RESEARCH ARTICLE

Physical Activity and Abnormal Blood Glucose Among

Healthy Weight Adults
Arch G. Mainous III, PhD,1,2 Rebecca J. Tanner, MA,1 Stephen D. Anton, PhD,3 Ara Jo, MS,1
Maya C. Luetke, MSPH1

Introduction: Physical activity has been linked to prevention and treatment of prediabetes and
diabetes in overweight and obese adults. This study examines the relationship between low physical
activity levels and risk of abnormal blood glucose (prediabetes or undiagnosed diabetes) in healthy
weight adults.

Methods: Data from the 2014 Health Survey for England were analyzed in July 2016, focusing on
adults with a BMI Z18.5 and o25 who had never been diagnosed with diabetes (N¼1,153).
Abnormal blood glucose was defined as hemoglobin A1c Z5.7. Physical activity was measured
through the International Physical Activity Questionnaire. Bivariate analyses and Poisson models
were conducted on the effect of physical activity on abnormal blood glucose, controlling for age, sex,
waist to hip ratio, sitting time, age X physical activity interaction, sex X physical activity, and race.

Results: Abnormal blood glucose was detected in 23.7% of individuals with low activity levels,
14.8% of those with medium activity levels, and 12.2% of those with high activity levels (po0.003).
Similarly, 25.4% of inactive individuals (physically active for o30 minutes per week) were more
likely to have abnormal blood glucose levels than active individuals (13.4%, po0.0001). Higher
physical activity was associated with a lower likelihood of abnormal blood glucose in an adjusted
Poisson regression.

Conclusions: Among healthy weight adults, low physical activity levels are significantly associated
with abnormal blood glucose (prediabetes and undiagnosed diabetes). These findings suggest that
healthy weight individuals may benefit from physical exercise.
Am J Prev Med 2016;](]):]]]–]]]. & 2016 American Journal of Preventive Medicine. Published by Elsevier Inc.
All rights reserved.

INTRODUCTION detection of abnormal glucose tend to focus on individ-

uals who are overweight or obese.4,6 Recent data have

D
iabetes has reached epidemic proportions.
Worldwide, the number of people with diabetes
has risen from 108 million in 1980 to 422
million in 2014.1 Diabetes is a major cause of kidney
failure, heart attacks, stroke, and lower limb amputa-
tions.2 In addition to the morbidity and mortality
associated with diabetes, the cost of diabetes care is
substantial.3
Early detection and screening for undiagnosed Type 2
diabetes is needed because of the utility of treatment to
prevent Type 2 diabetes complications.4 Second, there is
an equal or more important need to detect prediabetes, a
state of abnormal blood glucose that indicates high risk
for the development of Type 2 diabetes.5 Strategies for
indicated, however, that a substantial proportion of
individuals at a “healthy weight” (BMI between 18.5
and 24.9) have prediabetes.7 In fact, among healthy
weight individuals aged 45 years and older in the U.S.,
From the 1Department of Health Services Research, Management and
Policy, University of Florida, Gainesville, Florida; 2Department of Com-
munity Health and Family Medicine, University of Florida, Gainesville,
Florida; and 3Department of Aging and Geriatric Research, University of
Florida, Gainesville, Florida
Address correspondence to: Arch G. Mainous III, PhD, Department of
Health Services Research, Management, and Policy, University of Florida,
Health Sciences Center, PO Box 100195, Gainesville FL 32610. E-mail:
arch.mainous@phhp.ufl.edu.
0749-3797/$36.00
http://dx.doi.org/10.1016/j.amepre.2016.11.027

& 2016 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights Am J Prev Med 2016;](]):]]]–]]] 1
reserved.
2 Mainous III et al / Am J Prev Med 2016;](]):]]]–]]]
the prevalence of prediabetes was 33.1% in 2012.7 One
theory for this substantial proportion of healthy weight,
prediabetic adults in the U.S. is that sedentary lifestyle
may contribute to unhealthy changes in body composi-
tion, even among individuals who are not overweight.
