PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
 
PLoS One. 2016; 11(6): e0157592.
Published online 2016 June 20. doi:  10.1371/journal.pone.0157592
PMCID: PMC4913922

Tobacco Use, Insulin Resistance, and Risk of Type 2 Diabetes: Results from the Multi-Ethnic Study of Atherosclerosis

Ying-Mei Feng, Editor

Abstract

Introduction

Tobacco use is associated with insulin resistance and incident diabetes. Given the racial/ethnic differences in smoking patterns and incident type 2 diabetes our objective was to evaluate the association between tobacco use and insulin resistance (IR) as well as incident type 2 diabetes mellitus in a contemporary multiethnic cohort.

Methods and Results

We studied 5,931 Multi- Ethnic Study of Atherosclerosis (MESA) participants who at baseline were free of type 2 diabetes (fasting glucose ≥7.0 mmol/l (126 mg/dl) and/or use of insulin or oral hypoglycemic medications) categorized by self-reported tobacco status and reclassified by urinary cotinine (available in 58% of participants) as never, current or former tobacco users. The association between tobacco use, IR (fasting plasma glucose, insulin, and the homeostatic model assessment of insulin resistance (HOMA-IR)) and incident diabetes over 10 years was evaluated using multivariable linear regression and Cox proportional hazards models, respectively. Mean age of the participants was 62 (±10) years, 46% were male, 41% Caucasian, 12% Chinese, 26% African American and 21% Hispanic/Latino. IR biomarkers did not significantly differ between current, former, and never cigarette users (P >0.10) but showed limited unadjusted differences for users of cigar, pipe and smokeless tobacco (All P <0.05). Fully adjusted models showed no association between dose or intensity of tobacco exposure and any index of IR. When stratified into participants that quit smoking vs. those who continued smoking during the 10-year study there was no difference in serum glucose levels or frequency of diabetes. In fully adjusted models, there was no significant difference in diabetes risk between former or current cigarette smokers compared to never smokers [HR (95% CI) 1.02 (0.77,1.37) and 0.81 (0.52,1.26) respectively].

Conclusion

In a contemporary multi-ethnic cohort, there was no independent association between tobacco use and IR or incident type 2 diabetes. The role smoking plays in causing diabetes may be more complicated than originally thought and warrants more in-depth large contemporary multi-ethnic studies.

Introduction

Tobacco smoke and tobacco products, such as cigars, pipes, and smokeless tobacco, contain many harmful and potentially harmful constituents (HPHC) [1, 2] that affect different organ systems and physiological processes in a tissue-specific manner. Smoking is reported to be a risk factor for type 2 diabetes mellitus [3, 4]. Furthermore, the most recent Surgeon General’s report concluded that smoking is a cause of type 2 diabetes and that the risk of developing type 2 diabetes increases with the cumulative number of cigarettes smoked [5] and this idea has been supported by several meta-analysis as well [6, 7]. In fact, compared with non-smokers, current smokers have been reported to have a 30–40% increased risk of type 2 diabetes [5]. Less is currently understood about other forms of tobacco relationship with type 2 diabetes, but there are some reported links to an increased risk for type 2 diabetes [8, 9]. Multiple mechanisms of tobacco exposure’s causation of diabetes have been proposed, including that tobacco exposure directly causes insulin resistance [10, 11].

The relationship between tobacco use and insulin resistance may be confounded by multiple associated variables. For example, Insulin resistance is often associated with an increase body habitus and increased BMI is a known risk factor for diabetes [12]. Smoking has also been linked to an overall lower BMI than that seen in non-smokers [13] however, smoking has be linked to increased central adiposity [11]. The increase in central adiposity may contribute to insulin resistance and diabetes proposed in smokers [11]. Moreover, there is evidence of different patterns of insulin resistance based on race [14] and gender [15]. Furthermore, smoking trends differ between race/ethnicity and genders, which may create different exposure patterns based on race/ethnicity or gender leading to different patterns of IR or incident diabetes. Of note the Hispanic/Latino populations tend to have low rates of smoking [16] while African American populations tend to have greater use of cigarettes that contain menthol [17].

Temporal trends in tobacco use may also influence the association of IR and incident diabetes. Over the decades the changing composition of cigarettes including the introduction of filters on cigarettes in the mid to late 1950’s which did not gain popularity until the 1980’s [18] and introduction of low tar cigarettes in the 1980’s [19]. Moreover, curing of tobacco and flavoring of tobacco affects the amount and type of HPHC in the product [20, 21], therefore different tobacco use patterns lead to varying exposure to these HPHC’s. There is little research into how these different HPHC’s impact IR and incident diabetes between races.

Indeed, there is a body of evidence showing no relationship between tobacco use and IR or incident diabetes. Current research demonstrates heterogeneity in the association between smoking and glycated hemoglobin (HbA1c) levels, fasting blood glucose levels and 2 hour- post-challenge glycaemia (2H-PG) [2225]. Furthermore it has been suggested the effect of smoking is not seen in lean individuals, instead only in obese men [26]. Two meta-analysis used to support the causal relationship between tobacco use and diabetes show considerable heterogeneity in the evidence used to complete the study [5, 6]. Indeed, extant meta-analyses used prior studies that often examined the association of smoking and type 2 diabetes in participants of similar race and gender, typically Caucasian males [5, 7].

