Work,physical activity,and metabolic health: Understanding insulin sensitivity of long-haul truck drivers

Abstract

BACKGROUND:

Long-haul truck drivers are disproportionately exposed to metabolic risk; however, little is known about their metabolic health and the role of physical activity and other risk factors in metabolic outcomes.

OBJECTIVE:

This study compares truck drivers’ insulin sensitivity, and associations between metabolic risk factors and insulin sensitivity, with those of the general population.

METHODS:

Survey, anthropometric, and biometric data were collected from 115 long-haul truckers, which were then compared to the general population data using the National Health and Nutrition Examination Survey (NHANES) dataset. The quantitative insulin sensitivity check index (QUICKI) was used to estimate insulin sensitivity.

RESULTS:

Truck drivers had lower QUICKI scores than the general population cohort. Sagittal abdominal diameter and exercise were predictive for QUICKI among combined cohorts. Waist circumference and perceived health were more predictive for QUICKI among truck drivers, and sagittal abdominal diameter and income were more predictive for QUICKI among the general population.

CONCLUSIONS:

Long-haul truckers appear to represent a subset of the general population regarding the impact of physical activity and other metabolic risk factors on QUICKI. Accordingly, comprehensive efforts which target these factors are needed to improve truckers’ physical activity levels and other metabolic risks.

Keywords

Diabetes exercise commercial transport abdominal obesity

1 Introduction

Type 2 diabetes mellitus (T2D) has risen over several decades and shows little sign of slowing [1 –3]. Currently, 29.1 million Americans have been diagnosed with T2D, and approximately 79 million Americans are living with pre-diabetes, 70%of whom will eventually develop full-blown T2D [1 –4]. Individuals with diabetes have double the mortality risk, and many leading causes of death are related to the increased prevalence of obesity and associated metabolic dysfunctions, including the loss of insulin sensitivity [2 , 5].

Insulin sensitivity—the amount of insulin required to deposit an amount of glucose into tissues—is a component of metabolic health. The progressive loss of insulin sensitivity (i.e., the development of insulin resistance) is a hallmark of T2D leading to hyperinsulinemia and eventually pancreatic β cell failure [6]. Overt T2D is only one condition associated with diminished insulin sensitivity, as impaired insulin sensitivity itself is a component of metabolic syndrome and, in conjunction with impaired glucose tolerance, can lead to increased cardiovascular risk [5, 7]. Therefore, assessing and managing insulin sensitivity is important for health promotion as well as disease prevention.

Sedentariness, poor diet, and associated adiposity are among key factors which reduce insulin sensitivity. Insulin sensitivity, along with post-prandial glucose tolerance and fasting blood glucose concentrations, improve with increased physical activity [8 –10]. Previously-exercised muscles have increased responsiveness to insulin stimulated glucose uptake, leading to lower post-prandial glycemic excursions, even among insulin resistant populations [9, 11]. Active lifestyle, including regular physical activity, improves HbA1c over time, reducing the risk of diabetic complications and the eventual development of the disease [12, 13]. Likewise, increasing sedentariness leads to increased adiposity and reduced insulin sensitivity [8, 14]. Sedentariness independently red-uces insulin sensitivity, even among individuals who regularly exercise [15, 16]. Due to this relationship between physical activity and insulin sensitivity, populations exhibiting high levels of sedentariness are at risk of reduced insulin sensitivity [17, 18]. Diets high in fat and sugar have also been shown to contribute to increased adiposity and reduced insulin sensitivity [19, 20]. Together, sedentariness and poor diet interact to increase adiposity, pro-inflammatory cytokines, and reduce insulin sensitivity in several tissues, leading to increased risk of metabolic disease and dysfunction [7 , 21–23].

Truck drivers experience disproportionate meta-bolic risks compared to other occupational segments. Low levels of physical activity are inherent to commercial driving as: 1) job duties require sitting for many hours at a time behind the wheel; 2) long and irregular work hours frequently disrupt regular functions of the body (e.g., sleep) and health-sup-portive routines (e.g., engaging in physical activity); and 3) workplace built environments (e.g., truckstops) present barriers to physical activity and healthy eating [24 –26]. High levels of work-induced sedentariness and low levels of physical activity are exacerbated by unhealthful dietary patterns, and thus elevate adiposity—both general and abdominal—is endemic in this population [25 , 28]. Besides sedentariness and limited access to healthful foods, characteristics of truck driving that independently exacerbate risk for reduced insulin sensitivity include long work hours, frequent shiftwork, fatigue, job strain, chronically elevated stress, and low access to health-promotive resources [24, 29]. This unique constellation of metabolic risks enhances the likelihood of truck drivers developing multiple chronic conditions associated with reduced insulin sensitivity, including T2D and cardiovascular disease [30, 31].