Some individuals at healthy weights have what has been
referred to as “normal-weight obesity,” a condition
characterized by high body fat and lower lean muscle
mass while still falling within the BMI parameters of
“healthy” weight.8
This study is innovative in examining abnormal blood
glucose in healthy weight adults, a population previously
believed to be at low risk of glucose abnormalities.
Because typical strategies to prevent prediabetes or Type
2 diabetes (e.g., calorie restriction and weight reduction)
are inappropriate for individuals who are not overweight,
it is unclear whether physical activity moderates the
presence of abnormal glucose in this healthy weight
population. The purpose of this study was to examine the
relationship of physical activity with abnormal blood
glucose among adults at a healthy weight who previously
had not been diagnosed with diabetes.
METHODS
Data from the 2014 Health Survey for England (HSE) were
analyzed. The HSE is an annual survey that monitors trends in
the health of England’s residents. The HSE uses a stratified random
probability sample of households. Information on the sampling
methodology of the HSE is available from the United Kingdom
Data Service.9 The HSE includes both a computer-assisted
personal interview and a nurse visit. For the 2014 HSE, of those
respondents with full interviews, 5,941 adults aged Z16 years and
1,249 children aged 0–15 years had a visit with a nurse.9 During the
nurse visit, a nurse obtained blood, saliva, and urine samples, along
with physical measurements of the participant.
To obtain accurate estimates, the HSE is weighted by the Joint
Health Surveys Unit of NatCen Social Research and University
College London to account for sampling design, non-response, and
the type of data used.10,11 Using the weighting variable allows
researchers to make estimates for the entire population in England
and reduce possible biases.9
The current project focused on adults aged Z20 years, who had
a nurse-measured BMI Z18.5 and o25, who reported never being
diagnosed with diabetes, and who had data available on glycated
hemoglobin (HbA1c). Based on these criteria, there were a total of
1,153 respondents (weighted sample size, 1,219). The weighted
sample size is the sample size when the weighting variable is
applied to the sample. Data were analyzed in July 2016. The study
used publicly available de-identified data and was exempted by the
University of Florida IRB.
Measures
Individuals were considered to have diagnosed diabetes if they
reported that they had been diagnosed with diabetes (other than
gestational) by a healthcare provider.
Participants with abnormal blood glucose were identified using
HbA1c. Abnormal blood glucose was defined as HbA1c Z5.7%.6
Owing to their increased risk of all-cause mortality, individuals
with HbA1c o4.0 were removed from the analysis.12 On August 4,
2016, HSE 2014 users were notified of the need to correct a
calibration problem with the HbA1c values in the HSE for 2014
(K Dennison, UK Data Service, personal/written communication,
2016). The HbA1c levels used for this paper have been corrected as
recommended by NatCen Social Research, the organization that
jointly conducts the HSE on behalf of the Health and Social Care
Information Centre. For HbA1c values between 3.5 and 6.62, 0.1
was added to the given HbA1c level. For HbA1c values between 6.3
and 8.9, 0.2 was added to the given HbA1c level. For HbA1c values
48.9, 0.3 was added to the given HbA1c level. There were 953
individuals of normal BMI who did not have diabetes and were
aged 420 years and did not have HbA1c results. Although this has
the potential to introduce bias, the weighting variable took into
account respondents who refused or were unable to give a blood
sample to partially mitigate the effect of missing data.
Physical activity was defined in two ways. First, the HSE includes a
derived variable of tertiles of moderate or vigorous intensive minutes
of activity per week based on the International Physical Activity
Questionnaire questions in the HSE.13 The tertiles are sex specific,
and exclude walking.14 Activity level is characterized as low, medium,
and high. For men, low physical activity was 0–120 minutes of
moderate- to vigorous-intensity physical activity (MVPA) per week;
medium MVPA was 121–840 minutes per week; and high MVPA
was Z841 minutes of MVPA per week. For women, low MVPA was
defined as 0 minutes of MVPA per week, medium MVPA was
defined as between 10 and 496 minutes of MVPA, and high MVPA
was defined as Z496 minutes MVPA per week.15 The HSE MVPA
measures included some values that were extremely high. Out of
concern that these extremely high and unrealistic values would bias
the analysis, respondents reporting 45,000 minutes of MVPA per
week were excluded from the analysis. In doing so, 26 respondents
were recoded as missing. Additionally, of individuals aged Z20 years
with available HbA1c measures and a normal BMI, a total of 156 were
missing data on physical activity.