In light of the heterogeneity of evidence available we aimed to identify whether the association between tobacco use IR and type 2 diabetes suggested in previous studies is also evident in large well characterized multi-ethnic cohort with multiple validated markers of insulin resistance (IR). Moreover, we sought to identify if there were markers of IR sensitive to smoking that may help identify a group of tobacco product users who are at an increased risk of type 2 diabetes. Identifying IR and incident diabetes risk in these in a contemporary multi-ethnic population seemed important as tobacco exposure patterns and metabolism of tobacco differs by race [2730], and gender, [25, 28, 31]. We therefore examined the relationship between tobacco exposure, cross-sectional insulin resistance (IR) at baseline, and prospective incident type 2 diabetes over the course of 10 years in a contemporary multi-ethnic cohort.

Methods

The Multi-Ethnic Study of Atherosclerosis (MESA) is a prospective epidemiological study of the prevalence, risk factors, and progression of subclinical CVD in a multi-ethnic cohort. The study design and methods have been previously published [32]. Briefly, 6,814 participants aged 45–84 who identified themselves as Caucasian, African-American, Hispanic/Latino, or Chinese were recruited from 6 U.S. communities (Baltimore City and Baltimore County, Maryland; Chicago, Illinois; Forsyth County, North Carolina; Los Angeles County, California; New York, New York and St. Paul, Minnesota) from 2000–2002. Participants were free of clinical CVD at enrollment. The protocols were approved by the institutional review boards of all collaborating institutions including John Hopkins University, Northwestern University, Wake Forest School of Medicine, University of California Los Angeles, Columbia University, University of Minnesota Clinical and Translational Science Institute, and by the National Heart, Lung, and Blood Institute. All participants provided written informed consent.

Exclusion Criteria

Criteria for exclusion included baseline type 2 diabetes [fasting glucose ≥7.0 mmol/l (126 mg/dl)] or previous medical diagnosis and/or use of insulin and/or oral hypoglycemic medications at baseline (n = 688, 10%). Prevalence of diabetes at baseline was 13%. Of note, there was no significant difference in the proportion of smokers excluded due to type 2 diabetes at baseline and those who were included in our study (results not shown). Participants were also excluded if they were missing tobacco exposure status (n = 35, 0.5%), duration of tobacco exposure (n = 129, 1.9%), and baseline fasting glucose or insulin levels (n = 31, 0.5%). The final sample size consisted of 5,931 participants.

Tobacco Exposure

Tobacco use included cigarettes, cigars, pipes, and smokeless tobacco defined as chewing tobacco and/or snuff. Subjects were classified as never (less than 100 lifetime cigarettes or less than 20 cigars, pipefuls, or smokeless tobacco uses), former (no use in past 30 days but more than 100 lifetime cigarettes or 20 cigars, pipefuls, or smokeless tobacco uses), or current users of tobacco products (used tobacco in last 30 days and more than 100 lifetime cigarettes or more than 20 cigars, pipefuls, or smokeless tobacco uses). Self-reported never smokers at baseline who reported being former smokers at visit 2 (n = 143) were reclassified as baseline former smokers. Baseline urinary cotinine, a product of nicotine metabolism that reflects tobacco exposure levels [33], was measured in 3,943 total participants (about half the cohort). 69 subjects (12%) who identified themselves as never or former smokers and had urinary cotinine levels greater than 500 ng/mL were reclassified as being current smokers. For former and current users, dose of exposure was defined in pack years: calculated as number of packs per day of cigarettes smoked multiplied by the number of years of smoking. Dose of pipe, cigars, and smokeless tobacco use was defined as number used per day multiplied by the number of years. Intensity of tobacco exposure was defined as the average number of cigarettes, cigars, pipes, or smokeless tobacco per day.

Outcomes

Cross-sectional IR was assessed by fasting plasma glucose levels, insulin levels, and HOMA-IR. Fasting blood glucose in serum was measured by the glucose oxidase method using the Vitros 950 analyzer (Johnson & Johnson Clinical Diagnostics, Rochester, New York). Fasting serum insulin was measured by the Linco Human Insulin Specific Radioimmunoassay kit (Linco Research, Inc., St. Charles, Missouri) [34]. HOMA-IR was calculated using the equation: glucose (mg/dL) x insulin (mU/L) / 405. Prediabetes was defined according to the American Diabetes Association as fasting glucose ≥100 & <126 mg/dL. Incident type 2 diabetes was accessed prospectively over a median 10 years of follow-up and was defined as a fasting glucose ≥7.0 mmol/l (126 mg/dl) and/or use of insulin or oral hypoglycemic medications.

Measurement of Covariates

Information on demographics, medical history, socioeconomic status, diet and physical activity was collected using questionnaires as previously described [32]. Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. Systolic and diastolic blood pressure (SBP, DBP) were measured three times using an automated sphygmomanometer and the mean of the last two measurements was used. Triglycerides, total and high-density lipoprotein cholesterol (HDL-C) were measured at a central laboratory (Fairview-University Medical Center, Minneapolis, MN) after a 12-hour fast. Low-density lipoprotein cholesterol (LDL-C) was calculated using the Friedewald equation. High-sensitivity C-reactive protein (hsCRP) was measured using a particle-enhanced immunonepholometric assay on the BNII nephelometer (Dade-Behring, Inc., Deerfield, IL) at the University of Vermont, Burlington, Vermont.

