848735

research-article2019
DSTXXX10.1177/1932296819848735Journal of Diabetes Science and TechnologyLee et al

Symposium/Special Issue

Journal of Diabetes Science and Technology

Moderate-to-Vigorous-Intensity
2019, Vol. 13(5) 827­–835
© 2019 Diabetes Technology Society
Article reuse guidelines:
Physical Activity Observed in People sagepub.com/journals-permissions
DOI: 10.1177/1932296819848735
https://doi.org/10.1177/1932296819848735

With Diabetes-Related Foot Ulcers journals.sagepub.com/home/dst

Over a One-Week Period

Maggie Lee, BSc1,2, Jaap J. van Netten, PhD1,3,

Helen Sheahan, BAppSc4,5, and Peter A. Lazzarini, PhD1,4,6

Abstract
Background: Regular moderate-to-vigorous-intensity physical activity results in health benefits in people with diabetes. No
study has observed the moderate-to-vigorous-intensity physical activity typically performed by people with diabetes-related
foot ulcers (DFU) in their everyday free-living environments. We observed the bouts, and accumulated time, spent doing
moderate-to-vigorous-intensity physical activity in cases with DFU compared with diabetes-related peripheral neuropathy
(DPN) and diabetes (DM) controls over a one-week period.

Methods: This was a secondary analysis of a cross-sectional case-control study. Participants wore a multisensor device
for >5 days (>22 hours per day). Primary outcomes included the number, duration (minutes) and intensity (metabolic
equivalent tasks [METs]) of bouts of moderate-to-vigorous-intensity physical activity (defined as at least >3 METs for >10
consecutive minutes). Secondary outcomes included the total accumulated times spent doing moderate-to-vigorous-intensity
physical activity (>3 METs) and doing sedentary-intensity activity (<1.5 METs). DFU subgroups with minor amputations and
nonremovable offloading devices were also analyzed.

Results: Overall, 15 DM, 23 DPN, and 27 DFU participants were included. All groups recorded similar low median daily
numbers (0.33, 0.29, 0.25 numbers, respectively), duration (15, 17, 14 minutes), and intensity of daily bouts of moderate-to-
vigorous-intensity physical activity (4.1, 4.3, 3.9 METs) (all, P > .1). Median accumulated daily time spent doing moderate-to-
vigorous-intensity physical activity was also similar (40, 37, 36 minutes; P > .8). Those with DFU had more mean accumulated
daily time spent doing sedentary-intensity activity (796 minutes) compared to DPN (720 minutes; P < .05), but not compared
to DM (728 minutes; P < .08). DFU subgroups with minor amputations had more median accumulated daily time spent
doing moderate-to-vigorous-intensity physical activity (66, 28 minutes; P < .05) and less mean time doing sedentary-intensity
activity (745, 837; P < .05) than those without amputations.

Conclusions: People with DFU performed similar low numbers of daily bouts of moderate-to-vigorous-intensity physical
activity to controls, but spend more time doing sedentary-intensity activities. Interventions that gradually increase the
moderate-to-vigorous-intensity physical activity in people with DFU should be investigated.

Keywords
bout, diabetes-related foot ulcers, diabetes-related peripheral neuropathy, physical activity, sedentary activity

Diabetes-related complications are rapidly becoming one of Performing regular moderate-to-vigorous-intensity physi-
the leading causes of global morbidity, mortality and health cal activity has been reported to prevent or delay the onset of
care cost burdens.1-3 Diabetes-related foot ulcers (DFU) are diabetes and other diabetes-related complications.8,9 Specific
now very much a leading cause of these large diabetes- health benefits gained from performing regular moderate-to-
related burdens.4-6 Foot ulcers typically develop in people vigorous-intensity physical activity in people with diabetes
with the diabetes-related peripheral neuropathy (DPN).4,7 includes improved insulin sensitivity,10 improved endothelial
Thus, to reduce diabetes-related burdens more needs to be function,11 lowered glycated HbA1c,12-14 reduced body mass
done to understand, prevent, delay and manage these diabe- index,12 reduced systolic blood pressure,15 and reduced risk
tes-related foot complications. of complications.8,9,15
828 Journal of Diabetes Science and Technology 13(5)