Despite elevated metabolic risks endemic to truck driving, as well as the dire implications of reduced insulin sensitivity for truck drivers, other motorists, and pedestrians, little is known about truck drivers’ metabolic risk profile, including the most salient risk factors for negative metabolic outcomes and the role of physical activity in these outcomes. Therefore, this study has two objectives: 1) to determine how the fasting insulin sensitivity of long-haul truck drivers compares to that of the general population; and 2) to discern those risk factors influencing insulin sensitivity among truck drivers and the general population, with an emphasis on the role of physical activity. It was hypothesized that: (a) truck drivers have impaired insulin sensitivity due to their occupational milieu, which is marked by sedentariness; and (b) truck drivers share the same metabolic risk factors that affect insulin sensitivity with the general population, including physical activity.

2 Materials and methods

The study procedures and cohort details have been described previously [27, 31]. Briefly, long-haul truck drivers were recruited from a TravelCenters of America truckstop over a 6-month period. Researchers approached truck drivers between 18:00–23 : 00 and asked screening questions to determine eligibility for the study, which ensured they were long-haul drivers (rather than local or regional drivers) and that they had an overnight layover at the truckstop. Approximately half of all drivers approached met inclusion criteria, who completed a questionnaire about their health status, work, and health behaviors. Upon the completion of the interviewer-administered surveys and collection of anthropometric data, drivers were given an appointment card for the following morning (between 04:00–06:00) and instructed to fast for the remainder of the evening. Procedures were approved by the Institutional Review Board of the University of North Carolina at Greensboro. Of the 262 drivers who participated in the first portion of the study, 115 returned for morning blood draws.

2.1 Survey data

Demographic, behavioral, and health status data were collected using a comprehensive survey instrument. The survey was developed from insights glea-ned from other relevant instruments and literature as well as previous work with truck drivers [32 –35]. Key self-reported variables for this study included age, education, annual income, perceived health, screen time (computer and TV), and time spent exercising.

2.2 Anthropometric data

Height, weight, waist circumference (WC), and sagittal abdominal diameter (SAD) were measured. Weight was measured using an Elite XXL scale, height was measured using a portable stadiometer (Seca, Chino, CA, USA), and body mass index (BMI) was calculated from these measurements (kg/m²). WC was measured using a Gulick II tape measure and was assessed at the natural waist. SAD was measured using a Rosscraft Campbell caliper #20 (Vancouver, BC), rounding values to the nearest tenth of a centimeter. WC and SAD measures were taken twice and then averaged. All anthropometric data were rounded to the nearest tenth.

2.3 Blood draws and analysis

Following Occupational & Safety Health Administration (OSHA) regulations [United States Department of Labor, 2003], blood was taken from the antecubital space by a trained phlebotomist. Blood was collected in serum separator tubes and allowed to clot at room temperature for 15–20 minutes before being centrifuged at 3000 rpm for 15 minutes, using a portable centrifuge (LW Scientific E8 Portafuge, Atlanta, GA, USA), and then transported on ice to the laboratory. Serum was divided into aliquots and stored at –80°C for future analysis.

All serum samples were run in duplicate in the university’s exercise physiology lab. Mercodia insulin enzyme linked immunosorbent assay (ELISA) kits (Mercodia; Uppsala, Sweden) were used for the measurement of insulin, and Cayman glucose colorimetric assay kits (Cayman Chemicals; Ann Arbor, Michigan) were used for measurement of glucose. An Epoch plate reader (BioTek, Winooski, VT, USA) was used to read all assays. The average values of the duplicate samples were used in analyses, and only accepted with a coefficient of variance less than 15%between samples.

2.4 Insulin sensitivity

Although the hyperinsulinemic euglycemic clamp technique is the optimal method for assessing insulin sensitivity, its implementation is time and labor intensive [36]. As a pragmatic alternative, an individual’s insulin sensitivity can be estimated during fasting conditions using the quantitative insulin sensitivity check index (QUICKI), which is calculated from fasting insulin and glucose concentrations [36, 37]. QUICKI has been shown to correspond well to values obtained using a clamp technique [36, 37]. QUICKI values were calculated from the fasting insulin and glucose concentrations for the truck drivers.