Statistical Analysis

Baseline characteristics of the population by cigarette smoking status were reported as means ± standard deviation for normally distributed variables and medians (interquartile range) for skewed variables. Differences were tested using ANOVA, equality-of-medians test and chi-squared test. For non-cigarette tobacco exposure, all analyses were restricted to those not currently using cigarettes (n = 5123) in order to minimize confounding due to concurrent cigarette smoking. Linear regression was used to assess for a possible dose-dependent relationship between dose and intensity of tobacco exposure and levels of IR biomarkers. Cox proportional hazards models were used to study the association between tobacco exposure and incident type 2 diabetes. The proportionality assumption was tested using graphic methods. Models were adjusted for age, gender, race/ethnicity, BMI, income, SBP, DBP, anti-hypertensive medication use, HDL-C, LDL-C, lipid-lowering medication use, physical activity, healthy diet, alcohol use and hsCRP.

To further explore and strengthen any potential cause-and-effect relationships between smoking, insulin resistance, and diabetes, we performed several longitudinal data analyses using multiple linear regression adjusting for the same covariates and also including change in insulin resistance and change in smoking exposure. Only serum glucose was available at each of the 5 MESA visits and therefore we could not model change in insulin or HOMA-IR. We accounted for change in smoking status using 3 approaches. In the first approach we restricted our analyses to individuals who maintained their smoking status at all MESA visits. In the second we analyzed individuals who quit smoking cigarettes at visit 2 and maintained their status as former smokers at each of the subsequent MESA visits. In the third approach, we adjusted for time-sensitive terms for quitting cigarette smoking at visits 2, 3, 4, or 5.

As sensitivity analyses, we did not adjust for hsCRP as it may mediate the association between tobacco use and IR/incident diabetes. We also studied the association between tobacco exposure status and pre-diabetes defined as fasting serum glucose between 100 and 126 mg/dL (S1 Data).

All analyses were performed using STATA (version 13) and a p-value of <0.05 was considered statistically significant (two-sided).

Results

Mean age of the participants was 62 (±10) years, 46% were male, 41% were Caucasian, 12% Chinese, 26% African American, and 21% Hispanic/Latino. Current cigarette smokers were younger compared to never and former smokers (p-values <0.001). Never smokers were more likely to be female while former smokers were more likely to be male. Compared with never smokers, current and former users were more likely to be African American (36% and 27% vs. 23%) and less likely to be Chinese (6% and 7% vs. 18%). Compared with never and former smokers, current smokers had lower SBP, less use of anti-hypertensive medications, lower HDL-C, less use of lipid lowering medications, and higher hsCRP (all p<0.05) (Table 1).

Table 1
Baseline characteristics of the population by cigarette smoking status.

IR and Type of Tobacco Exposure

Among never, current, and former cigarette smokers, there was no significant difference in crude levels of glucose, insulin, or HOMA-IR (p>0.05) (Table 1). Former cigar users had higher levels of all IR biomarkers compared with never and current users (all p<0.05) (Table 2). Current pipe users had higher levels of insulin and HOMA-IR in comparison with former and never smokers (all p<0.05) (Table 2). Current smokeless tobacco users had higher levels of all IR biomarkers compared with never and formers users (all p<0.05) (Table 2). After adjustment for age, gender, race/ethnicity, BMI, income, SBP, DBP, anti-hypertensive medication, HDL, LDL, lipid-lowering medication, and alcohol use former cigar retained higher levels of glucose and HOMA-IR, former pipe users retained higher levels of glucose, while current smokeless tobacco users retained higher levels of all IR biomarkers (all p<0.05) (Table 3).

Table 2
Duration of exposure and levels of insulin resistance biomarkers in non-cigarette tobacco.
Table 3
Adjusted§ levels of IR biomarkers by tobacco exposure status.

IR, Dose and Intensity of Tobacco Exposure

There was a significant association between dose of tobacco exposure and levels of IR biomarkers in unadjusted or partially adjusted models, but not in fully adjusted models (Table 4). Similarly, there was no significant association between intensity of tobacco exposure and levels of IR biomarkers in fully adjusted models (Table 4). There was no association between urinary cotinine levels and levels of IR biomarkers (Table 5).

Table 4
Beta coefficients (95% CI) for the cross-sectional association of dose and intensity of tobacco exposure and levels of insulin resistance biomarkers.
Table 5
Beta coefficients (95% CI) for the cross-sectional association of cotinine levels (per 10^5 ng/mL) and measures of insulin resistance biomarkers.

