glucose iAUC was re-
duced by 24%–44% after WALK and CYCLE. Although the
clinical significance of these reductions in a healthy popula-
tion is uncertain, elevated nonfasting blood glucose is a risk
factor for cardiovascular disease even in apparently healthy
adults without diabetes (32), and postprandial hyperglycemia
has been reported to better predict cardiovascular disease than
fasting blood glucose in normoglycemic adults (9).
It was further hypothesized that the treatment effect would
be larger for the increased energy expenditure conditions.
This was supported by the finding that 24-h glucose during
WALK and CYCLE was lower compared with STAND, and
this was validated by the Bioharness data in which greater
HR was observed during WALK and CYCLE compared
with SIT and STAND. During SLEEP, only the CYCLE
condition exhibited a significantly lower glucose compared
with SIT. To our knowledge, this is the first study to show
that light-intensity activity in the form of multiple bouts
of slow cycling during the day exhibits beneficial effects on
glucose during sleep time.
We anticipated that a combination of posture change and
increasing energy expenditure via walking would produce
the most significant reduction in 24-h and postprandial glu-
cose. However, the CYCLE condition demonstrated the
most pronounced treatment effect. The greater effect for the
CYCLE condition was not expected. We carefully controlled
cycling power output and cadence to match energy expen-
diture and step rate associated with walking at 1.0 mph (6),
and Bioharness data confirmed that there was no signifi-
can t difference in HR between WALK and CYCLE. Therefore,
the larger treatment effect produced under the CYCLE
condition cannot be attributed to higher energy expenditure
than WALK. One possible explanation may be the highly
localized muscle activation of the quadriceps, which may
promote a higher uptake of glucose than the global muscle
activation experienced during walking. Although this could
possibly explain the results during the 8 h spent in the lab-
oratory, it is unlikely to explain the EVE and SLEEP phase
results because contraction-stimulated glucose uptake is
no longer observed within a couple h after exercise (33).
Thus, mechanisms related to increased insulin sensitivity
may play a greater role during the postwork evening hours
and during sleep. Quadriceps GLUT4 activity is signif-
icantly correlated with whole-body insulin sensitivity (25),
and cycling could be expected to activate quadriceps muscle
more than walking. Further research is required to determine
why the treatment effect is higher for cycling compared
with walking when energy expenditure is not significantly
different between treatment conditions. These results may
have high relevance to a work-based environment given that
pedaling in a seated position may provide the most benefit
for glucose control and may be better suited for working
while seated (8).
Several recent studies have demonstrated that interrupting
sitting with walking or standing reduces postprandial blood
glucose (2,3,11,23,41). Walking at Q2.0 mph (2,11,23) or
60% of maximum oxygen uptake (41) may not be as suitable
for working at a walking workstation compared with walk-
ing at 1.0 mph (34,45). Our study supports that walking at
1.0 mph (~2 METs) is sufficient to reduce mean and post-
prandial glucose not only during work hours but during the
evening hours after work. Our results also provide insight
into the effect of standing on glucose control. Prolonged
periods of standing, ranging from three continuous hours in
one afternoon (7) to nearly 10 h (combined with light-
intensity activity) accumulated for 24 h (43), have been
shown to reduce postprandial blood glucose excursions (7)
and prevent deterioration of insulin action (43) compared
with prolonged periods of uninterrupted sitting. Standing
breaks of 30 min every hour for 8 h have also been shown to
reduce the postprandial blood glucose response (44). Al-
though 2-min standing breaks every 20 min for 8 h did not
reduce postprandial blood glucose (2), it was recently
reported that 5-min standing breaks every 30 min for 7.5 h
did (23). In our study, the reduced glucose concentration
during STAND was already evident in the morning hours,
when the duration of each standing break was only 10–15 min
each h (Fig. 1). Thus, lesser amounts of standing than previ-
ously reported (7,43,44) may be useful for reducing 24-h
glucose concentrations and postprandial glucose responses
that are observed during uninterrupted sitting.