The second way physical activity was operationalized was
whether a person was active Z30 minutes per week, according
to their responses on the International Physical Activity Ques-
tionnaire. Individuals were considered either active (Z30 minutes
of MVPA per week) or inactive (o30 minutes MVPA per week).
Sitting time was analyzed. The HSE asks respondents how many
minutes they usually spend sitting on a weekday.
Waist to hip ratio (WHR) was assessed because of its utility as a
measure of abdominal obesity, which has been associated with
metabolic problems.14,16,17 The HSE includes a derived variable,
mean WHR. Waist circumference was measured midway between
the iliac crest and the costal margin twice.9 The hip circumference
was measured at the widest circumference over the buttocks and
below the iliac crest twice.9 The means of each number were used
to create a mean WHR.9 A WHR 40.85 in women or 40.90 in
men was considered unhealthy.18
For this analysis, respondents were split into two age groups,
those aged 20–44 years and those aged Z45 years. The American
Diabetes Association recommends screening for abnormal blood
glucose for all adults aged 445 years, as the risk of developing
diabetes increases with age.4 Race/ethnicity was categorized as
white, Asian, and other. Sex was defined as male and female.
www.ajpmonline.org
 Mainous III et al / Am J Prev Med 2016;](]):]]]–]]] 3
Statistical Analysis Table 2. Physical Activity Characteristics
This study used SAS, version 9.4, for all analyses. Bivariate analyses
were conducted on the relationship between physical activity and Characteristic % of sample
abnormal blood glucose. The sample was stratified based on sex
Physical activity tertile
and age, examining the bivariate relationships among men and
Low activity 26.2
women and for those aged 20–44 years and Z45 years. Unadjusted
Medium activity 38.8
Poisson models with robust SEs were computed examining the
effect of each physical activity variable on abnormal blood glucose. High activity 35.0
Adjusted Poisson models with robust SEs were computed exam- Physical activity level
ining the effect of each physical activity variable on abnormal Inactive/o30 minutes’ MVPA per week 23.3
blood glucose, controlling for age, sex, WHR, sitting time, age X Z30 minutes’ MVPA per week 76.7
physical activity interaction, and sex X physical activity interac- Usual minutes sitting per weekday
tion. To check for potential multicollinearity between physical r270 47.2
activity variables and sitting time, the correlation between sitting 4270 52.8
time and minutes of MVPA per week was computed. The
MVPA, moderate to vigorous physical activity.
correlation was –0.15. All analyses were weighted using the weight
appropriate for variables derived from the blood sample to account
for survey design and non-response. The weighting variable medium MVPA (12.4%) and high MVPA (9.9%).
allowed the authors to make estimates of the English population.9 Among women, the relationship between activity level
and abnormal blood glucose was not statistically signifi-
cant (low MVPA, 23.6%; medium MVPA, 16.5%; high
RESULTS MVPA, 14.2%; p=0.06).
The sample comprised adults aged Z20 years who had Among those aged 20–44 years, there was not a
not been diagnosed with diabetes by a healthcare significant difference in abnormal blood glucose by
provider and had a healthy BMI. Tables 1 and 2 show activity level (low MVPA, 6.1%; medium MVPA, 5.8%;
the demographic and physical activity characteristics of high MVPA, 3.0%; p=0.25). Among adults aged Z45
the sample. The relationship between abnormal blood years, there was a significant difference in the proportion
glucose and MVPA tertile was statistically significant with abnormal blood glucose between activity levels (low
(p=0.0003). Among individuals with low MVPA, 23.7% MVPA, 43.5%; medium MVPA, 29.5%; high MVPA,
had abnormal blood glucose compared with 14.8% of 27.3%; p=0.009). Table 3 shows the prevalence ratios
those with medium MVPA, and 12.2% of those with high from adjusted and unadjusted Poisson regression models
MVPA. Among men, physical activity level was signifi- examining the effect of MVPA tertile on abnormal blood
cantly associated with abnormal blood glucose glucose. The results of Poisson regression models indi-
(p=0.003). Among men with low MVPA, 23.7% had cated that low physical activity was associated with the
abnormal blood glucose, compared with those with presence of abnormal blood glucose. The adjusted results
presented here do not include the age X physical activity
Table 1. Demographic Characteristics (Unweighted Sample or sex X physical activity interaction terms as they were
Size 1,153; Weighted Sample Size¼1,219) not significant in the model.