Incident Type 2 Diabetes and Tobacco Exposure

There was no association between tobacco exposure and baseline pre-diabetes (Table 6). In unadjusted models, former tobacco users had a significantly increased risk of incident diabetes over a median follow-up of 10.2 years compared with never smokers [Hazard Ratio (95% confidence interval) = 1.35 (1.08,1.68), 1.74 (1.13,2.69), 1.46 (1.02,2.10) and 3.38 (1.85,6.16) for cigarette, cigar, pipe, and smokeless tobacco users respectively]. However, no association was seen between tobacco use and incident diabetes after adjustment for age, gender, race/ethnicity, BMI, income, SBP, DBP, anti-hypertensive medication, HDL, LDL, lipid-lowering medication, physical activity, healthy diet, alcohol use (Table 7). Absence of significant findings made analysis of mediation by IR unwarranted.

Table 6
Odds Ratios (95%) for the association of tobacco exposure and baseline prediabetes.
Table 7
Hazard ratios (95% CI) for the association of tobacco exposure and incident diabetes.

Interactions with Race, Age, and Gender

In an exploratory analysis, several outcomes, including glucose, insulin, HOMA-IR, and incident diabetes, or exposures, including cigarettes, cigars, pipe, and smokeless tobacco, were tested for interactions with race/ethnicity, age (<60 vs. >60 years of age) and gender. There were several p-values significant in the cigar, pipe and smokeless tobacco groups, including a relationship between cigarettes, race, and HOMA-IR (Table 8). However, given the small sample size of non-cigarette users and the multiple number of comparisons performed these findings may be secondary to chance alone.

Table 8
Interaction of race/ethnicity, age, and sex with cross-sectional insulin resistance and incident diabetes mellitus.

Sensitivity analysis

In longitudinal data analyses, there was no association between dose or intensity of cigarette exposure and change in serum glucose over the course of MESA follow-up. Similarly, there was no association between dose or intensity of cigarette exposure and change in glucose from visit 1 to visit 5. There was no significant association between sustained quitters vs. sustained smokers and change in glucose levels based on dose or intensity of cigarette exposure (S1 Data). In addition, including time-varying terms for change in smoking status over time did not alter the null association between cigarette smoking status and incident diabetes (results not shown).

Discussion

Tobacco has been shown to be associated with both subclinical markers of atherosclerosis (hsCRP, interleukin 6, fibrinogen, carotid intima-media thickness, coronary artery calcification, and ankle brachial index) [35] and the incident cardiovascular events in MESA [36]. Given these findings, we sought to examine the relationship between tobacco exposure, IR and incident diabetes as a potential mediator of tobacco induced atherosclerosis and cardiovascular events in MESA. We found no consistent association between tobacco use (cigarettes, cigars, pipe, and smokeless tobacco), and multiple measures of IR (fasting glucose, insulin, HOMA-IR) or incident diabetes regardless of whether the participants were sustained quitters or sustained smokers over 10 years of follow-up in this contemporary multiethnic cohort free of diabetes at baseline. These findings are consistent with several prior studies [22, 37, 38], but discordant with other studies, including the Surgeons General’s report, that showed a positive association between tobacco use and IR [10, 11, 39] and reports which established tobacco as a potential cause of diabetes [57]. We believe the discordant results in our study may be related to several factors, including: 1) changing composition of tobacco and modes of exposure through the decades, 2) differences between subjects studied, 3) inadequate adjustment for confounding, and 4) publication bias.

Temporal trends may partly be responsible for the different results seen in our study and those that report an association between tobacco use and incident DM. Indeed about half of all the studies in the Surgeon General’s report began over a quarter century ago and eight of studies examined were from individuals studied in the 1950’s through 1970’s [5], several decades before our modern cohort. Changing composition of cigarettes including the introduction of filters on cigarettes in the mid to late 1950’s which did not gain popularity until the 1980’s [18] and introduction of low tar cigarettes in the 1980’s [19] may help explain the disparate results between our study and other older cohorts. Moreover, curing of tobacco and flavoring of tobacco affect the amount and type of HPHC in the product [20, 21], which can lead to varying types of exposure to these HPHC’s.

Our study population differs from many of the epidemiological studies used in the meta-analysis, which used single race cohorts (mostly Caucasian in the Surgeon General and Willi et al. and mostly Asian in the Pan et. al. study) or did not report ethnicity [57]. This may lead to some of the disparate results since tobacco exposure patterns and metabolism of tobacco differs by race [20, 21, 27, 28], and gender [25, 28, 31]. Differences in smoking patterns include low rates of smoking among Hispanic/Latinos [16, 40] and greater use of cigarettes that contain menthol in African Americans [17]. Furthermore IR and diabetes rates differ by race [40, 41], gender [42], and BMI [43, 44]. MESA is a multi-ethnic, gender balanced cohort fully characterized for known risk factors of IR and DM, allowing for a uniquely complete evaluation of the association between tobacco use and IR/DM while controlling for confounders.

Unmeasured confounders in the association between IR, incident diabetes and tobacco exposure may also explain the discrepancy between studies. In our study we were able to adjust for many potential confounders (age, gender, ethnicity, BMI, income, SBP, DBP, anti-hypertensive medication, HDL, LDL, lipid-lowering medication, alcohol use). The studies included in the Surgeon General’s meta-analysis demonstrated significant heterogeneity in the number of confounders which ranged from <3 to >8. While stratified analysis of studies used in the Surgeon General’s report based on the number of these confounders did not yield substantial differences in the relationship between tobacco use and incident diabetes, many confounders were not accounted for in any of the studies used in the meta-analysis [5].