While a total of 115 truck drivers participated in the blood-draw portion of the study, only 104 samples were used for the analysis following the removal of outliers—defined as individuals with insulin or glucose values greater than 2 standard deviations above or below the mean of the group (with outliers included in mean). All participants from either cohort who were on any glucose-lowering medications were included, accounting for 13.46%of truck drivers and 1.17%of the National Health and Nutrition Examination Survey (NHANES) cohort. All data on medications were self-reported.

2.5 NHANES data for comparison to general population

NHANES 2011-2012 data were obtained from the Centers for Disease Control and Prevention (CDC) website [38]. The data were filtered to match the truck drivers in gender and age. To match truck drivers in our sample, only male NHANES respondents between the ages of 25 and 66 were included. Ethnicity/race and BMI were not filtered, due to their relationship to both insulin sensitivity and the profiles of truck drivers. NHANES participants on glucose-lowering medications were assessed through self-report. Outliers were only removed from the NHANES cohort if there were also outliers for the same variable in the truck driver group. This was because the NHANES data are generally large and contain fewer outliers that would substantially skew the results of analyses.

2.6 Statistical analyses

A χ² test was run for differences in race/ethnicity between cohorts, due to the fact that some minority groups are at higher risk of reduced insulin sensiti-vity and are differently represented within the truck driver cohort compared to NHANES participants [39]. An initial one-way analysis of variance (ANOVA) was run between truck drivers and matched NHANES 2011-2012 respondents for mean difference in QUICKI. Initial bivariate correlations showed that most of the environmental and behavioral variables were not significantly correlated directly with QUICKI; therefore, these variables were not used in the regression models. Any variable with a significant or near significant bivariate correlation, or precedent of correlation with insulin sensitivity, was entered into the stepwise regression. These included the following: age, BMI, WC, SAD, education, income, perceived health, screen time (computer and TV), and exercise.

An exploratory stepwise regression was run reg-ressing QUICKI on these variables among truck driver and NHANES samples. Independent variables included in the stepwise regressions included: BMI, perceived health (1 = worst to 5 = best), WC, SAD, level of education, level of income, time spent watching television, time spent using computers, and time spent exercising. Criterion for entry into the model was probability of F < 0.05, and criterion for removal was probability of F > 0.1. A follow-up two-way ANCOVA was run to determine whether there was an interaction between group membership (truck drivers versus NHANES respondents) and a common variable for QUICKI, which was seen in the initial step-wise regression, and using ethnicity and SAD (the other significant independent variables from the regression) as covariates. This analysis was com-pleted to determine whether there are differences between truck drivers and the general population given the significant variables that were revealed from the initial regression, as well as to determine whether truck drivers showed the same relationship between QUICKI and exercise as the general population.

Individual stepwise multiple regressions were run within each group for QUICKI using the same initial set of independent variables. Finally, a one-way ANOVA was run for means of WC between groups, and, because of heterogeneity of variance, non-parametric tests (Mann-Whitney U) were run for perceived health, SAD, amount of exercise, income category, and blood glucose and insulin concentrations.

3 Results

Findings related to race/ethnicity, diabetes, BMI category, insulin, and glucose for the NHANES res-pondents and truck drivers can be found in Table 1. The mean age was 46.3±0.4 years for NHANES respondents and 47.6±0.9 years for truck drivers (p > 0.05), as reported in previous manuscripts [27, 31]. BMI for NHANES respondents was 27.84±0.12 and 32.98±0.55 for truck drivers, representing a significant difference (Mann-Whitney U = 40896.0, p < 0.01). There was a significant difference in the racial/ethnic distribution of the truck drivers and NHANES cohorts (χ² = 41.952, p < 0.01) as shown in Table 1. The mean insulin concentration was higher among truck drivers compared to NHANES participants (19.45 mU/l vs. 13.08 mU/l, respectively, Mann-Whitney U = 54984.5, p < 0.01), but mean glucose concentrations were lower among the truck drivers compared to NHANES participants (92.1mg/dl vs. 102.7 mg/dl, respectively, Mann Whitney U = 48580.0, p < 0.01). This corresponded to a significant difference (F = 14.62, p < 0.01) in mean QUICKI values between truck drivers and NHANES participants, as seen in Fig. 1.