There was a statistically significant difference in blood
Lowest Middle Highest
glucose levels between inactive and active individuals.
% of activity activity activity
Characteristic sample tertile tertile tertile Among inactive individuals (those physically active for
o30 minutes per week), 25.4% had abnormal blood
Age, years
glucose compared with 13.4% of those who had Z30
20–44 58.8 53.1 62.2 62.0
minutes of MVPA per week (po0.0001). Among men,
Z45 41.2 46.9 37.8 38.0
physical activity level was significantly associated with
Sex
abnormal blood glucose (po0.0001). Among inactive
Male 44.4 41.7 42.0 58.0
Female 55.6 58.3 46.1 53.9
men, 29.4% had abnormal blood glucose, compared with
Race 11.0% among men with Z30 minutes of MVPA per week
White 89.1 87.2 91.3 89.8 (po0.0001). Among women, the relationship between
Asian 7.3 10.6 4.7 7.4 activity level and abnormal blood glucose was also
Other 3.6 2.2 4.0 2.8 significant, with 23.5% of inactive women with abnormal
Unhealthy WHR 21.2 33.3 19.5 12.4 blood glucose compared with 15.5% of women with Z30
Abnormal blood 16.8 23.7 14.8 12.2 minutes of MVPA per week (p¼0.02).
glucose Among those aged 20–44 years, there was not a
WHR, waist to hip ratio. significant difference in abnormal blood glucose by

] 2016
4 Mainous III et al / Am J Prev Med 2016;](]):]]]–]]]
Table 3. Unadjusted and Adjusted Prevalence Ratios for These findings are consistent with previous evidence
Physical Activity Predicting Abnormal Blood Glucose that healthy weight populations with lower physical
activity levels exhibit higher Type 2 diabetes prevalence
Prevalence ratio rates.19 Previous studies have found that low physical
Model (95% CI)
activity contributes to risk of prediabetes or Type 2
Unadjusted results diabetes.20–22 However, these have focused almost exclu-
Physical activity tertiles sively on overweight/obese populations. This is one of the
Low physical activity 1.94 (1.35, 2.78) first studies that looks at diabetes prevalence and physical
Medium physical activity 1.21 (0.84, 1.75) activity in the healthy weight population. Insulin sensi-
High physical activity 1.00
tivity and other metabolic improvements have been
Physical activity level
associated with exercise-induced visceral fat loss, rather
Inactive/o30 minutes per week 1.90 (1.41, 2.56)
than overall weight loss.23,24 Further, targeted exercise
Active for Z30 minutes per week 1.00
programs have been shown to relieve symptoms and
Adjusted results
Physical activity tertilesa
comorbidities in adults with Type 2 diabetes regardless of
Low physical activity 1.48 (1.04, 2.11) weight loss.25 Therefore, it is important to examine the
Medium physical activity 1.15 (0.81, 1.64) phenomenon of abnormal blood glucose in healthy
High physical activity 1.00 weight individuals and to investigate the effect of physical
Physical activity levelb activity on the blood glucose levels of these individuals.
Inactive/o30 minutes per week 1.66 (1.02, 2.70) Exercise has been shown to be a modifiable and cost-
Active for Z30 minutes per week 1.00 effective lifestyle intervention in reducing unnecessary
a
Adjusted for age, sex, race, sitting time, unhealthy WHR. weight gain and controlling blood glucose.26–28 Histor-
b
Adjusted for age, sex, age X physical activity interaction, sex X physical ically, the focus of Type 2 diabetes prevention and
activity interaction, race, sitting time, unhealthy WHR.