Publication bias follows from the fact that “positive” studies may be as much as three times more likely to be published than negative studies [45]. In fact, data suggests that magnitude and direction of a study's results may be a more important determinant of publication than study design, relevance, or quality [46, 47]. Publication bias is thought to be particularly prevalent for observational studies [48] and may be one contributing factor for the disparate results seen in our study and the conclusions drawn from a meta-analysis by the Surgeon General’s report.

Our analysis was performed in subjects with no known IR or diabetes at baseline in order to study the association between baseline tobacco exposure and incident type 2 diabetes. This is distinct from studies of the impact of tobacco use on IR among known type 2 diabetics or individuals with impaired glucose tolerance at the start of the study where smoking increased the magnitude of insulin resistance or glucose dysregulation [22, 24, 49, 50].

The strengths of our study include use of a well-characterized modern multi-ethnic US-based cohort with over 10 years of follow-up for incident diabetes. Cotinine measurements allowed us to reclassify tobacco use status and minimize self-report bias. Detailed information on tobacco use enabled us to evaluate for a dose response with respect to IR. Furthermore we were able to evaluate multiple forms of tobacco exposure, not just cigarettes, in relationship to IR and incident type 2 diabetes.

Limitations

Limitations to this study include how former smokers were defined as not having used tobacco in the last 30 days. There is evidence of reversal of some of the adverse effects of tobacco may not be seen at 30 days [51]. The small sample size of non-cigarette tobacco users may not be sufficiently powered to study the association of IR and incident type 2 diabetes in these users. We acknowledge that some findings with this population may have been due to chance and therefore would benefit from further research. Furthermore, there may be a bias in our smoking cohort as they are older than some of the other populations studied. Since they are older, this cohort may have inadvertently selected for a population of smoking that had made it to older age since they were healthier than other smokers. However, we were not able to find any meaningful interactions with age in our study. In our study our smoking group had a similar or slightly higher BMI and weight compared to non and former smokers (results not shown). Previous studies have shown that smokers tend to have a lower overall BMI [11], though often a higher waist to hip ratio indicating central adiposity [52]. As BMI has been associated with IR and incident diabetes it was important to control for this covariate in our modeling. Adding this covariate separately (modeling not shown) did not change our results. Finally, we only have single cotinine measures in a subset of subjects. Since current cigarette users at the time of enrollment had different patterns of tobacco exposure (time since last cigarette), there would be considerable random error in cotinine levels especially since cotinine may only remain elevated for 18–20 hours after smoking [53].

In conclusion we found no consistent association between tobacco exposures and both cross-sectional IR and incident type 2 diabetes. Given the discordance of results with respect to tobacco use and IR / incident diabetes and the importance of type 2 diabetes to global health we believe this issue warrants further study.

Supporting Information

S1 Data

Supplementary analyses_MESA IR.

(DOCX)

Acknowledgments

Research reported in this publication was supported by the AHA Tobacco Regulation and Addiction Center (A-TRAC) and FDA Center for Tobacco Products (CTP). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the Food and Drug Administration. The authors thank the other investigators, the staff, and the participants of the MESA study for their valuable contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org. Exam 1 and 5 Urinary Cotinine measurements are available to the MESA study courtesy of MESA Lung contract HL077612.

Funding Statement

This work was supported by the National Heart, Blood, and Lung Institute (NHLBI) (HL071739, N01-HC-95159 through, and N01-HC-95169), the AHA Tobacco Regulation and Addiction Center (A-TRAC), and FDA Center for Tobacco Products (CTP) (HL120163). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability

Data Availability

The Multi-Ethnic Study of Atherosclerosis (MESA) is sponsored by the National Heart Lung and Blood Institute of the National Institutes of Health. MESA data is freely available directly through the NIH via the BIOLINCC program at https://biolincc.nhlbi.nih.gov/studies/mesa/. Additionally the database is also accessible through MESA directly but requires a proposal for the data be reviewed by the MESA steering committee. Instruction for use of MESA data can be found at: https://www.mesa-nhlbi.org/default.aspx.