To achieve these health benefits international guidelines environment over a one-week period.23 The original study
recommend adults need to perform an accumulated 30 min- recruited participants from three diabetes-related foot outpa-
utes of exercise on at least five days of each week.8,16 They tient clinics in Brisbane, Australia.23 Findings from the original
define exercise as being a bout of at least 10 consecutive min- study on the outcomes of average daily steps, energy expendi-
utes of any moderate-to-vigorous-intensity physical activity ture (kJ), lying and sleeping durations have previously been
(defined as >3 metabolic equivalent tasks [METs]). One published.23 Ethical approval was granted for this study by the
MET is defined as approximately 1 kcal/kg/hour and includes Human Research Ethics Committees (HREC) of the Prince
activities such as sleeping, lying quietly or meditating.17 Charles Hospital (No. HREC/17/QPCH/13) and Queensland
Activities achieving levels of moderate-intensity physical University of Technology (No. 1700000228), Australia.
activity (3-6 METs) typically include brisk walking, swim- Written informed consent was voluntarily obtained from all
ming leisurely, or cycling leisurely,8,16,17 whereas activities individual participants. The design and methodology have been
achieving vigorous-intensity physical activity (>6 METs) described in detail elsewhere,23 and will be summarized below.
include jogging, cycling fast, or playing a game of football.17
To help achieve these moderate-to-vigorous-intensity levels
Participants
of physical activity guidelines also recommend that people
with diabetes should decrease their sedentary activity. All participants were adults with type 2 diabetes mellitus.
Sedentary activities (<1.5 METs) typically includes lying, Participants were grouped into those with: (1) DPN and a
sitting, eating or standing quietly.17 Therefore, people with current DFU (DFU group), (2) DPN and no DFU history
diabetes should be aiming to perform at least one bout of (DPN group), and (3) no DPN or DFU history (DM
moderate-to-vigorous-intensity physical activity on most group).23 The DFU group was further subdivided into
days, and reduce their daily sedentary activity, if they are to those (1) with a history of minor amputation(s) compared
achieve optimal health benefits.8,16 with those without any amputation history and (2) those
Many studies have now begun to investigate if people wearing a nonremovable knee-high offloading device
with type 2 diabetes are achieving these physical activity compared with those wearing another offloading device.23
recommendations.18-22 However, to our knowledge none DPN was defined as an inability to detect a 10-gram
have investigated this in people with DFU. Therefore, the monofilament on at least two of three sites on the plantar
primary aim of this study was to investigate the bouts of forefoot.7,24 DFU was defined as an active full thickness
moderate-to-vigorous-intensity physical activity performed wound on the plantar aspect of the foot.7,24 Minor amputa-
by people with DFU and to compare it to those with DPN tion was defined as a previous amputation below the
without DFU history (DPN) and those with diabetes without ankle.7,24,25 Nonremovable knee-high offloading devices
DPN or DFU (DM) in their everyday free-living environ- were defined as either a total contact cast or removable
ments. Secondary aims were to investigate the accumulated knee-high walker made irremovable.25,26 Other offloading
daily time spent doing any moderate-to-vigorous-intensity devices were defined as either removable ankle-high
physical activity and any sedentary-intensity activity. Last, devices (such as a postoperative shoe), therapeutic foot-
we also investigated these outcomes in DFU subgroups with wear, or felted foam applied in appropriate footwear.25,26
and without minor amputations, and with and without non- The exclusion criteria for the original study were designed
removable offloading devices. to avoid conditions that may confound the impact of DPN
and DFU on typical daily activity and sleep patterns and
included those with (1) peripheral arterial disease (defined as
Methods a toe pressure of <70 mmHg), (2) major amputation (previ-
ous amputation above the ankle), (3) mobility impairment
Study Design
(inability to walk without an aide), (4) a DFU on the contra-
This study was a secondary analysis of data collected in an lateral foot, (5) painful DPN, (6) sleep apnea, or (7) those
original cross-sectional case-control study that investigated the deemed to be cognitively unable or unwilling to perform the
activity of people with DFU, DPN, or DM in their free-living study procedures by the researchers.23,24

1
School of Clinical Sciences, Queensland University of Technology, Brisbane, Australia
2
Podiatry Department, Ng Teng Fong General Hospital, Singapore
3
Amsterdam UMC, Department of Rehabilitation, University of Amsterdam, Amsterdam Movement Sciences, Amsterdam, the Netherlands
4
Allied Health Research Collaborative, The Prince Charles Hospital, Brisbane, Australia
5
Department of Diabetes and Endocrinology, Sunshine Coast University Hospital, Birtinya, Australia
6
School of Public Health and Social Work, Queensland University of Technology, Brisbane, Australia

Corresponding Author:
Peter A. Lazzarini, PhD, School of Public Health and Social Work, Queensland University of Technology, Victoria Park Rd, Kelvin Grove, Brisbane,
Queensland 4059, Australia.
Email: [email protected]
Lee et al 829

Sample Size Outcome Measurement

Sample size was calculated for the original study based on The primary outcomes of interest for this particular study
previously reported mean daily steps in people with DM,27,28 were the number of bouts of moderate-to-vigorous-intensity
DPN,27,29,30 and DFU,27,29 a power of 80% and an alpha of physical activity (defined as activity recording an intensity of
.05.23 These calculations suggested 25 participants were >3 METs per minute for >10 consecutive minutes),8,16 and
required in each group to determine a statistical difference the average duration (minutes per bout) and intensity (METs
for daily steps which is considered a surrogate marker of per bout) of these bouts. Secondary outcomes included the
daily activity.23,24 overall accumulated time spent doing any moderate-to-vigor-
ous-intensity physical activity (total of any nonsleeping min-
utes recording intensity >3 METs per minute) and the overall
Participant Characteristics accumulated time spent doing any sedentary-intensity activ-
Participant’s characteristics were recorded using the ity time (total of any nonsleeping minutes recording intensity
Queensland High Risk Foot Form (QHRFF)23,25 and included <1.5 METs per minute) and intensity (METs per minutes).8,16
demographics (age, gender, body mass index [BMI]), diabe- The data from the five or more eligible days were averaged to
tes history (diabetes duration, HbA1c), comorbidity history determine daily outcomes.
(hypertension, dyslipidemia, cardiovascular disease, chronic
kidney disease), foot complications (foot deformity, DPN,
Statistical Analysis
DFU, and amputation history), and treatment (offloading
devices).23,25 Data were analyzed using SPSS 23.0 for Windows (IBM
Corp, Armonk, NY, USA). For categorical variables, pro-
portions were reported and Person’s chi-squared or
Procedure Fisher’s exact test used to test differences between three
A SenseWear armband (Model MF-SW; BodyMedia Inc, groups and two groups respectively. For continuous vari-
Pittsburgh, PA, USA) was used to measure physical activ- ables, Kolmogorov-Smirnov tests were used to assess nor-
ity in all participants.23 All participants were instructed to mality of distribution. For those with normal distributions,
wear the armband on their mid upper arm continuously for means and standard deviations (±SD) were reported and
one week at all times except when engaging in water-based ANOVA with Fisher’s least significant difference were
activities such as showering or swimming.23 No further used to test differences between three groups and Student’s
advice was provided as the study aim was to observe typi- t-tests between two groups. For those nonnormal distribu-
cal daily activity of participants in their normal free-living tions, median, and interquartile ranges (IQR) were
everyday environment.23 At least five days of data from reported and Kruskal-Wallis and Mann-Whitney U-tests
wearing the device for at least 22 hours on each of those were used to test differences between three and two groups
days, with at least one of those days being a weekend day, respectively. A significance of P < .05 was used through-
was required to be eligible for the participant’s data to be out. In addition, for those outcomes with P < .2 and nor-
included, as this period has been previously validated to be mal distributions we estimated effect sizes (standardized
representative of typical daily activity.31-34 Days in which a mean differences) between pairs of groups using Cohen’s
participant wore the device for less than 22 hours were d.40 The degree of the effect size difference was graded
excluded. based on Cohen’s d values as <0.1 trivial difference, 0.1-
The armband consists of multiple sensors such as tri- 0.2 small difference, 0.2-0.6 medium difference, 0.6-1.2
axial accelerometer (measures motion), skin temperature large difference, and >1.2 very large difference.41
(measures armband use), galvanic skin response (measures
skin impedance), and heat flux (measures the rate of heat
Results
dissipation).23,35,36 The armband uses validated proprietary
algorithms to transform data captured from the multiple Seventy-seven participants were recruited in the original
sensors along with demographic characteristics to detect cross-sectional study, including 23 DM, 24 DPN, and 30
and record wear time (compliance with wearing the device) DFU. Five (6.5%) participants were excluded from the origi-
and calculate activity parameters such as METs (mean error nal study (four DM and one DPN) as their armband failed to
<5%), steps (mean error <9%), physical activity intensity collect any data. A further seven (9.1%) participants also had
and duration (mean error <5%), lying and sleep duration to be excluded from this analysis (four DM and three DFU)
(mean error <10%), and estimated energy expenditure as the armband failed to collect all continuous minute-by-
(mean error <5%).23,35-37 The use of the armband has been minute physical activity data over the five days (>22 hours
validated in comparisons with gold standard criterion mea- per day) to determine the primary outcome of bouts.
sures for a range of daily physical activity parameters, Therefore, 65 participants were included in this secondary
including METs.31,32,34-36,38,39 analysis, including 15 DM, 23 DPN, and 27 DFU.
830 Journal of Diabetes Science and Technology 13(5)