Table 1
Race/ethnicity, diabetes, BMI category, insulin, and glucose for the NHANES respondents and long-haul truck drivers. Values are either percentages or the mean±SEM

NHANES Truck Drivers

Ethnicity White 37.36% 65.38%

Black 24.51% 25.00%

Hispanic 22.09% 7.69%

Other 16.04% 1.92%

Diabetes (doctor diagnosed) 8.88% 16.35%

Medicated (for glucose) 1.17% 13.46%

BMI Category Normal weight 29.73% 6.73%

Overweight 40.41% 25.96%

Obese 29.86% 67.31%

Insulin (mU/L) 13.08±0.21 19.45±1.39

Glucose (mg/dL) 102.7±0.39 92.1±2.62

		NHANES	Truck Drivers
Ethnicity	White	37.36%	65.38%
Black	24.51%	25.00%
Hispanic	22.09%	7.69%
Other	16.04%	1.92%
Diabetes	(doctor diagnosed)	8.88%	16.35%
Medicated	(for glucose)	1.17%	13.46%
BMI Category	Normal weight	29.73%	6.73%
	Overweight	40.41%	25.96%
	Obese	29.86%	67.31%
Insulin (mU/L)		13.08±0.21	19.45±1.39
Glucose (mg/dL)		102.7±0.39	92.1±2.62

Fig. 1

The mean difference in QUICKI between NHANES and long-haul truck drivers. Values are mean±SD. ^*Significant difference at P < 0.05.

The rates of physician-diagnosed T2D were nearly twice as high among truck drivers compared to NHANES (16.35 vs. 8.88%, respectively), and 13.46%of the drivers were medicated for elevated glucose levels, compared to only 1.17%of NHANES participants. Additionally, truck drivers were less active on average than NHANES participants (Mann Whitney U = 714737.5, p = 0.01); however, the variability of physical activity among the two cohorts was quite different. Among truck drivers, 38.5%reported no physical activity and only 1%reported greater than 3 hours of weekly physical activity. In contrast, from NHANES, 43.6%reported getting no physical activity and 19.7%getting more than 3 hours of physical activity per week. There was no significant difference in physical activity reported by truck drivers who gave blood samples versus those who only completed the survey (F = 0.003, p = 0.954), suggesting that drivers who gave blood samples are representative of the larger population of truck drivers.

The significant variables from the initial stepwise regression for combined NHANES and truck drivers were SAD (β= –0.32, p < 0.01) and exercise per week (β= 0.05, p = 0.032), as seen in Table 2, with an overall adjusted R² of 0.103 (p < 0.01) for QUICKI. All other independent variables were excluded from the stepwise regression (as discussed in Methods). An initial analysis of covariance (ANCOVA), adjusting for BMI, SAD, and exercise category, indicated that there was a small but significant difference in QUICKI between truck drivers (QUICKI = 0.32±0.003) and NHANES respondents (QUICKI = 0.33±0.001) (F = 12.271, p < 0.01). The ANCOVA confirmed the significance of group membership for the dependent variable QUICKI (F = 7.839, p < 0.01), but showed that there was not a significant difference for exercise duration when accounting for racial/ethnic groups (F = 17.286, p < 0.01) and SAD (F = 203.01, p < 0.01). Additionally, there was no interaction between exercise duration and group membership (F = 1.042, p = 0.37).

Table 2

Predictive capacity of independent variables on insulin sensitivity (QUICKI) for combined groups, NHANES only and long-haul truck drivers only

Variable	Unstandardized	Std. Error	Standardized	P value
Combined model including NHANES and LHTD
Constant	0.382	0.004		< 0.01
SAD	–0.002	0.000	–0.318	< 0.01
Exercise	0.001	0.001	0.049	0.03
Model including only NHANES
Constant	0.392	0.005		< 0.01
SAD	–0.003	0.000	–0.312	< 0.01
Income	–0.003	0.001	–0.050	0.040
Exercise	0.001	0.001	0.048	0.048
Model including only LHTD
Constant	0.391	0.026		< 0.01
Waist Circumference	–0.001	0.000	–0.379	< 0.01
Perceived Health	0.007	0.003	0.214	0.021

Significant variables in the stepwise regression model specific for the NHANES respondents were SAD, income category, and amount of exercise, as seen in Table 2. However, the significant variables in the stepwise regression model for truck drivers were WC and perceived health. These regressions yielded higher R² for truck drivers compared to NHANES respondents (R² = 0.22 versus R² = 0.10, respectively). All significant independent variables within groups were also significantly different between groups, as seen in Table 3. However, BMI, level of education, time spent on the computer, and time spent watching TV were not significantly related to QUICKI in any stepwise regressions.