References

1. Haussmann HJ. Use of hazard indices for a theoretical evaluation of cigarette smoke composition. Chemical research in toxicology. 2012;25(4):794–810. Epub 2012/02/23. doi: 10.1021/tx200536w . [PubMed]
2. Stepanov I, Jensen J, Hatsukami D, Hecht SS. New and traditional smokeless tobacco: comparison of toxicant and carcinogen levels. Nicotine & tobacco research: official journal of the Society for Research on Nicotine and Tobacco. 2008;10(12):1773–82. doi: 10.1080/14622200802443544 ; PubMed Central PMCID: PMC2892835. [PMC free article] [PubMed]
3. Eliasson B. Cigarette smoking and diabetes. Progress in cardiovascular diseases. 2003;45(5):405–13. Epub 2003/04/22. doi: 10.1053/pcad.2003.00103 . [PubMed]
4. Seet RC, Loke WM, Khoo CM, Chew SE, Chong WL, Quek AM, et al. Acute effects of cigarette smoking on insulin resistance and arterial stiffness in young adults. Atherosclerosis. 2012;224(1):195–200. Epub 2012/07/31. doi: 10.1016/j.atherosclerosis.2012.06.060 . [PubMed]
5. Services USDoHaH. The Health Consequences of Smoking: 50 Years of Progress. A Report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health., 2014.
6. Pan A, Wang Y, Talaei M, Hu FB, Wu T. Relation of active, passive, and quitting smoking with incident type 2 diabetes: a systematic review and meta-analysis. The lancet Diabetes & endocrinology. 2015. Epub 2015/09/22. doi: 10.1016/s2213-8587(15)00316-2 . [PMC free article] [PubMed]
7. Willi C, Bodenmann P, Ghali WA, Faris PD, Cornuz J. Active smoking and the risk of type 2 diabetes: a systematic review and meta-analysis. Jama. 2007;298(22):2654–64. Epub 2007/12/13. doi: 10.1001/jama.298.22.2654 . [PubMed]
8. Ostenson CG, Hilding A, Grill V, Efendic S. High consumption of smokeless tobacco ("snus") predicts increased risk of type 2 diabetes in a 10-year prospective study of middle-aged Swedish men. Scandinavian journal of public health. 2012;40(8):730–7. Epub 2012/11/03. doi: 10.1177/1403494812459814 . [PubMed]
9. Persson PG, Carlsson S, Svanstrom L, Ostenson CG, Efendic S, Grill V. Cigarette smoking, oral moist snuff use and glucose intolerance. Journal of internal medicine. 2000;248(2):103–10. Epub 2000/08/18. . [PubMed]
10. Facchini FS, Hollenbeck CB, Jeppesen J, Chen YD, Reaven GM. Insulin resistance and cigarette smoking. Lancet. 1992;339(8802):1128–30. . [PubMed]
11. Chiolero A, Faeh D, Paccaud F, Cornuz J. Consequences of smoking for body weight, body fat distribution, and insulin resistance. The American journal of clinical nutrition. 2008;87(4):801–9. Epub 2008/04/11. . [PubMed]
12. Twig G, Afek A, Derazne E, Tzur D, Cukierman-Yaffe T, Gerstein HC, et al. Diabetes Risk Among Overweight and Obese Metabolically Healthy Young Adults. Diabetes care. 2014. Epub 2014/08/21. doi: 10.2337/dc14-0869 . [PubMed]
13. Canoy D, Wareham N, Luben R, Welch A, Bingham S, Day N, et al. Cigarette smoking and fat distribution in 21,828 British men and women: a population-based study. Obesity research. 2005;13(8):1466–75. Epub 2005/09/01. doi: 10.1038/oby.2005.177 . [PubMed]
14. Haffner SM, D'Agostino R, Saad MF, Rewers M, Mykkanen L, Selby J, et al. Increased insulin resistance and insulin secretion in nondiabetic African-Americans and Hispanics compared with non-Hispanic whites. The Insulin Resistance Atherosclerosis Study. Diabetes. 1996;45(6):742–8. Epub 1996/06/01. . [PubMed]
15. Henninger J, Hammarstedt A, Rawshani A, Eliasson B. Metabolic predictors of impaired glucose tolerance and type 2 diabetes in a predisposed population—A prospective cohort study. BMC endocrine disorders. 2015;15(1):51 Epub 2015/09/27. doi: 10.1186/s12902-015-0048-8 ; PubMed Central PMCID: PMCPmc4583989. [PMC free article] [PubMed]
16. Zinser MC, Pampel FC, Flores E. Distinct Beliefs, Attitudes, and Experiences of Latino Smokers: Relevance for Cessation Interventions. American Journal of Health Promotion. 2011;25(sp5):eS1–eS15. doi: 10.4278/ajhp.100616-QUAN-200 [PMC free article] [PubMed]
17. Mustonen TK, Spencer SM, Hoskinson RA, Sachs DP, Garvey AJ. The influence of gender, race, and menthol content on tobacco exposure measures. Nicotine & tobacco research: official journal of the Society for Research on Nicotine and Tobacco. 2005;7(4):581–90. doi: 10.1080/14622200500185199 . [PubMed]
18. Prevention CfDCaPNCfCDPaHPOoSaHCfDCa. The Changing Cigarette. How Tobacco Smoke Causes Disease: The Biology and Behavioral Basis for Smoking-Attributable Disease: A Report of the Surgeon General2010. [PubMed]