Table 1. Participant Baseline Characteristics and Outcomes of Interest for Each Group Stated in Mean ± SD, Median [IQR], or
Number (%).

Total DM DPN DFU P value

Participant characteristics
Numbers 65 15 23 27 —
Age 61 ± 11 59 ± 11 68 ± 9** 56 ± 11 .001*
Men 42 (64.6%) 7 (46.7%) 14 (60.9%) 21 (77.8%) .116
BMI 32.6 [29.0-35.7] 32.6 [30.4-33.7] 33.2 [29.8-35.5] 32.6 [28.6-36.8] 1.000
Diabetes duration (years) 13 ± 9 10 ± 7 15 ± 9 13 ± 10 .275
HbA1c (%) 6.9 [6.3-8.4] 7.9 [6.4-9.2] 6.6 [6.2-7.6] 7.0 [6.9-1.0] .259
Hypertension 37 (56.9%) 9 (60.0%) 14 (60.9%) 14 (51.9%) .784
CVD 14 (21.5%) 3 (20.0%) 5 (21.7%) 6 (22.2%) 1.000
CKD 6 (9.2%) 1 (6.7%) 1 (4.3%) 4 (14.8%) .554
Dyslipidemia 30 (46.2%) 8 (53.3%) 12 (52.2%) 10 (37.0%) .461
Foot deformity 24 (36.9%) 0 4 (16.7%) 20 (83.3%)** <.001*
DPN 50 (76.9%) 0 23 (100%) 27 (100%) NA
Previous DFU 25 (38.5%) 0 0 25 (92.6%) NA
Previous amputation 12 (18.5%) 0 0 12 (44.4%) NA
Outcomes of interest
Daily number of bouts of MVPAa (number per day) 0.33 [0-0.90] 0.33 [0-0.67] 0.29 [0-1.33] 0.25 [0-0.80] .981
Duration per bout of MVPAb (minutes per bout) 15 [12-18] 15 [12-17] 17 [14-21] 14 [11-18] .142
Intensity per bout of MVPAc (METs per bout) 4.1 [3.8-4.4] 4.1 [4.0-4.4] 4.3 [3.8-4.5] 3.9 [3.6-4.2] .343
Accumulated daily time spent doing MVPAd 37 [20-65] 40 [21-65] 37 [19-76] 36 [17-64] .893
(minutes per day)
Accumulated daily time spent doing sedentary 754 ± 131 728 ± 123^ 720 ± 140*** 796 ± 120***^ .086
activitye (minutes per day)
Intensity per sedentary activityf (METs per 0.92 ± 0.09 0.89 ± 0.09 0.90 ± 0.07 0.95 ± 0.10** .041*
sedentary activity)

*Three groups are significantly different (P < .05). **Group that is significantly different from the other two groups (P < .05). ***Two groups that are
significantly different from each other (P < .05); ^groups that show a trend to being different from each other using post hoc tests (P < .08).
a
>3 METs for >10 consecutive minutes. bConsecutive time until eligible bout ends. cAverage METs per eligible bout. dAny time recording >3 METs per
minute. eAny time recording <1.5 METs per minute. fAverage METs per sedentary-intensity activity.