WHR, waist to hip ratio.
treatment programs has been primarily on calorie
restriction and weight loss. However, this study and
physical activity versus inactivity (inactive, 7.0%; Z30 others suggest that physical activity, not solely weight
minutes of MVPA, 4.3%; p¼0.21). Among adults aged loss, provides the protective and therapeutic effects on
Z45 years, there was a significant difference in the blood glucose levels.29,30 Accordingly, physical activity
proportion with abnormal blood glucose between phys- should be a crucial tenet of future prevention interven-
ical activity and inactivity (inactive, 45.7%; Z30 minutes tions, especially in the healthy weight population.
of MVPA, 28.3%; p¼0.0006). Table 3 shows the preva- Studies have shown that physical activity boosts
lence ratios from adjusted and unadjusted Poisson resting metabolic rates and increases lean muscle mass,
regression models examining the effect of physical which have been linked to improved insulin action.31–33
activity versus inactivity on abnormal blood glucose. Muscle strength, decreases in visceral adiposity,
Being inactive was associated with an increased preva- and improved insulin sensitivity have all been associated
lence of abnormal blood glucose. These presented results with higher levels of physical activity.34,35 Future
do include the interaction terms for age X physical studies should investigate diabetes risk, exercise, lean
activity and sex X physical activity. Men who were muscle mass, and visceral adiposity in healthy weight
inactive had a higher prevalence ratio of abnormal blood individuals.
glucose than men who were active (1.95, 95% CI¼1.05, Two different measures of physical activity were used.
3.60). Respondents aged Z45 years who were inactive Physical activity is a more general concept than exercise,
also had a higher prevalence ratio of abnormal blood which refers to structured or planned activities, and is
glucose (1.52, 95% CI¼1.14, 2.03). commonly evaluated as leisure time exercise. Many of the
recommendations on physical activity are actually
focused on the subset of activity for leisure time exercise,
DISCUSSION thereby missing activity in the workplace.36 In this study
The key finding of this study is that low physical activity using two different measures of physical activity, both
levels are associated with abnormal blood glucose in measures produced similar results, showing that a
healthy weight adults. These results help us to better healthy weight population with low levels of physical
understand a potential intervention for healthy weight activity is more likely to have prediabetes or undiagnosed
individuals who may not be metabolically healthy. Type 2 diabetes than those who report higher levels of
Further, these results from England add to previous physical activity. This finding provides support for the
findings from the U.S. on the increasing prevalence of association of low levels of physical activity with
abnormal blood glucose among healthy weight adults.7 increased prevalence of abnormal blood glucose among

www.ajpmonline.org
 Mainous III et al / Am J Prev Med 2016;](]):]]]–]]]
this population. In these results, differences between
physical activity tertiles were significant in men but not
in women. This difference in significance between sexes
could be due to not having a large enough sample size for
women or because there may be an optimal level that
men, who were more active, were more likely to meet
than women. Similarly, the fact that results were signifi-
cant in the population aged Z45 years but not in the
population aged 20–44 years is likely the result of the
recognized effect of aging on Type 2 diabetes risk.4,6
Additionally, when terms accounting for the interaction
between age and physical activity and sex and physical
activity were evaluated, the association between physical
activity and abnormal blood glucose was only apparent in
men with no activity in the active versus inactive model
and among those aged Z45 years with no activity in the
same model.
Limitations
The strength of the present study is the use of a nationally
representative survey that includes biomarkers. It pro-
vides evidence from an industrialized country other than
the U.S. This study also goes beyond the focus on
overweight/obesity when examining abnormal blood
glucose and covers new territory by looking at abnormal
blood glucose in the healthy weight population.6 How-
ever, there are some limitations. First, although the
physical activity variables were based on a valid and
reliable scale, the International Physical Activity Ques-
tionnaire, it is based on self-reported data. Another
limitation is that this study did not examine the impact
of dietary factors. Third, the HSE did not collect data on
family history of diabetes. Although family history can
represent both culture and inherited predispositions,
assessing the extent to which genetics plays a role in
abnormal glucose among the healthy weight population
was not within the analytic capabilities of the HSE.