Table 3

Values of independent variables used in regression analyses for NHANES respondents and Long-haul truck drivers. Values are mean±SEM

Variable	NHANES (mean±SEM)	Truck Drivers (mean±SEM)	Significance
BMI (kg/m²)	27.84±0.12	32.98±0.55	p < 0.01
Waist Circumference (cm)	98.66±0.32	114.11±1.33	p < 0.01
Perceived Health	3.38±0.03	2.94±0.90	p < 0.01
SAD (cm)	22.81±0.09	32.07±0.48	p < 0.01
Income Category	0.63±0.02	1.35±0.07	p < 0.01
Exercise Category	1.19±0.03	0.77±0.07	p < 0.01

4 Discussion

Long-haul truck drivers had lower insulin sensitivity, even when accounting for BMI, SAD, and amount of exercise, despite having lower fasting blood-glucose concentrations on average, compared to the general population, affirming recent CDC-funded studies showing increased rates of T2D among truck drivers [28]. However, this disparity may underestimate the true degree of metabolic risk experienced by truck drivers compared to the general population. The lower levels of insulin sensitivity among the truck drivers were despite a higher percentage of truck drivers being medicated to control blood glucose concentrations compared to the NHANES participants. Also, within the general population, some individuals with T2D may have been attempting to use lifestyle modification to control blood glucose concentrations, whereas lifestyle modification would be particularly difficult for truck drivers given their occupational milieu [24]. Finally, truck drivers are subject to occupational regulations, including health screenings every two years, increasing the likelihood that truck drivers with T2D are medicated to maintain their job eligibility [40]. Because of these regulations, truck drivers who are unable to successfully control their blood glucose levels eventually leave the occupation, thus creating a ceiling effect regarding insulin sensitivity outcomes in this population [41].

Factors which most strongly influenced insulin sensitivity among truck drivers were different from those for the general population with greater variability among truck drivers. In particular, the amount of exercise was a significant predictor of fasting insulin sensitivity in the general population, but not for truck drivers. However, there was no interaction when examining the effects of exercise on insulin sensitivity between the general population and truck drivers, indicating that regardless of group, exercise had similar effects on insulin sensitivity. This suggests that truck drivers do not respond differently to exercise compared to the general population. Thus, truck drivers may have similar improvements in insulin sensitivity in response to exercise [8 –13] and are a unique subset of the general population of males between the ages of 25 and 66 due to their disproportionate exposure to metabolic risk factors. Interestingly, income category was associated with insulin resistance for the general population but not truck drivers. This may be due to a greater diversity in access to health-supportive resources among the general population when compared to truck drivers. Individuals in the general population with higher incomes likely have better access to health-promotive resources such as health care, fitness resources, and healthy food [42, 43]. However, truck drivers with higher incomes are still unlikely to overcome pervasive barriers which prevent access to these health-promotive resources; in particular, trucker worksites rarely offer these resources, regardless of affordability [25].

Similar to other studies, measures of regional body fat distribution (WC or SAD) were indicators of insulin sensitivity in both cohorts [44, 45]. WC and SAD were both predictors, but only one was maintained in each regression, as they have substantial overlap. Thus, either anthropometric measure has similar predictive potential of insulin sensitivity for both groups, and measures like SAD and WC may be the easiest methods for estimating metabolic health among truck drivers as well as the general population. This is especially important for the former, as their work organization is marked by constant mobility [24] and may make it difficult for companies or other stakeholders to monitor drivers’ health. Measuring SAD and WC may provide valuable and easily-obtained markers for monitoring the cardiometabolic health of this population.

The significant contribution from perceived health among truck drivers may be due to a general lack of health care among truck drivers [30]. This lack of health care may be a reason that truck drivers had significantly lower perceived health problems compared to the general population. Ironically, there was less variability in responses of perceived health among the general population compared to truck drivers, allowing more of the variability in truck drivers to account for differences in insulin resistance. This variability in perceived health among truck drivers is somewhat surprising given that federal policies require a health screening for truck drivers every two years, ensuring that the least healthy individuals are removed from the profession [46]. It is difficult to know what factors truck drivers consider when assessing their own health, and it is possible that other psychosocial factors –especially job stress [24] –are pervasive factors in their perceived health.