Table 1 displays participant characteristics and outcomes sedentary activity time was significantly higher for the DFU
of interest for the three groups. Most characteristics were group compared to the DPN group (DFU 0.95 ± 0.10, DPN
similar between groups (P > .05), except for those with DPN 0.90 ± 0.07 METs per minute; P < .05; d = 0.58) and com-
being older (DPN 68 ± 9, DM 59 ± 11 and DFU 56 ± 11 pared to the DM group (DFU 0.95 ± 0.10, DM 0.89 ± 0.09
years; P = .001) and those with DFU having more foot METs per minute, P < .05; d = 0.63).
deformities (DM 0%, DPN 16.7%, DFU 83.3%; P < .001). Table 2 displays the participant characteristics and out-
All groups recorded similar low median [IQR] numbers of comes of interest for the subgroups of those 27 with DFU,
daily bouts of moderate-to-vigorous-intensity physical activ- including 12 (44%) with minor amputation(s) and 15 (56%)
ity (DM 0.33 [0-0.67], DPN 0.29 [0-1.33], DFU 0.25 [0-0.8] without a minor amputation, and 7 (26%) wearing a nonre-
numbers), median durations per these bouts (DM 15 [12-17], movable offloading device and 20 (74%) wearing other
DPN 17 [14-21], DFU 14 [11-18] minutes) and median offloading devices. Participants with minor amputation(s)
intensity per bout (DM 4.1 [4.0-4.4], DPN 4.3 [3.8-4.5], compared to those without any amputation had more, albeit
DFU 3.9 [3.6-4.2] METs) (all P > .1). All groups also nonsignificant, median daily bouts of moderate-to-vigorous-
recorded similar median accumulated daily time spent doing intensity physical activity (0.58 [0.17-1.89], 0.20 [0-0.63]; P
moderate-to-vigorous-intensity physical activity (DM 40 < .08), significantly higher median intensity during their
[21-65], DPN 37 [19-76], DFU 36 [17-64] minutes per day; bouts (4.2 [3.9-4.3], 3.7 [3.5-4.0] METs; P < .05), signifi-
P > .8). The DFU group recorded significantly more mean cantly more median accumulated daily time spent doing
accumulated daily time spent doing sedentary-intensity moderate-to-vigorous-intensity physical activity time (66
activity than the DPN group (DFU 796 ± 120, DPN 720 ± [24-109], 28 [12-42] minutes per day; P < .05), significantly
140 minutes per day; P < .05; d = 0.58), but not signifi- less mean accumulated daily time spent doing sedentary-
cantly more than the DM group (DFU 796 ± 120, DM 728 intensity activity (745 ± 125, 837 ± 103 minutes per day; P
± 123; P < .08; d = 0.56). The mean intensity during this < .05; d = 0.80), and significantly more mean intensity
Lee et al 831

Table 2. Participant Baseline Characteristics and Outcomes of Interest for Each DFU Subgroup Stated in Mean ± SD, Median [IQR], or
Number (%).

No Minor Nonremovable Other offloading

amputation amputation P value offloading device device P value
Participant characteristics
Numbers 15 12 — 7 20 —
Age 56 ± 13 57 ± 9 .840 57 [57-59] 56 [49-62] .739
Men 10 (66.7%) 11 (91.7%) .182 5 (71.4%) 16 (80.0%) .633
BMI 34.8 ± 7.5 31.3 ± 6.0 .204 29.0 [26.9-32.6] 33.3 [29.6-39.4] .184
Diabetes duration (years) 9±5 19 ± 12 .017* 18 ± 15 12 ± 8 .219
HbA1c (%) 7.5 [6.0-10.2] 6.5 [6.2-7.7] .349 7.0 [6.7-11.0] 6.5 [6.3-10.0] .230
Hypertension 8 (53.3%) 6 (50.0%) .863 2 (28.6%) 12 (60.0%) .209
CVD 4 (26.7%) 2 (16.7%) .662 1 (14.3%) 5 (25.0%) 1.000
CKD 2 (13.3%) 2 (16.7%) 1.000 2 (28.6%) 2 (10.0%) .269
Dyslipidemia 5 (33.3%) 5 (41.7%) .706 2 (28.6%) 8 (40.0%) .678
Foot deformity 9 (60.0%) 11 (91.7%) .091 6 (85.7%) 14 (70.0%) .633
Previous DFU 13 (86.7%) 12 (100%) .487 6 (85.7%) 19 (95.0%) .459
Outcomes of interest
Daily number of bouts of MVPAa 0.20 [0-0.63] 0.58 [0.17-1.89] .079^ 0.20 [0-0.67] 0.29 [0-0.95] .758
(number per day)
Duration per bout of MVPAb 15 [12-18] 13 [11-17] .631 12 [11-13] 15 [13-18] .266
(minutes per bout)
Intensity per bout of MVPAc (METs per bout) 3.7 [3.5-4.0] 4.2 [3.9-4.3] .023* 4.1 [3.6-4.2] 3.9 [3.7-4.2] .800
Accumulated daily time spent doing MVPAd 28 [12-42] 66 [24-109] .017* 32 [12-69] 37 [18-64] .782
(minutes per day)
Accumulated daily time spent doing sedentary 837 ± 103 745 ± 125 .046* 800 ± 116 795 ± 125 .921
activitye (minutes per day)
Intensity per sedentary activityf (METs per 0.92 ± 0.07 1.00 ± 0.11 .017* 0.95 ± 0.11 0.97 ± 0.10 .894
sedentary activity)

*Groups that are significantly different from each other (P < .05). ^Groups that show a trend to being different from each other (P < .08).
a
>3 METs for >10 consecutive minutes. bConsecutive time until eligible bout ends. cAverage METs per eligible bout. dAny time recording >3 METs per
minute. eAny time recording <1.5 METs per minute. fAverage METs per minute doing any sedentary-intensity activity.