However, regardless of differences in diet and the genetic
component of prediabetes and Type 2 diabetes, physical
activity has been shown to elicit positive physiologic
changes in blood glucose levels and insulin sensitiv-
ity.37,38 Another limitation is that it is possible that
people with abnormal blood glucose might be sicker and
thus less likely to engage in physical activity. However, a
lack of physical activity has been shown to contribute to
unhealthy blood glucose levels.39,40 Additionally, though
sitting time was controlled for, the HSE did not allow for
assessment of duration of sitting episodes. Finally, this
study is cross-sectional and therefore does not examine
the study variables over time. In order to further validate
the connection between low physical activity levels and
abnormal blood glucose in healthy weight individuals, an
] 2016
5
RCT or an observational cohort study should be
conducted.
CONCLUSIONS
The present study provides strong evidence that physical
activity levels are significantly associated with abnormal
blood glucose in healthy weight adults. As noted above,
this population has previously not been considered to be
at high risk of abnormal blood glucose owing to their
healthy BMI. However, these findings suggest that
healthy weight individuals with low levels of activity
represent a population with a higher prevalence of
abnormal blood glucose than previously thought. An
important next step is to examine how physical activity
and other lifestyle interventions can effectively reduce the
risk of abnormal blood glucose among individuals who
are of healthy weight.
ACKNOWLEDGMENTS
No financial disclosures were reported by the authors of
this paper.
REFERENCES
1. WHO. Global report on diabetes 2016. www.who.int/diabetes/global-
report/en/. Accessed August 4, 2016.
2. CDC. National Diabetes Statistics Report: Estimates of Diabetes and Its
Burden in the United States, 2014. Atlanta, GA: U.S. DHHS; 2014.
3. American Diabetes Association. Economic costs of diabetes in the U.S.
in 2012. Diabetes Care. 2013;36(4):1033–1046. http://dx.doi.org/
10.2337/dc12-2625.
4. American Diabetes Association. Diagnosis and classification of dia-
betes mellitus. Diabetes Care. 2013;36(suppl 1):S67–S74. http://dx.doi.
org/10.2337/dc13-S067.
5. Tabák AG, Herder C, Rathmann W, Brunner EJ, Kivimäki M.
Prediabetes: a high-risk state for diabetes development. Lancet.
2012;379(9833):2279–2290. http://dx.doi.org/10.1016/S0140-6736(12)
60283-9.
6. U.S. Preventive Services Task Force (USPSTF). Final recommendation:
abnormal blood glucose and Type 2 diabetes mellitus: screening. www.
uspreventiveservicestaskforce.org/Page/Document/Recommendation
StatementFinal/screening-for-abnormal-blood-glucose-and-type-2-dia
betes. Published 2015. Accessed August 4, 2016.
7. Mainous AG 3rd, Tanner RJ, Jo A, Anton SD. Prevalence of prediabetes
and abdominal obesity among healthy-weight adults: 18-year trend. Ann
Fam Med. 2016;14(4):304–310. http://dx.doi.org/10.1370/afm.1946.
8. Romero-Corral A, Somers VK, Sierra-Johnson J, et al. Normal weight
obesity: a risk factor for cardiometabolic dysregulation and cardiovas-
cular mortality. Eur Heart J. 2010;31(6):737–746. http://dx.doi.org/
10.1093/eurheartj/ehp487.
9. Health Survey for England 2014 User Guide. https://discover.ukdata
service.ac.uk/catalogue/?sn=7919&type=Data%20catalogue. Accessed
July 15, 2016.
10. Sproston K, Primatesta P, eds. Health Survey for England, 2003;
Volume 3: Methodology and Documentation. London, England: The
Stationery Office; 2004.
6 Mainous III et al / Am J Prev Med 2016;](]):]]]–]]]
11. Craig R, Fuller E, Mindell J. Health Survey for England, 2014; Volume 2:
Methods and Documentation. London, England: Health and Social
Care Information Centre 2015.
12. Carson AP, Fox CS, McGuire DK, et al. Low hemoglobin A1c and risk
of all-cause mortality among U.S. adults without diabetes. Circ
Cardiovasc Qual Outcomes. 2010;3(6):661–667. http://dx.doi.org/
10.1161/CIRCOUTCOMES.110.957936.