FIGURE 2—Postprandialmean interstitial glucose profiles forbreakfast,
lunch, and dinner during each condition (SIT, STAND, WALK, and
CYCLE), n= 9. Error bars represent 95% confidence intervals and are
displayed only at 20-min intervals to improve clarity. See text and Table 2
for details and statistical comparisons.
http://www.acsm-msse.org2508 Official Journal of the American College of Sports Medicine
APPLIED SCIENCES
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
The mechanisms for the reduced mean glucose during
STAND are largely unknown. The gastrocnemius muscle is
constantly active during standing (25), and sustained con-
tractile activity in this muscle may facilitate muscle glucose
uptake via recruitment of GLUT4 transporters (16,36).
Our study has several strengths. This is the first study
to compare the effects of standing and both light-intensity
walking and cycling with uninterrupted sitting on 24-h and
postprandial glucose. The activPAL and GENEActiv data
confirmed planned differences in posture allocation among
the four conditions. Furthermore, activPAL and GENEActiv
data indicated that any differences between conditions dur-
ing the measurement periods outside the LAB phase were
not due to compensatory behavior during the evening hours.
The total number of lower-limb movements during the WALK
and CYCLE active periods was controlled by matching cycling
cadence to the step rate of walking at 1.0 mph. The use of
continuous glucose monitoring allowed us to examine in-
terstitial glucose during the 24-h period. Consequently, our
study also provides new information to show that the effects
of standing and very light-intensity physical activity on glu-
cose responses extend for several hours beyond the period
during which the activities are performed.
The primary weaknesses of our study are the small sample
size and the fact that only two of the nine participants had
impaired fasting blood glucose. However, apparently healthy
adults without diabetes may benefit from lower postprandial
(9) and 24-h glucose concentrations (32). Our significant
findings on 24-h and postprandial glucose on a small sample
of adults with relatively normal glycemia highlight the po-
tential effectiveness of our intervention. The small sample
size reduced our ability to detect statistically significant re-
ductions in 24-h glucose AUC that would be expected on the
basis of the results for the mean glucose data. Similarly,
because of the reduction of multiple data points to one value
per AUC and iAUC calculation, incomplete data sets were
excluded to avoid the underestimation of AUC and iAUC
results. This was required to protect the integrity of the re-
sults and further reduced the sample size. A larger sample
size may reduce the effects of missing data and may enable
the detection of significant AUC and iAUC differences be-
tween conditions. By contrast, LMM analyses of glucose
data included a greater number of measurements (i.e., every
5 min) per participant for each condition.
We did not control for menstrual cycle phase, and five
of our seven female participants were between the ages of
18 and 44 yr. Insulin sensitivity has been shown to change
during the menstrual cycle (14), but it has also been reported
that menstrual cycle phase had no effect on glucose toler-
ance or insulin secretion (5). Because several other studies
similar to ours have not reported whether they controlled
for menstrual cycle phase (1,2,7,11,41), this is a potentially
important methodological consideration that needs to be
incorporated into the design of future studies (44).
Although the timing and the duration of the breakfast and
lunch meals were strictly controlled, the consumption of the
dinner meal occurred outside the laboratory. Thus, we were
unable to verify subject compliance with instructions for the
consumption of the dinner meal. The inspection of CGM
data indicated that for a given participant, the dinner meal
was consumed at approximately the same time of evening
under all conditions.
We acknowledge that there is limited CGM accuracy in
comparison with capillary blood glucose measurement (4).
However, given the rigor of the experimental design, com-
pleting multiple blood draws would have been too burden-
some. The standardization of meals was also generalized for
all participants and was not adjusted to accommodate dif-
ferences in resting metabolic rates. However, any potential
glycemic effects would have been present under all condi-
tions, and such practices are regularly conducted with OGTT
methods in which all subjects are provided with the same
glucose solution concentration. This also would not con-
found any differences between the WALK and the CYCLE
conditions for which energy expenditure was matched. The
standardization of meals was enforced the evening before
each test day, but dietary intake during the entire 24- to 48-h
period before each test day was not controlled. Future studies
may benefit from the inclusion of subjective measures of ap-
petite and may control dietary intake 24–48 h before each
test day to investigate potential confounding effects.