during this sedentary time (1.00 ± 0.11, 0.92 ± 0.07; P < of accumulated moderate-to-vigorous-intensity physical
.05; d = 0.87). There were no significant differences between activity on average per day.8,16 However, as this recommenda-
those who wore nonremovable knee-high devices and those tion defines exercise as a bout of at least 10 consecutive min-
wearing other offloading devices in any of their baseline utes of moderate-to-vigorous-intensity physical activity, our
characteristics (P > .15) or outcomes of interests (P > .25). findings actually show that no group met these exercise rec-
ommendations as none performed at least one of these bouts
per day to be defined as exercise.8,16 We observed that people
Discussion
with diabetes, DPN, or DFU rarely perform a 10-minute bout
This is the first study to observe the bouts and accumulated of moderate-to-vigorous-intensity physical activity, and thus
time people with DFU spent performing moderate-to-vigor- rarely achieve any of the health benefits that regular exercise
ous-intensity physical activity in their everyday environ- provides.8,16 In people with established diabetes, these find-
ment. We observed people with DFU performed similar low ings may indicate that the presence of DPN or DFU does not
numbers of daily bouts and accumulated time doing moder- impact on moderate-intensity physical activity levels.
ate-to-vigorous-intensity physical activity to those with DPN However, all groups had low levels of moderate-to-vigorous-
or DM. However, people with DFU spent more daily time intensity physical activity indicating that people with DM,
doing activity of sedentary intensity. In DFU subgroups, DPN, and DFU might not be willing or able to sustain a mini-
those with minor amputations spent more time performing mum of 10 consecutive minutes of moderate-intensity physi-
moderate-to-vigorous-intensity physical activity and less cal activity. Further studies are required to investigate first the
time sedentary than those without minor amputations. ability of people with DPN and DFU to perform regular bouts
Our findings initially suggest all three groups met interna- of moderate-to-vigorous-intensity physical activity (exer-
tional guideline recommendations for performing 30 minutes cise), and second the impact that achieving these recom-
of exercise per day, with all groups achieving over 30 minutes mended regular bouts have on their general health as well as
832 Journal of Diabetes Science and Technology 13(5)

their foot health, such as foot ulcer development, prevention DFU subgroup, and (3) being generally sicker with more
and healing.26,42,43 comorbidities as there were no differences in comorbidities
We also observed that all groups were highly sedentary, between groups. Although we are unable to interpret if the dif-
spending at least 12 waking hours each day (>720 minutes) ferences in mean sedentary activity intensity between the DFU
performing activities that are akin to sitting quietly. Further, group and other groups is clinically important, this difference
our findings showed that the DFU group spent an additional was reported to be statistically different with a moderate-large
hour each day sedentary compared to controls. A simple effect size, and thus any clinical difference cannot be discounted
explanation for this may be that the DFU group slept less until future studies can confirm. Therefore, this again all sug-
and had an extra hour of waking time than the other groups; gests that the most likely hypotheses for this increase in average
however, our original cross-sectional study found no differ- sedentary-intensity activity is either from the increased energy
ences in daily sleep time between these same groups or DFU required to heal the foot ulcer or from a more energy inefficient
subgroups.23 Previous studies have reported that foot com- compensatory gait in those with DFU compared with controls.
plications such as chronic foot pain or foot deformity However, future research is still required to verify these hypoth-
resulted in reductions in physical activity.44 This may partly eses and investigate if there is any general health impact of such
explain our findings as the DFU participants had signifi- increases in sedentary activity intensity.
cantly more foot deformities than controls; however, they Finally, our study observed that those with DFU and minor
did not perform any less physical activity than the other amputations recorded more accumulated daily time doing
groups. Another explanation may be that treating clinicians moderate-to-vigorous-intensity physical activity, less accumu-
often recommend that those with DFU reduce their ambula- lated time doing sedentary-intensity activity and a trend
tory activity to reduce plantar pressure and in turn facilitate toward higher numbers and intensity of bouts moderate-to-
improved DFU healing.42 In our study we provided no vigorous-intensity physical activity than those without minor
advice to participants that should have biased their everyday amputations. Many studies have investigated the impact of
physical activity, although it cannot be discounted that pre- major amputations on a range of physical activity parameters
vious advice may have impacted. Regardless, our findings and have overwhelming found negative impacts;47-49 however,
do suggest that those with DFU are not performing any our results suggest this may not be the case for minor amputa-
alternative bouts of moderate-to-vigorous-intensity physical tions. To our knowledge only one other study has investigated
activity such as upper body exercises and this may be fur- the impact on physical activity parameters of minor amputa-
ther negatively impacting on their general health. tions in people with DFU and also found no negative impact
An interesting finding from this study that supports those on physical activity.29 This suggests minor amputations might
from other studies was that those with DFU had a higher aver- not impact on physical activity and supports reports suggest-
age intensity of sedentary activity. This sounds contradictory, ing that minor amputations should be viewed as a positive pro-
but indicates that people with DFU require more energy or phylactic procedure with limited negative impact on general
higher intensity to perform the same (sedentary) activities as health that may prevent major amputations that have signifi-
controls. The original cross-sectional study reported that these cant detrimental impacts on general health.50
same participants with DFU generally walked less but expended
more overall energy to do this than controls.23 Hypotheses for
Strengths and Limitations
the original study’s findings included that DFU patients may (1)
perform other nonambulatory activity that were not accounted There were several strengths to this study. First, this study
for by simply measuring walking activity, (2) have had to ambu- excluded participants with conditions that have been shown
late in heavier offloading devices, (3) have more comorbidities to limit physical activity and would have biased findings,
and are generally sicker, which in turn requires more energy to such as peripheral arterial disease, mobility impairment and
function, (4) require more energy to heal their foot ulcer than painful DPN. Second, participant characteristics were gener-
those without ulcers, and (5) have a more inefficient gait that is ally well-matched between groups, except for age and foot
compensating for the DFU itself, foot deformity, minor amputa- deformity. However, it is already well known that people
tions (as indicated by our subgroup results), postural instability with DFU have much higher prevalence of foot deformities
or soft tissue rigidity.23,45,46 Our findings in this study seem to than controls. Regarding age, the DFU group was younger
suggest we may be able to discount the hypotheses of (1) per- than the DPN group and, if anything, could have been
forming other nonambulatory physical activity as we measured expected to be more active, not less active as found. While
all activity in this study and found no differences, (2) ambulat- we could have adjusted for age in our analysis we considered
ing in heavier offloading devices as the majority of people in the against this due to the small numbers in each group present-
DFU group wore removable ankle-high devices (such as post- ing a significant risk of being underpowered for such an
operative shoes) which can be considered similar in weight to adjustment, and the most likely possibility that this would
footwear worn by those in the DPN and DFU group, and the have increased the effect rather than decreased the effect of
small DFU subgroup wearing nonremovable offloading devices our findings and thus not impacted our overall findings. Last,
showed no differences to the removable ankle-high devices participant characteristics and physical activity were both
Lee et al 833