Together, these findings indicate potential directions for improving metabolic health outcomes among truck drivers. Physical activity appears to hold similar value for non-pharmacological improvements in insulin sensitivity for truck drivers as well as the general population, but truck drivers have much more room for improvement. Interventions for truck drivers which target physical activity must take into account the multitude of other metabolic risk factors, such as limited access to affordable healthful foods (not included in this study), and work organization factors, which may influence physical activity efforts. Additionally, as indicated in our findings regarding the influence of income on insulin resistance, the unique worksite environments which truck drivers encounter constitute barriers which make it difficult for individually-focused interventions to be effective [25, 26]. While physical activity should constitute key components of interventions to improve metabolic health for truck drivers, the highest-leverage opportunities for increasing physical activity should be identified and implemented through federal and corporate policies and built environment changes to maximize the efficacy of these efforts. Thus, corresponding interventions should be comprehensive, accounting for this unique milieu of metabolic risks [24 , 47]. Along with the current need for interventions that target metabolic health outcomes among truck drivers, it is critical that extant and future initiatives are rigorously evaluated to improve their efficacy in mitigating metabolic risks, as well as to encourage and facilitate their proliferation across multiple commercial driving contexts.

There are three primary limitations of this study that have been partially addressed in other papers [27, 31]. First, the small sample size for truck drivers willing to give blood samples, along with the comparatively larger number of participants included in the NHANES dataset, could have influenced our findings. To counter this limitation, additional analyses were run for each group separately, indicating the predictors unique to long-haul truck drivers. Second, truck drivers may have self-selected to participate in this study for reasons which are unknown but may have influenced our results due to a biased sample. This is unlikely though, as several of our findings were in accordance with previous studies. Finally, because survey data are reliant on self-reporting, any questionnaire measures may have been skewed; however, because both datasets used in this study relied on survey data, these differences did not necessarily skew comparisons between the samples.

In conclusion, truck drivers have reduced insulin sensitivity, and therefore have greater metabolic health risk, compared to the general population. Differences in the predictive value of variables included in this study–especially physical activity–were likely due to the excessive metabolic health risks endemic to long-haul trucking, including long hours, high stress, poor access to healthful foods, and lack of physical activity. However, truck drivers do not appear to be uniquely affected by these risks; instead, they are disproportionately exposed to these conditions, explaining their reduced insulin sensitivity. Because truck drivers represent a unique subset of the general population, non-pharmacological interventions which are well established in the general population, especially emphasizing physical activity, should be implemented. However, rather than being individually-focused, these efforts should constitute comprehensive interventions, focusing on policies and built environment changes to account for the milieu of metabolic risks faced by truck drivers.

Footnotes

Acknowledgments

We would like to thank Mr. Tom Liutkus, Vice President of Marketing and Public Relations for Travel Centers of America (TA) and Mr. Jerald Brisson, General Manager of the Whitsett, NC TA truck stop and his staff for their instrumental support in data collection. We also thank the long-haul truckers who participated in this study and extend our thanks to our graduate student Kiki Hatzudis for her invaluable assistance in various phases of data collection.

Conflict of interest

None to report.

Funding

This study was funded by an award through the University of North Carolina, Greensboro.

References

Selvin

, Parrinello

, Sacks

, Coresh

. Trends in prevalence and control of diabetes in the United States, 1988-1994 and 1999-2010. Annals of Internal Medicine. 2014;160:517–25.

Centers for Disease Control and Prevention. Diabetes report card 2014. Atlanta, GA: US Department of Health and Human Services; 2015.

Centers for Disease Control and Prevention. National diabetes statistics report: Estimates of diabetes and its burden in the United States 2014. Atlanta, GA: US Department of Health and Human Services; 2014.

Nathan

, Davidson

, DeFronzo

, Heine

, Henry

, Pratley

, et al. Impaired fasting glucose and impaired glucose tolerance: Implications for care. Diabetes Care. 2007;30:753–9.

Cavalot

, Pagliarino

, Valle

, Di Martino

, Bonomo

, Massucco

, et al. Postprandial blood glucose predicts cardiovascular events and all-cause mortality in type 2 diabetes in a 14-year follow-up: lessons from the San Luigi Gonzaga Diabetes Study. Diabetes care. 2011;34:2237–43.