Our experimental design of accumulating 2.5 h of non-
sitting time, with progressively longer durations throughout
the workday, may prove to be more than needed to lower 24-h
glucose. We used this design on the basis of our previous
work on ambulatory blood pressure (49), which was another
outcome measure in this study (48). Figure 1 shows that
during the LAB phase, the greatest reductions in glucose
during STAND, WALK, and CYCLE occurred during the
first half of the LAB phase, when the durations of nonsitting
time were the shortest. It has been documented that accu-
mulating 2-min bouts of light-intensity (2.0 mph) and
moderate-intensity (3.6–4.0 mph) walking every 20 min for
5 h has been shown to alter the expression of nume-
rous skeletal muscle genes, including those associated with
glucose metabolism (30). It would be useful to determine
whether posture change or very light-intensity activity
(~2 METs) alters skeletal muscle gene expression.
Investigating the effects of posture change or very light-
intensity activity for an extended period would also provide
useful data regarding the true effects of sitting as an occupa-
tional hazard. Whether there is a dose–response relationship
and an optimal duration of posture change and/or light-
intensity activity that is both effective and suitable for the
workplace environment is yet to be determined.
The authors thank the School of Nutrition and Health Promotion
for funding support of this study.
The authors have no conflicts of interest to declare.
The results of the present study do not constitute endorsement by
the American College of Sports Medicine.
STANDING, LIGHT-INTENSITY ACTIVITY, AND GLYCEMIA Medicine & Science in Sports & Exercise
d
2509
APPLIED SCIENCES
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
REFERENCES
1. Altenburg TM, Rotteveel J, Dunstan DW, Salmon J, Chinapaw
MJ. The effect of interrupting prolonged sitting time with short,
hourly, moderate-intensity cycling bouts on cardiometabolic risk
factors in healthy, young adults. J Appl Physiol (1985). 2013;
115(12):1751–6.
2. 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. 2015;18(3):
294–8.
3. Benatti FB, Ried-Larsen M. The effects of breaking up prolonged
sitting time: a review of experimental studies. Med Sci Sports
Exerc. 2015;47(10):2053–61.
4. Blevins TC, Bode BW, Garg SK, et al. Statement by the American
Association of Clinical Endocrinologists Consensus Panel on
continuous glucose monitoring. Endocr Pract. 2010;16(5):730–45.
5. Bonora E, Zavaroni I, Alpi O, et al. Influence of the menstrual
cycle on glucose tolerance and insulin secretion. Am J Obstet
Gynecol. 1987;157(1):140–1.
6. Browning RC, Modica JR, Kram R, Goswami A. The effects of
adding mass to the legs on the energetics and biomechanics of
walking. Med Sci Sports Exerc. 2007;39(3):515–25.
7. Buckley JP, Mellor DD, Morris M, Joseph F. Standing-based of-
fice work shows encouraging signs of attenuating post-prandial
glycaemic excursion. Occup Environ Med. 2014;71(2):109–11.
8. Carr LJ, Maeda H, Luther B, Rider P, Tucker SJ, Leonhard C.
Acceptability and effects of a seated active workstation during
sedentary work: a proof of concept study. Int J Workplace Health
Manag. 2014;7(1):2–15.
9. DECODE Study Group, the European Diabetes Epidemiology
Group. Glucose tolerance and cardiovascular mortality: compari-
son of fasting and 2-hour diagnostic criteria. Arch Intern Med.
2001;161(3):397–405.
10. Dunstan DW, Barr EL, Healy GN, et al. Television viewing time
and mortality: the Australian Diabetes, Obesity and Lifestyle Study
(AusDiab). Circulation. 2010;121(3):384–91.
11. Dunstan DW, Kingwell BA, Larsen R, et al. Breaking up pro-
longed sitting reduces postprandial glucose and insulin responses.
Diabetes Care. 2012;35(5):976–83.
12. Dunstan DW, Salmon J, Healy GN, et al. Association of television
viewing with fasting and 2-h postchallenge plasma glucose levels
in adults without diagnosed diabetes. Diabetes Care. 2007;30(3):
516–22.