recorded using tools and devices respectively found to be duration or offloading treatment duration. A previous study
reliable and valid to collect such variables.25,31-33,35,38 identified that activity may change in people with DFU over
There were also several limitations to this study. First, time,51 and thus future longitudinal studies are recommended
while we recruited and retained the group numbers required to observe if physical activity changes do occur over time in
by our original sample size calculations (n = 25) for the DFU these populations.
group (n = 27), this wasn’t quite the case for the DPN group
(n = 23) and especially not the case for the DM group (n =
Conclusion
15). Thus, this study was underpowered when comparing
against the DM group in particular; however, our use of two We found that people with DFU, DPN or DM did not meet
“control” groups (DPN and DM groups) and the reported the international guideline recommendations for regular
effect sizes give some context of the potential differences in exercise (bouts of moderate-to-vigorous-intensity physical
physical activity outcomes for cases with DFU compared to activity) that have been found to achieve general health ben-
control groups.24 Second, our comparison of DFU subgroups efits. We also found all groups were highly sedentary, and
resulted in some small subgroups and potentially underpow- those with DFU more so. Interestingly, we observed that
ered findings, particularly for the nonremovable knee-high people with DFU and minor amputations were more physi-
offloading device group with seven participants. While the cally active and less sedentary compared to those with DFU
significance and effect size findings of these subgroups pro- without a minor amputation. More work is now needed to
vide some direction they should be viewed with caution until investigate interventions that may increase exercise of mod-
future adequately powered studies can confirm these find- erate-to-vigorous-intensity physical activity, and reduce sed-
ings. Third, due to these small subgroups we lacked the power entary-intensity activity, in people with DPN and DFU to
to be able to investigate combined subgroups of those with determine their effects on general and foot health.
minor amputations wearing nonremovable knee-high offload-
ing devices compared to other offloading devices for exam- Abbreviations
ple. Such further analyses in larger groups may have shed ANOVA, analysis of variance; BMI, body mass index; CKD,
light on the reasons for differences seen in the DFU subgroup chronic kidney disease; CVD, cardiovascular disease; DFU, diabe-
with minor amputations in particular. Fourth, although the tes-related foot ulcer; DM, diabetes mellitus; DPN, diabetes-related
other offloading device subgroup consisted mainly of similar peripheral neuropathy; HbA1c, glycosylated hemoglobin; HREC,
types of removable ankle-high devices or footwear, unlike the human research ethics committee; IQR, interquartile range; LSD,
nonremovable offloading device group this was not a homog- least significance difference; MET, metabolic equivalent task;
enous offloading device group. Future studies should investi- MVPA, moderate-to-vigorous-intensity physical activity; PA,
gate the impact of specific offloading device types on daily physical activity; QHRFF, Queensland High Risk Foot Form; SD,
standard deviation.
activity to determine their specific impact on daily activity
and offloading. Fifth, while we captured and reported nearly
all demographic and comorbidity variables recommended for Acknowledgments
collection by diabetic foot studies, we did not capture educa- We gratefully acknowledge the help and support of the staff and
tional, socioeconomic or employment status24 which may patients at the Community Diabetes Service and Department of
have also had some impacts on daily activity. Sixth, although Podiatry, Metro North Hospital and Health Service, Australia. We
the SenseWear armband has been demonstrated to be valid would also like to thank the Allied Health Research Collaborative
from The Prince Charles Hospital for data storage and availability.
and reliable for recording physical activity parameters in dif-
ferent populations with chronic conditions,31-33,35,38,39 it has
yet to be validated in a DFU population. Seventh, the interna- Declaration of Conflicting Interests
tional exercise guideline recommendations used as the basis The author(s) declared no potential conflicts of interest with respect
for this study were for 18- to 65-year-old adults and may not to the research, authorship, and/or publication of this article.
readily apply to our older population.8,16 However, our DFU
population in particular had a mean age of 56 years, so most Funding
of these participants would have still been in the age range for The author(s) disclosed receipt of the following financial support
these recommendations. Eighth, our definition of a bout was for the research, authorship, and/or publication of this article: This
very tight8,16 and did not account for any long moderate-to- work was originally supported by a research grant from the Prince
vigorous-intensity physical activities that may have been bro- Charles Hospital Foundation. The funder had no involvement in the
ken up by a minute or two of <3 METs or for any long bout study design, data collection, data analysis, manuscript preparation,
durations (such as a 10-minute bout being counted the same or publication decisions. Furthermore, no funding was provided to
way as a 60-minute bout). However, this was the same for all carry out the secondary analyses for this particular study.
groups and the duration and intensity of bouts were found to
also be low along with the number. Last, this study was not ORCID iD
controlled to represent a particular phase of DFU healing, Peter A Lazzarini https://orcid.org/0000-0002-8235-7964
834 Journal of Diabetes Science and Technology 13(5)