Tahrani

, Bailey

, Del Prato

, Barnett

. Management of type 2 diabetes: new and future developments in treatment. Lancet. 2011;378:182–97.

Kaur

. A comprehensive review on metabolic syndrome. Cardiol Res Pract. 2014;2014:943162.

Mikus

, Oberlin

, Libla

, Taylor

, Booth

, Thyfault

. Lowering physical activity impairs glycemic control in healthy volunteers. Medicine and science in sports and exercise. 2012;44:225–31.

Mikus

, Oberlin

, Libla

, Boyle

, Thyfault

. Glycaemic control is improved by 7 days of aerobic exercise training in patients with type 2 diabetes. Diabetologia. 2012;55:1417–23.

10.

Oberlin

, Mikus

, Kearney

, Hinton

, Manrique

, Leidy

, et al. One bout of exercise alters free-living postprandial glycemia in type 2 diabetes. Medicine and Science in Sports and Exercise. 2014;46:232–8.

11.

Richter

, Kiens

, Mizuno

, Strange

. Insulin action in human thighs after one-legged immobilization. J Appl Physiol. 1989;67:19–23.

12.

Conn

, Koopman

, Ruppar

, Phillips

, Mehr

, Hafdahl

. Insulin Sensitivity Following Exercise Interventions: Systematic Review and Meta-Analysis of Outcomes Among Healthy Adults. J Prim Care Community Health. 2014;5:211–22.

13.

Snowling

, Hopkins

. Effects of different modes of exercise training on glucose control and risk factors for complications in type 2 diabetic patients: a meta-analysis. Diabetes Care. 2006;29:2518–27.

14.

Krogh-Madsen

, Thyfault

, Broholm

, Mortensen

, Olsen

, Mounier

, et al. A 2-wk reduction of ambulatory activity attenuates peripheral insulin sensitivity. J Appl Physiol. 2010;108:1034–40.

15.

Duvivier

, Schaper

, Bremers

, van Crombrugge

, Menheere

, Kars

, et al. Minimal intensity physical activity (standing and walking) of longer duration improves insulin action and plasma lipids more than shorter periods of moderate to vigorous exercise (cycling) in sedentary subjects when energy expenditure is comparable. PloS one. 2013;8:e55542.

16.

Healy

, Wijndaele

, Dunstan

, Shaw

, Salmon

, Zimmet

, et al. Objectively measured sedentary time, physical activity, and metabolic risk: the Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Diabetes Care. 2008;31:369–71.

17.

Colberg

, Albright

, Blissmer

, Braun

, Chasan-Taber

, Fernhall

, 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. Medicine and Science in Sports and Exercise. 2010;42:2282–303.

18.

Hawley

, Lessard

. Exercise training-induced improvements in insulin action. Acta Physiol (Oxf). 2008;192:127–35.

19.

Riccardi

, Giacco

, Rivellese

. Dietary fat, insulin sensitivity and the metabolic syndrome. Clin Nutr. 2004;23:447–56.

20.

Cordain

, Eaton

, Sebastian

, Mann

, Lindeberg

, Watkins

, et al. Origins and evolution of the Western diet: health implications for the 21st century. Am J Clin Nutr. 2005;81:341–54.

21.

Yudkin

, Stehouwer

, Emeis

, Coppack

. C-reactive protein in healthy subjects: associations with obesity, insulin resistance, and endothelial dysfunction: a potential role for cytokines originating from adipose tissue? Arteriosclerosis, Thrombosis, and Vascular Biology. 1999;19:972–8.

22.

Kahn

, Hull

, Utzschneider

. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature. 2006;444:840–6.

23.

Nazare

, Smith

, Borel

, Almeras

, Tremblay

, Bergeron

, et al. Changes in both global diet quality and physical activity level synergistically reduce visceral adiposity in men with features of metabolic syndrome. J Nutr. 2013;143:1074–83.

24.

Apostolopoulos

, Lemke

, Sonmez

. Risks endemic to long-haul trucking in North America: Strategies to protect and promote driver well-being. New Solut. 2014;24:57–81.

25.

Apostolopoulos

, Lemke

, Sönmez

, Hege

. The obesogenic environment of commerical trucking: A worksite environmental audit and implications for systems-based interventions. American Journal of Health Education. 2016;47:85–93.