13. Elmer SJ, Martin JC. A cycling workstation to facilitate physical
activity in office settings. Appl Ergon. 2014;45(4):1240–6.
14. Escalante Pulido JM, Alpizar Salazar M. Changes in insulin sen-
sitivity, secretion and glucose effectiveness during menstrual cy-
cle. Arch Med Res. 1999;30(1):19–22.
15. Esliger DW, Rowlands AV, Hurst TL, Catt M, Murray P, Eston
RG. Validation of the GENEA Accelerometer. Med Sci Sports
Exerc. 2011;43(6):1085–93.
16. Galante P, Mosthaf L, Kellerer M, et al. Acute hyperglycemia
provides an insulin-independent inducer for GLUT4 translocation
in C
2
C
12
myotubes and rat skeletal muscle. Diabetes. 1995;44(6):
646–51.
17. Grant PM, Ryan CG, Tigbe WW, Granat MH. The validation of a
novel activity monitor in the measurement of posture and motion
during everyday activities. Br J Sports Med. 2006;40(12):992–7.
18. Hamilton MT, Healy GN, Dunstan DW, Zderic TW, Owen N. Too
little exercise and too much sitting: inactivity physiology and
the need for new recommendations on sedentary behavior. Curr
Cardiovasc Risk Rep. 2008;2(4):292–8.
19. 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.
20. Healy GN, Matthews CE, Dunstan DW, Winkler EAH, Owen N.
Sedentary time and cardio-metabolic biomarkers in US adults:
NHANES 2003–06. Eur Heart J. 2011;32(5):590–7.
21. Healy GN, Wijndaele K, Dunstan DW, et al. Objectively measured
sedentary time, physical activity, and metabolic risk: the Australian
Diabetes, Obesity and Lifestyle Study (AusDiab). Diabetes Care.
2008;31(2):369–71.
22. Helmerhorst HJF, Wijndaele K, Brage S, Wareham NJ, Ekelund
U. Objectively measured sedentary time may predict insulin re-
sistance independent of moderate- and vigorous-intensity physical
activity. Diabetes. 2009;58(8):1776–9.
23. Henson J, Davies MJ, Bodicoat DH, et al. Breaking up prolonged
sitting with standing or walking attenuates the postprandial meta-
bolic response in postmenopausal women: a randomized acute
study. Diabetes Care. 2016;39(1):130–8.
24. Hill NR, Oliver NS, Choudhary P, Levy JC, Hindmarsh P, Matthews
DR. Normal reference range for mean tissue glucose and glycemic
variability derived from continuous monitoring for subjects with-
out diabetes in different ethnic groups. Diabetes Technol Ther.
2011;13(9):921–8.
25. Houmard JA, Weidner MD, Dolan PL, et al. Skeletal muscle
GLUT4 protein concentration and aging in humans. Diabetes.
1995;44(5):555–60.
26. Hu FB, Li TY, Colditz GA, Willett WC, Manson JE. Television
watching and other sedentary behaviors in relation to risk of obe-
sity and type 2 diabetes mellitus in women. JAMA. 2003;289(14):
1785–91.
27. Jans MP, Proper KI, Hildebrandt VH. Sedentary behavior in Dutch
workers: differences between occupations and business sectors.
Am J Prev Med. 2007;33(6):450–4.
28. Katzmarzyk PT, Church TS, Craig CL, Bouchard C. Sitting time
and mortality from all causes, cardiovascular disease, and cancer.
Med Sci Sports Exerc. 2009;41(5):998–1005.
29. Kokonozi A, Astaras A, Semertzidis P, et al. Development and
clinical evaluation of a physiological data acquisition device for
monitoring and exercise guidance of heart failure and chronic heart
disease patients. Computing in Cardiology. 2010;37:1099–1102.
30. Latouche C, Jowett JB, Carey AL, et al. Effects of breaking up
prolonged sitting on skeletal muscle gene expression. J Appl
Physiol (1985). 2013;114(4):453–60.