References American College of Sports Medicine and the American Heart

Association. Med Sci Sports Exerc. 2007;39:1423-1434.
1. Vos T, Allen C, Arora M, et al. Global, regional, and national
17. Ainsworth BE, Haskell WL, Herrmann SD, et al. 2011 com-
incidence, prevalence, and years lived with disability for 310
pendium of physical activities: a second update of codes and
diseases and injuries, 1990-2015: a systematic analysis for the
MET values. Med Sci Sports Exerc. 2011;43:1575-1581.
Global Burden of Disease Study 2015. Lancet. 2016;388:1545-
18. Nolan RC, Raynor AJ, Berry NM, May EJ. Self-reported
1602.
physical activity using the International Physical Activity
2. Bommer C, Heesemann E, Sagalova V, et al. The global eco-
Questionnaire (IPAQ) in Australian adults with type 2 diabe-
nomic burden of diabetes in adults aged 20-79 years: a cost-of-
tes, with and without peripheral neuropathy. Can J Diabetes.
illness study. Lancet Diabetes Endocrinol. 5:423-430.
2016;40:576-579.
3. Dieleman JL, Baral R, Birger M, et al. US spending on per-
19. Vanden Bosch ML, Robbins LB, Anderson K. Correlates of
sonal health care and public health, 1996-2013. JAMA. 2016;
physical activity in middle-aged women with and without dia-
316:2627-2646.
betes. West J Nurs Res. 2015;37:1581-1603.
4. Armstrong DG, Boulton AJM, Bus SA. Diabetic foot ulcers
20. Guillory IK, Cullen KW, Thompson D, Watson KB. Physical
and their recurrence. N Engl J Med. 2017;376:2367-2375.
activity in youth with well-controlled versus poorly controlled
5. Lazzarini PA, Pacella RE, Armstrong DG, Van Netten JJ.
type 2 diabetes. Clin Pediatr. 2012;51:354-358.
Diabetes-related lower-extremity complications are a lead-
21. Shaibi GQ, Michaliszyn SB, Fritschi C, Quinn L, Faulkner
ing cause of the global burden of disability. Diabetic Med.
MS. Type 2 diabetes in youth: a phenotype of poor cardiore-
2018;35:1297-1299.
spiratory fitness and low physical activity. Int J Pediatr Obes.
6. Lazzarini PA, Hurn SE, Kuys SS, et al. The silent overall bur-
2009;4:332-337.
den of foot disease in a representative hospitalised population.
22. Jakicic JM, Gregg E, Knowler W, et al. Activity patterns of
Int Wound J. 2017;14:716-728.
obese adults with type 2 diabetes in the look AHEAD study.
7. Schaper NC, Van Netten JJ, Apelqvist J, Lipsky BA, Bakker K;
Med Sci Sports Exerc. 2010;42:1995-2005.
on behalf of the International Working Group on the Diabetic
23. Sheahan H, Canning K, Refausse N, et al. Differences in the
Foot. Prevention and management of foot problems in diabe-
daily activity of patients with diabetic foot ulcers compared
tes: a Summary Guidance for Daily Practice 2015, based on
to controls in their free-living environments. Int Wound J.
the IWGDF Guidance Documents. Diabetes Metab Res Rev.
2017;14:1175-1182.
2016;32:7-15.
24. Jeffcoate WJ, Bus SA, Game FL, Hinchliffe RJ, Price PE,
8. Colberg SR, Sigal RJ, Yardley JE, et al. Physical activity/
Schaper NC. Reporting standards of studies and papers on
exercise and diabetes: a position statement of the American
the prevention and management of foot ulcers in diabetes:
Diabetes Association. Diabetes Care. 2016;39:2065-2079.
required details and markers of good quality. Lancet Diabetes
9. Thomas DE, Elliott EJ, Naughton GA. Exercise for type 2 dia-
Endocrinol. 2016;4:781-788.
betes mellitus. Cochrane Database Syst Rev. 2006;Cd002968.
10. Way KL, Hackett DA, Baker MK, Johnson NA. The effect of 25. Lazzarini PA, Ng V, Kinnear EM, et al. The Queensland High
regular exercise on insulin sensitivity in type 2 diabetes mel- Risk Foot Form (QHRFF)—is it a reliable and valid clinical
litus: a systematic review and meta-analysis. Diabetes Metab research tool for foot disease? J Foot Ankle Surg. 2014;7:7.
J. 2016;40:253-271. 26. Bus SA, Armstrong DG, van Deursen RW, et al. IWGDF guid-
11. Francois ME, Durrer C, Pistawka KJ, Halperin FA, Little JP. ance on footwear and offloading interventions to prevent and
Resistance-based interval exercise acutely improves endo- heal foot ulcers in patients with diabetes. Diabetes Metab Res
thelial function in type 2 diabetes. Am J Physiol Heart Circ Rev. 2016;32:25-36.
Physiol. 2016;311:H1258-H1267. 27. Maluf KS, Mueller MJ. Novel Award 2002. Comparison of
12. Bakrania K, Yates T, Edwardson CL, et al. Associations of physical activity and cumulative plantar tissue stress among
moderate-to-vigorous-intensity physical activity and body subjects with and without diabetes mellitus and a history
mass index with glycated haemoglobin within the general pop- of recurrent plantar ulcers. Clin Biomech (Bristol, Avon).
ulation: a cross-sectional analysis of the 2008 Health Survey 2003;18:567-575.
for England. BMJ Open. 2017;7:e014456. 28. Ponsonby AL, Sun C, Ukoumunne OC, Pezic A, Venn A,
13. de Lade CG, Marins JC, Lima LM, et al. Effects of different Shaw JE, et al. Objectively measured physical activity and
exercise programs and minimal detectable changes in hemo- the subsequent risk of incident dysglycemia: The Australian
globin A1c in patients with type 2 diabetes. Diabetol Metab Diabetes, Obesity and Lifestyle Study (AusDiab). Diabetes
Syndr. 2016;8:13. Care. 2011;34:1497-1502.
14. Liubaoerjijin Y, Terada T, Fletcher K, Boule NG. Effect of 29. Kanade RV, van Deursen RWM, Harding K, Price P. Walking
aerobic exercise intensity on glycemic control in type 2 dia- performance in people with diabetic neuropathy: benefits and
betes: a meta-analysis of head-to-head randomized trials. Acta threats. Diabetologia. 2006;49:1747-1754.
Diabetol. 2016;53:769-781. 30. Najafi B, Crews RT, Wrobel JS. Importance of time spent
15. Lamb MJ, Westgate K, Brage S, et al. Prospective associa- standing for those at risk of diabetic foot ulceration. Diabetes
tions between sedentary time, physical activity, fitness and Care. 2010;33:2448-2450.
cardiometabolic risk factors in people with type 2 diabetes. 31. Mignault D, St.-Onge M, Karelis AD, Allison DB, Rabasa-
Diabetologia. 2016;59:110-120. Lhoret R. Evaluation of the portable HealthWear Armband: a
16. Haskell WL, Lee IM, Pate RR, et al. Physical activity and device to measure total daily energy expenditure in free-living
public health: updated recommendation for adults from the type 2 diabetic individuals. Diabetes Care. 2005;28:225-227.
Lee et al 835