13. Craig CL, Marshall AL, Sjöström M, et al. International Physical
Activity Questionnaire: 12-country reliability and validity. Med Sci
Sports Exerc. 2003;35(8):1381–1395. http://dx.doi.org/10.1249/01.
MSS.0000078924.61453.FB.
14. Gómez-Ambrosi J, Silva C, Galofré JC, et al. Body mass index
classification misses subjects with increased cardiometabolic risk
factors related to elevated adiposity. Int J Obes (Lond). 2012;36:286–
294. http://dx.doi.org/10.1038/ijo.2011.100.
15. Health Survey for England 2014. Lists of variables and derived
variables. https://discover.ukdataservice.ac.uk/catalogue/?sn=7919&
type=Data%20catalogue. Accessed July 15, 2016.
16. Yusuf S, Hawken S, Ounpuu S, et al. Obesity and the risk of myocardial
infarction in 27,000 participants from 52 countries: a case-control
study. Lancet. 2005;366(9497):1640–1649. http://dx.doi.org/10.1016/
S0140-6736(05)67663-5.
17. de Koning L, Merchant AT, Pogue J, Anand SS. Waist circumference
and waist-to-hip ratio as predictors of cardiovascular events: meta-
regression analysis of prospective studies. Eur Heart J. 2007;28(7):850–
856. http://dx.doi.org/10.1093/eurheartj/ehm026.
18. WHO. Waist circumference and waist–hip ratio. Report of a WHO
expert consultation, Geneva, 8-11 December 2008. http://apps.who.int/
iris/bitstream/10665/44583/1/9789241501491_eng.pdf?ua=1. Accessed
August 4, 2016.
19. Crump C, Sundquist J, Winkleby MA, Sieh W, Sundquist K. Physical
fitness among Swedish military conscripts and long-term risk for Type
2 diabetes mellitus. Ann Intern Med. 2016;164(9):577–584. http://dx.
doi.org/10.7326/M15-2002.
20. Church TS, Blair SN, Cocreham S, et al. Effects of aerobic and
resistance training on hemoglobin A1c levels in patients with Type 2
diabetes: a randomized controlled trial. JAMA. 2010;304(20):2253–
2262. http://dx.doi.org/10.1001/jama.2010.1710.
21. Hordern MD, Dunstan DW, Prins JB, Baker MK, Singh MA, Coombes JS.
Exercise prescription for patients with Type 2 diabetes and pre-diabetes: a
position statement from Exercise and Sport Science Australia. J Sci Med
Sport. 2012;15(1):25–31. http://dx.doi.org/10.1016/j.jsams.2011.04.005.
22. Stevens JW, Khunti K, Harvey R, et al. Preventing the progression to
Type 2 diabetes mellitus in adults at high risk: a systematic review and
network meta-analysis of lifestyle, pharmacological and surgical
interventions. Diabetes Res Clin Pract. 2015;107(3):320–331. http:
//dx.doi.org/10.1016/j.diabres.2015.01.027.
23. Mourier A, Gautier JF, De Kerviler E, et al. Mobilization of visceral
adipose tissue related to the improvement in insulin sensitivity in
response to physical training in NIDDM: effects of branched-chain
amino acid supplements. Diabetes Care. 1997;20(3):385–391. http://dx.
doi.org/10.2337/diacare.20.3.385.
24. Giannopoulou I, Ploutz-Snyder LL, Carhart R, et al. Exercise is
required for visceral fat loss in postmenopausal women with Type 2
diabetes. J Clin Endocrinol Metab. 2005;90(3):1511–1518. http://dx.doi.
org/10.1210/jc.2004-1782.
25. Colberg SR, Albright AL, Blissmer BJ, et al. Exercise and Type 2
diabetes: American College of Sports Medicine and the American
Diabetes Association: joint position statement. Exercise and Type 2
diabetes. Med Sci Sports Exerc. 2010;42(12):2282–2303. http://dx.doi.
org/10.1249/MSS.0b013e3181eeb61c.
26. Diabetes Prevention Program Research Group. Reduction in the
incidence of Type 2 diabetes with lifestyle intervention or metformin.
N Engl J Med. 2002;346:393–403. http://dx.doi.org/10.1056/NEJMoa
012512.