19. Hoffmann D, Hoffmann I, El-Bayoumy K. The Less Harmful Cigarette: A Controversial Issue. A Tribute to Ernst L. Wynder. Chemical research in toxicology. 2001;14(7):767–90. doi: 10.1021/tx000260u [PubMed]
20. Hoffmann D, Brunnemann KD, Prokopczyk B, Djordjevic MV. Tobacco-specific N-nitrosamines and Areca-derived N-nitrosamines: chemistry, biochemistry, carcinogenicity, and relevance to humans. J Toxicol Environ Health. 1994;41(1):1–52. Epub 1994/01/01. doi: 10.1080/15287399409531825 . [PubMed]
21. Sarkar M, Wang J, Liang Q. Metabolism of nicotine and 4-(methylnitrosamino)-l-(3-pyridyl)-lbutanone (NNK) in menthol and non-menthol cigarette smokers. Drug Metab Lett. 2012;6(3):198–206. Epub 2012/11/13. . [PubMed]
22. Axelsson T, Jansson PA, Smith U, Eliasson B. Nicotine infusion acutely impairs insulin sensitivity in type 2 diabetic patients but not in healthy subjects. J Intern Med. 2001;249(6):539–44. Epub 2001/06/26. . [PubMed]
23. Beziaud F, Halimi JM, Lecomte P, Vol S, Tichet J. Cigarette smoking and diabetes mellitus. Diabetes & metabolism. 2004;30(2):161–6. Epub 2004/06/30. . [PubMed]
24. Madsbad S, McNair P, Christensen MS, Christiansen C, Faber OK, Binder C, et al. Influence of smoking on insulin requirement and metbolic status in diabetes mellitus. Diabetes care. 1980;3(1):41–3. Epub 1980/01/01. . [PubMed]
25. Soulimane S, Simon D, Herman WH, Lange C, Lee CM, Colagiuri S, et al. HbA1c, fasting and 2 h plasma glucose in current, ex- and never-smokers: a meta-analysis. Diabetologia. 2014;57(1):30–9. Epub 2013/09/26. doi: 10.1007/s00125-013-3058-y . [PMC free article] [PubMed]
26. Nagaya T, Yoshida H, Takahashi H, Kawai M. Heavy smoking raises risk for type 2 diabetes mellitus in obese men; but, light smoking reduces the risk in lean men: a follow-up study in Japan. Annals of epidemiology. 2008;18(2):113–8. Epub 2007/12/18. doi: 10.1016/j.annepidem.2007.07.107 . [PubMed]
27. Nakajima M, Fukami T, Yamanaka H, Higashi E, Sakai H, Yoshida R, et al. Comprehensive evaluation of variability in nicotine metabolism and CYP2A6 polymorphic alleles in four ethnic populations. Clinical pharmacology and therapeutics. 2006;80(3):282–97. doi: 10.1016/j.clpt.2006.05.012 . [PubMed]
28. Fukami T, Nakajima M, Yamanaka H, Fukushima Y, McLeod HL, Yokoi T. A novel duplication type of CYP2A6 gene in African-American population. Drug metabolism and disposition: the biological fate of chemicals. 2007;35(4):515–20. doi: 10.1124/dmd.106.013557 . [PubMed]
29. Xia Y, Bernert JT, Jain RB, Ashley DL, Pirkle JL. Tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) in smokers in the United States: NHANES 2007–2008. Biomarkers: biochemical indicators of exposure, response, and susceptibility to chemicals. 2011;16(2):112–9. Epub 2010/12/01. doi: 10.3109/1354750x.2010.533288 . [PubMed]
30. Wang L, Rundek T, Beecham A, Hudson B, Blanton SH, Zhao H, et al. Genome-wide interaction study identifies RCBTB1 as a modifier for smoking effect on carotid intima-media thickness. Arteriosclerosis, thrombosis, and vascular biology. 2014;34(1):219–25. doi: 10.1161/ATVBAHA.113.302706 . [PMC free article] [PubMed]
31. Henley SJ, Thun M, Connell C, Calle E. Two large prospective studies of mortality among men who use snuff or chewing tobacco (United States). Cancer Causes Control. 2005;16(4):347–58. doi: 10.1007/s10552-004-5519-6 [PubMed]
32. Bild DE, Bluemke DA, Burke GL, Detrano R, Diez Roux AV, Folsom AR, et al. Multi-ethnic study of atherosclerosis: objectives and design. American journal of epidemiology. 2002;156(9):871–81. Epub 2002/10/25. . [PubMed]
33. Benowitz NL. Biomarkers of environmental tobacco smoke exposure. Environmental health perspectives. 1999;107 Suppl 2:349–55. Epub 1999/06/03. ; PubMed Central PMCID: PMCPmc1566286. [PMC free article] [PubMed]
34. Levin G, Kestenbaum B, Ida Chen YD, Jacobs DR Jr., Psaty BM, Rotter JI, et al. Glucose, insulin, and incident hypertension in the multi-ethnic study of atherosclerosis. American journal of epidemiology. 2010;172(10):1144–54. Epub 2010/10/22. doi: 10.1093/aje/kwq266 ; PubMed Central PMCID: PMCPmc3004765. [PMC free article] [PubMed]
35. McEvoy JW, Nasir K, DeFilippis AP, Lima JA, Bluemke DA, Hundley WG, et al. Relationship of cigarette smoking with inflammation and subclinical vascular disease: the Multi-Ethnic Study of Atherosclerosis. Arteriosclerosis, thrombosis, and vascular biology. 2015;35(4):1002–10. Epub 2015/03/07. doi: 10.1161/atvbaha.114.304960 . [PMC free article] [PubMed]