32. St-Onge M, Mignault D, Allison DB, Rabasa-Lhoret R. physical activity and exercise in patients at risk of diabetic foot
Evaluation of a portable device to measure daily energy expen- ulcers: a critical review. Diabetes Metab Res Rev. 2016;32:791-
diture in free-living adults. Am J Clin Nutr. 2007;85:742-749. 804.
33. van Schooten KS, Rispens SM, Elders PJ, Lips P, van Dieen 43. Bus SA, van Netten JJ, Lavery LA, et al. IWGDF guidance on
JH, Pijnappels M. Assessing physical activity in older adults: the prevention of foot ulcers in at-risk patients with diabetes.
required days of trunk accelerometer measurements for reliable Diabetes Metab Res Rev. 2016;32:16-24.
estimation. J Aging Phys Act. 2015;23:9-17. 44. Muchna A, Najafi B, Wendel CS, Schwenk M, Armstrong
34. Scheers T, Philippaerts R, Lefevre J. Variability in physical DG, Mohler J. Foot problems in older adults associations with
activity patterns as measured by the SenseWear Armband: how incident falls, frailty syndrome, and sensor-derived gait, bal-
many days are needed? Eur J Appl Physiol Occup Physiol. ance, and physical activity measures. J Am Podiatr Med Assoc.
2012;112:1653-1662. 2018;108:126-139.
35. Van Remoortel H, Raste Y, Louvaris Z, et al. Validity of 45. Petrovic M, Deschamps K, Verschueren SM, et al. Is the meta-
six activity monitors in chronic obstructive pulmonary dis- bolic cost of walking higher in people with diabetes? J Appl
ease: a comparison with indirect calorimetry. PLOS ONE. Physiol. 2016;120:55-62.
2012;7:e39198-e. 46. Toosizadeh N, Mohler J, Armstrong DG, Talal TK, Najafi B.
36. Johannsen DL, Calabro MA, Stewart J, Franke W, Rood JC, The influence of diabetic peripheral neuropathy on local pos-
Welk GJ. Accuracy of armband monitors for measuring daily tural muscle and central sensory feedback balance control.
energy expenditure in healthy adults. Med Sci Sports Exerc. PLOS ONE. 2015;10:e0135255-e.
2010;42:2134-2140. 47. Starholm IM, Mirtaheri P, Kapetanovic N, et al. Energy
37. BodyMedia. SenseWear Wearable Multi-sensor Armband expenditure of transfemoral amputees during floor and tread-
Mittagong. Australia: Temple Healthcare Pty Ltd; 2016. mill walking with different speeds. Prosthet Orthot Int. 2016;
Available at: www.templehealthcare.com.au. 40:336-342.
38. Mackey DC, Manini TM, Schoeller DA, et al. Validation of an 48. Desveaux L, Goldstein RS, Mathur S, et al. Physical activity in
armband to measure daily energy expenditure in older adults. J adults with diabetes following prosthetic rehabilitation. Can J
Gerontol A Bio Sci Med Sci. 2011;66:1108-1113. Diabetes. 2016;40:336-341.
39. Tierney M, Fraser A, Purtill H, Kennedy N. Study to determine 49. Paxton RJ, Murray AM, Stevens-Lapsley JE, Sherk KA,
the criterion validity of the SenseWear Armband as a mea- Christiansen CL. Physical activity, ambulation, and comor-
sure of physical activity in people with rheumatoid arthritis. bidities in people with diabetes and lower-limb amputation. J
Arthritis Care Res. 2013;65:888-895. Rehabil Res Dev. 2016;53:1069-1078.
40. Cohen JA. Statistical Power Analysis for the Behavioral 50. Jeffcoate WJ, van Houtum WH. Amputation as a marker of the
Sciences. 2nd ed. Mahwah, NJ: Lawrence Erlbaum; 1988. quality of foot care in diabetes. Diabetologia. 2004;47:2051-
41. Hopkins WG. A scale of magnitudes for effect statistics. A new 2058.
view of statistics. Sports Sci. 2002. Available at: www.sportsci 51. Najafi B, Grewal GS, Bharara M, Menzies R, Talal TK,
.org/resource/stats/effectmag.html. Accessed December 25, 2011. Armstrong DG. Can’t stand the pressure: the association
42. Crews RT, Schneider KL, Yalla SV, Reeves ND, Vileikyte between unprotected standing, walking, and wound healing in
L. Physiological and psychological challenges of increasing people with diabetes. J Diabetes Sci Technol. 2017;11:657-667.