27. Lindström J, Peltonen M, Eriksson JG, et al. Improved lifestyle and
decreased diabetes risk over 13 years: long-term follow-up of the
randomised Finnish Diabetes Prevention Study (DPS). Diabetologia.
2013;56(2):284–293. http://dx.doi.org/10.1007/s00125-012-2752-5.
28. Larsen BA, Allison MA, Kang E, et al. Associations of physical activity and
sedentary behavior with regional fat deposition. Med Sci Sports Exerc.
2014;46(3):520. http://dx.doi.org/10.1249/MSS.0b013e3182a77220.
29. Boulé NG, Haddad E, Kenny GP, Wells GA, Sigal RJ. Effects of exercise
on glycemic control and body mass in Type 2 diabetes mellitus:
a meta-analysis of controlled clinical trials. JAMA. 2001;286(10):
1218–1227. http://dx.doi.org/10.1001/jama.286.10.1218.
30. Sigal RJ, Kenny GP, Wasserman DH, Castaneda-Sceppa C, White RD.
Physical activity/exercise and Type 2 diabetes: a consensus statement
from the American Diabetes Association. Diabetes Care. 2006;29
(6):1433–1438. http://dx.doi.org/10.2337/dc06-9910.
31. Poirier P, Després JP. Exercise in weight management of obesity.
Cardiol Clin. 2001;19(3):459–470. http://dx.doi.org/10.1016/
S0733-8651(05)70229-0.
32. Stiegler P, Cunliffe A. The role of diet and exercise for the maintenance of
fat-free mass and resting metabolic rate during weight loss. Sports Med.
2006;36(3):239–262. http://dx.doi.org/10.2165/00007256-200636030-00005.
33. Nomura T, Ikeda Y, Nakao S, et al. Muscle strength is a marker of
insulin resistance in patients with Type 2 diabetes: a pilot study.
Endocr J 2007;54(5):791–796. http://dx.doi.org/10.1507/endocrj.K07-055.
34. Sigal RJ, Kenny GP, Boulé NG, et al. Effects of aerobic training,
resistance training, or both on glycemic control in Type 2 diabetes:
A randomized trial. Ann Intern Med. 2007;147(6):357–369. http://dx.
doi.org/10.7326/0003-4819-147-6-200709180-00005.
35. Boulé NG, Haddad E, Kenny GP, Wells GA, Sigal RJ. Effects of exercise
on glycemic control and body mass in Type 2 diabetes mellitus: a meta-
analysis of controlled clinical trials. JAMA. 2001;286(10):1218–1227.
http://dx.doi.org/10.1001/jama.286.10.1218.
36. Knowler WC, Barrett-Connor E, Fowler SE, et al. Diabetes Prevention
Program Research Group. Reduction in the incidence of Type 2
diabetes with lifestyle intervention or metformin. N Engl J Med.
2002;346(6):393–403. http://dx.doi.org/10.1056/NEJMoa012512.
37. Dube JJ, Amati F, Stefanovic-Racic M, Toledo FG, Sauers SE, Good-
paster BH. Exercise-induced alterations in intramyocellular lipids and
insulin resistance: the athlete’s paradox revisited. Am J Physiol
Endocrinol Metab. 2008;294(5):E882–E888. http://dx.doi.org/10.1152/
ajpendo.00769.2007.
38. Oshida Y, Yamanouchi K, Hayamizu S, Sato Y. Long-term mild
jogging increases insulin action despite no influence on body mass
index or VO2 max. J Appl Physiol. 1989;66:2206–2210.
39. Duncan GE, Perri MG, Theriaque DW, Hutson AD, Eckel RH,
Stacpoole PW. Exercise training, without weight loss, increases insulin
sensitivity and postheparin plasma lipase activity in previously
sedentary adults. Diabetes Care. 2003;26(3):557–562. http://dx.doi.
org/10.2337/diacare.26.3.557.
40. Bassuk SS, Manson JE. Epidemiological evidence for the role of
physical activity in reducing risk of Type 2 diabetes and cardiovascular
disease. J Appl Physiol. 2005;99(3):1193–1204. http://dx.doi.org/
10.1152/japplphysiol.00160.2005.
www.ajpmonline.org