36. McEvoy JW, Blaha MJ, DeFilippis AP, Lima JA, Bluemke DA, Hundley WG, et al. Cigarette smoking and cardiovascular events: role of inflammation and subclinical atherosclerosis from the MultiEthnic Study of Atherosclerosis. Arteriosclerosis, thrombosis, and vascular biology. 2015;35(3):700–9. Epub 2015/01/13. doi: 10.1161/atvbaha.114.304562 ; PubMed Central PMCID: PMCPmc4404404. [PMC free article] [PubMed]
37. Onat A, Ozhan H, Esen AM, Albayrak S, Karabulut A, Can G, et al. Prospective epidemiologic evidence of a "protective" effect of smoking on metabolic syndrome and diabetes among Turkish women—without associated overall health benefit. Atherosclerosis. 2007;193(2):380–8. Epub 2006/08/24. doi: 10.1016/j.atherosclerosis.2006.07.002 . [PubMed]
38. McCulloch P, Lee S, Higgins R, McCall K, Schade DS. Effect of smoking on hemoglobin A1c and body mass index in patients with type 2 diabetes mellitus. Journal of investigative medicine: the official publication of the American Federation for Clinical Research. 2002;50(4):284–7. Epub 2002/07/12. . [PubMed]
39. Berlin I. Smoking-induced metabolic disorders: a review. Diabetes & metabolism. 2008;34(4 Pt 1):307–14. Epub 2008/05/13. doi: 10.1016/j.diabet.2008.01.008 . [PubMed]
40. Kurian AK, Cardarelli KM. Racial and ethnic differences in cardiovascular disease risk factors: a systematic review. Ethnicity & disease. 2007;17(1):143–52. Epub 2007/02/06. . [PubMed]
41. Hippisley-Cox J, Coupland C, Robson J, Sheikh A, Brindle P. Predicting risk of type 2 diabetes in England and Wales: prospective derivation and validation of QDScore. BMJ (Clinical research ed). 2009;338:b880 Epub 2009/03/20. doi: 10.1136/bmj.b880 ; PubMed Central PMCID: PMCPmc2659857. [PubMed]
42. Mickens L, Ameringer K, Brightman M, Leventhal AM. Epidemiology, determinants, and consequences of cigarette smoking in African American women: An integrative review. Addictive Behaviors. 2010;35(5):383–91. doi: 10.1016/j.addbeh.2009.12.014 [PMC free article] [PubMed]
43. Hu H, Huff CD, Yamamura Y, Wu X, Strom SS. The Relationship between Native American Ancestry, Body Mass Index and Diabetes Risk among Mexican-Americans. PloS one. 2015;10(10):e0141260 Epub 2015/10/27. doi: 10.1371/journal.pone.0141260 ; PubMed Central PMCID: PMCPmc4621045. [PMC free article] [PubMed]
44. Palaniappan LP, Carnethon MR, Fortmann SP. Heterogeneity in the relationship between ethnicity, BMI, and fasting insulin. Diabetes care. 2002;25(8):1351–7. Epub 2002/07/30. ; PubMed Central PMCID: PMCPmc3121929. [PMC free article] [PubMed]
45. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ (Clinical research ed). 1997;315(7109):629–34. Epub 1997/10/06. ; PubMed Central PMCID: PMCPmc2127453. [PMC free article] [PubMed]
46. Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR. Publication bias in clinical research. Lancet. 1991;337(8746):867–72. Epub 1991/04/13. . [PubMed]
47. Misakian AL, Bero LA. Publication bias and research on passive smoking: comparison of published and unpublished studies. JAMA: the journal of the American Medical Association. 1998;280(3):250–3. Epub 1998/07/24. . [PubMed]
48. Montori VM, Smieja M, Guyatt GH. Publication bias: a brief review for clinicians. Mayo Clinic proceedings. 2000;75(12):1284–8. Epub 2000/12/29. doi: 10.4065/75.12.1284 . [PubMed]
49. Foy CG, Bell RA, Farmer DF, Goff DC, Wagenknecht LE. Smoking and Incidence of Diabetes Among U.S. Adults: Findings from the Insulin Resistance Atherosclerosis Study. Diabetes care. 2005;28(10):2501–7. [PubMed]
50. Cugati S, Wang JJ, Rochtchina E, Mitchell P. Ten-year incidence of diabetes in older Australians: the Blue Mountains Eye Study. The Medical journal of Australia. 2007;186(3):131–5. Epub 2007/02/21. . [PubMed]
51. Wannamethee SG, Lowe GD, Shaper AG, Rumley A, Lennon L, Whincup PH. Associations between cigarette smoking, pipe/cigar smoking, and smoking cessation, and haemostatic and inflammatory markers for cardiovascular disease. European heart journal. 2005;26(17):1765–73. Epub 2005/04/09. doi: 10.1093/eurheartj/ehi183 . [PubMed]
52. Barrett-Connor E, Khaw KT. Cigarette smoking and increased central adiposity. Annals of internal medicine. 1989;111(10):783–7. Epub 1989/11/15. . [PubMed]
53. Dhar P. Measuring tobacco smoke exposure: quantifying nicotine/cotinine concentration in biological samples by colorimetry, chromatography and immunoassay methods. Journal of pharmaceutical and biomedical analysis. 2004;35(1):155–68. Epub 2004/03/20. doi: 10.1016/j.jpba.2004.01.009 . [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science