31. Levine JA, Lanningham-Foster LM, McCrady SK, et al. Inter-
individual variation in posture allocation: possible role in human
obesity. Science. 2005;307(5709):584–6.
32. Levitan EB, Song Y, Ford ES, Liu S. Is nondiabetic hyperglycemia
a risk factor for cardiovascular disease? A meta-analysis of pro-
spective studies. Arch Intern Med. 2004;164(19):2147–55.
33. Maarbjerg SJ, Sylow L, Richter EA. Current understanding of in-
creased insulin sensitivity after exercise—emerging candidates.
Acta Physiol (Oxf). 2011;202(3):323–35.
34. MacEwen BT, MacDonald DJ, Burr JF. A systematic review of standing
and treadmill desks in the workplace. Prev Med. 2015;70:50–8.
35. Matthews CE, Moore SC, Sampson J, et al. Mortality benefits for
replacing sitting time with different physical activities. Med Sci
Sports Exerc. 2015;47(9):1833–40.
36. Megeney LA, Neufer PD, Dohm GL, et al. Effects of muscle ac-
tivity and fiber composition on glucose transport and GLUT-4. Am
J Physiol. 1993;264(4 Pt 1):E583–93.
37. Miller R, Brown W. Steps and sitting in a working population. Int
J Behav Med. 2004;11(4):219–24.
38. Owen N, Bauman A, Brown W. Too much sitting: a novel and
important predictor of chronic disease risk? Br J Sports Med.
2009;43(2):81–3.
39. Pate RR, O_Neill JR, Lobelo F. The evolving definition of ‘‘sed-
entary.’’ Exerc Sport Sci Rev. 2008;36(4):173–8.
http://www.acsm-msse.org2510 Official Journal of the American College of Sports Medicine
APPLIED SCIENCES
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
40. Patel AV, Bernstein L, Deka A, et al. Leisure time spent sitting
in relation to total mortality in a prospective cohort of US adults.
Am J Epidemiol. 2010;172(4):419–29.
41. Peddie MC, Bone JL, Rehrer NJ, Skeaff CM, Gray AR, Perry TL.
Breaking prolonged sitting reduces postprandial glycemia in
healthy, normal-weight adults: a randomized crossover trial. Am J
Clin Nutr. 2013;98(2):358–66.
42. Rosenberg DE, Bull FC, Marshall AL, Sallis JF, Bauman AE.
Assessment of sedentary behavior with the International Physical
Activity Questionnaire. J Phys Act Health. 2008;5(1 Suppl):S30–44.
43. 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(7):941–9.
44. Thorp AA, Healy GN, Winkler E, et al. Prolonged sedentary time
and physical activity in workplace and non-work contexts: a cross-
sectional study of office, customer service and call centre em-
ployees. Int J Behav Nutr Phys Act. 2012;9:128.
45. Tudor-Locke C, Schuna JM Jr, Frensham LJ, Proenca M. Chang-
ing the way we work: elevating energy expenditure with work-
station alternatives. Int J Obes (Lond). 2014;38(6):755–65.
46. Wolever TM. Effect of blood sampling schedule and method of
calculating the area under the curve on validity and precision of
glycaemic index values. Br J Nutr. 2004;91(2):295–301.
47. World Health Organization. Preventing noncommunicable dis-
eases in the workplace through diet and physical activity. WHO/
World Economic Forum Report of a Joint Event. 2008 [cited June
2016]. Available from: http://www.weforum.org/pdf/Wellness/
WHOWEF_report.pdf.
48. Zeigler ZS, Mullane SL, Crespo NC, Buman MP, Gaesser GA.
Effects of standing and light-intensity activity on ambulatory blood
pressure. Med Sci Sports Exerc. 2016;48(2):175–81.
49. Zeigler ZS, Swan PD, Bhammar DM, Gaesser GA. Walking
workstation use reduces ambulatory blood pressure in adults with
prehypertension. J Phys Act Health. 2015;12(6 Suppl 1):S119–27.
STANDING, LIGHT-INTENSITY ACTIVITY, AND GLYCEMIA Medicine & Science in Sports & Exercise
d
2511
APPLIED SCIENCES
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