26.

Lemke

, Meissen

, Apostolopoulos

. Overcoming barriers in unhealthy settings: A phenomenological study of healthy truck drivers. Global Qualitative Nursing Research. 2016;3:1–9.

27.

Wideman

, Oberlin

, Sönmez

, Labban

, Lemke

, Apostolopoulos

. Obesity indices are predictive of elevated C-reactive protein in long-haul truck drivers. Am J Ind Med. 2016;59:665–75.

28.

Sieber

, Robinson

, Birdsey

, Chen

, Hitchcock

, Lincoln

, et al. Obesity and other risk factors: The National Survey of US Long-Haul Truck Driver Health and Injury. Am J Ind Med. 2014;57:615–26.

29.

Birdsey

, Sieber

, Chen

, Hitchcock

, Lincoln

, Nakata

, et al. National survey of US long-haul truck driver health and injury: Health behaviors. J Occup Environ Med. 2015;57:210–6.

30.

Apostolopoulos

, Sonmez

, Shattell

, Gonzales

, Fehrenbacher

. Health survey of U.S. long-haul truck drivers: Work environment, physical health, and healthcare access. Work. 2013;46:113–23.

31.

Apostolopoulos

, Lemke

, Hege

, Sönmez

, Sang

, Oberlin

, et al. Work and chronic disease: Comparison of cardiometabolic risk markers between truck drivers and the general U.S. population. J Occup Environ Med. 2016;58:1098–105.

32.

Netzer

, Stoohs

, Clark

, Strohl

. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern Med. 1999;131:485–91.

33.

Partinen

, Gislason

. Basic Nordic Sleep Questionnaire (BNSQ): A quantitated measure of subjective sleep complaints. J Sleep Res. 1995;4:150–5.

34.

Philip

, Åkerstedt

. Transport and industrial safety, how are they affected by sleepiness and sleep restriction? Sleep Med Rev. 2006;10:347–56.

35.

Ursin

, Baste

, Moen

. Sleep duration and sleep-related problems in different occupations in the Hordaland Health Study. Scand J Work Environ Health. 2009;35:193–202.

36.

Singh

, Saxena

. Surrogate markers of insulin resistance: A review. World J Diabetes. 2010;1:36–47.

37.

Katz

, Nambi

, Mather

, Baron

, Follmann

, Sullivan

, et al. Quantitative insulin sensitivity check index: A simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab. 2000;85:2402–10.

38.

Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey. National Center for Health Statistics: CDC; 10/30/2015.

39.

Kodama

, Tojjar

, Yamada

, Toda

, Patel

, Butte

. Ethnic differences in the relationship between insulin sensitivity and insulin response: a systematic review and meta-analysis. Diabetes Care. 2013;36:1789–96.

40.

Federal Motor Carrier Safety Administration. DOT medical exam and commercial motor vehicle certification.Washington, DC: U.S. Department of Transportation; 2015.

41.

Krueger

, Belzer

, Alvarez

, Knipling

, Husting

, Brewster

, et al. Health and wellness of commercial drivers. In: Petty A, editor. The Domain of Truck and Bus Safety Research.Washington, DC: Transportation Research Board; 2007. pp. 58–91.

42.

Gorton

, Bullen

, Mhurchu

. Environmental influences on food security in high-income countries. Nutr Rev. 2010;68:1–29.

43.

Sallis

, Slymen

, Conway

, Frank

, Saelens

, Cain

, et al. Income disparities in perceived neighborhood built and social environment attributes. Health & Place. 2011;17:1274–83.

44.

de Souza

, de Oliveira

. Sagittal abdominal diameter shows better correlation with cardiobascular risk factors than waist circumference and BMI. J Diab & Metab Dis. 2013;12:41–5.

45.

Pajunen

, Rissanen

, Laaksonen

, Heliovaara

, Reunanen

, Knekt

. Sagittal abdominal diameter as a new predictor for incident diabetes. Diabetes Care. 2013;36:283–8.

46.

Federal Motor Carrier Safety Administration. DOT medical exam and commercial motor vehicle certification.Washington, DC: U.S. Department of Transportation; 2015.

47.

Lemke

, Hege

, Perko

, Sönmez

, Apostolopoulos

. Work patterns, sleeping hours and excess weight in commercial drivers. Occup Med. 2015:kqv080.