Alterations in Multiple Lifestyle Factors in Subjects with the Metabolic Syndrome Independently of Obesity

Abstract

Background:

Many lifestyle factors have been associated with the metabolic syndrome (MetS). However, most of these studies have not considered the potential impact of obesity and have often only investigated one lifestyle factor at the time. We aimed to investigate the interplay between body mass index (BMI) and MetS with respect to multiple lifestyle factors.

Methods:

BMI and MetS [National Cholesterol Education Program (NCEP)/Adult Treatment Panel III criteria] were assessed in a sample of 18,880 subjects aged 45–75 years from the population-based EpiHealth study. Participants were categorized into six groups according to BMI category (normal weight/BMI <25 kg/m², overweight/BMI 25–30 kg/m², and obesity/BMI >30 kg/m²) and MetS status (+/−, NCEP criteria). A wide range of lifestyle factors related to physical activity, smoking, alcohol, sleep quality, working conditions, quality of life and stress, and eating patterns were assessed using a questionnaire.

Results:

Prevalent MetS (23% in the sample) was associated with less physical activity (P < 0.0001), more TV watching (P < 0.0001), more years of smoking (P < 0.0001), lower education level (P = 0.007), and experiencing a poor general quality of life (P < 0.0001). These lifestyle factors were all associated with MetS, independently of each other and independently of BMI. Similar results were generated when number of MetS components and presence/absence of individual MetS components were used as outcomes.

Conclusions:

This cross-sectional study identified alterations in a number of lifestyle factors associated with MetS independently of each other and independently of BMI. Future longitudinal studies are needed to assess causal and temporal relationships between lifestyle factors and MetS development.

Introduction

The Metabolic syndrome (MetS) denotes a clustering of cardiovascular risk factors in the same individual. The condition was first described in 1988.^1,2 Having MetS at middle-age approximately doubles the risk of a future cardiovascular event during the following decades.³

Alterations in a number of lifestyle factors, such as smoking, alcohol intake, food intake patterns, exercise habits and physical inactivity, work stress, education level, employment, and quality of life have all been linked to MetS in cross-sectional and prospective studies.^{4

–20} Some of these lifestyle factor deviations have been claimed to be involved in the pathogenesis of MetS, while others might rather be caused by MetS.

However, as obesity is one of the cornerstones of MetS, it is often not clear from previous studies if the lifestyle alterations described to be associated with MetS are driven by obesity or not. In addition, most studies in the literature focus on a small number of lifestyle factors and do not consider the possible interdependence of multiple lifestyle factors.

We therefore investigated multiple lifestyle factors in almost 20,000 individuals from the population-based EpiHealth study²¹ with the aims to identify lifestyle factors that are associated with MetS (i) independently of obesity/body mass index (BMI) and (ii) independently of each other. An additional aim was to investigate the relationships between lifestyle factors and MetS in subjects with obesity, as individuals with obesity without metabolic disturbances have been claimed to be protected from the cardiometabolic consequences otherwise associated with obesity.^22

–25

Materials and Methods

Materials

Starting from April 27th 2011, men and women aged 45–75 years in the Swedish towns of Uppsala and Malmö have been invited in a random manner to an ongoing health screening survey, EpiHealth.²¹ The present evaluation was conducted in a sample of 18,880 individuals included in EpiHealth between 2011 and 2015. The present study was approved by the local ethics committee and the participants have given their informed consent.

Measurements

MetS was defined according to the National Cholesterol Education Program/Adult Treatment Panel III (NCEP/ATP III) criteria.^26,27 Three of the following five criteria should be fulfilled: Blood pressure ≥130/85 mmHg or antihypertensive treatment, fasting plasma glucose ≥6.1 mmol/L, serum triglycerides ≥1.7 mmol/l, waist circumference >102 cm in men and >88 cm in women, and high-density lipoprotein (HDL) cholesterol <1.0 mmol/l in men and <1.3 in women. Blood samples were drawn after a minimum of 6 hr of fasting. Blood glucose and lipids were measured by standard laboratory techniques. Blood pressure was measured manually in triplicate in the sitting position. Height was recorded and waist circumference was measured at the umbilical level.

Questionnaire on lifestyle factors

Study participants answered a computer-based questionnaire with questions about alcohol intake (drinks per week); smoking habits (how many years they had been smoking during their life); leisure-time physical activity (five-point scale from sedentary to athlete) with specifications for how many minutes per day they spent watching TV, sitting working with their hands, or standing; general sleep quality (five-graded scale from very poor to very good); irregular eating habits (defined as not having breakfast, lunch, and dinner every day); and education level (≤9 years, 10–12 years, or >12 years, which equals university level). Participants were further asked about hours of work per week, current/previous shift work, current/previous nighttime work, unemployment, and premature retirement due to disease. Unemployment and premature retirement due to disease were combined in a “not working” category in further analyses. Six psychosocial working conditions (conflicting demands, freedom to make decisions, enough time to perform tasks, opportunity to learn new things, choose how to perform tasks, and repetitive job) were assessed on a four-graded scale from often to never. Participants were asked to rate their general quality of life on a scale from 0 to 10, where 10 denotes an excellent quality of life. General stress was assessed by asking the subjects to rate (on a five-graded scale from never to very often) how often during the last month they had felt that they were not in control of the important factors of their life, and how often they had felt nervous and stressed.

Statistical analyses

By defining normal weight as BMI <25 kg/m², overweight as BMI 25–30 kg/m², and obesity as BMI >30 kg/m², we grouped participants in six categories: normal weight without MetS, normal weight with MetS, overweight without MetS, overweight with MetS, obesity without MetS, and obesity with MetS.

Our primary test was to evaluate the overall differences between the BMI/MetS groups regarding lifestyle factors. All other tests performed were regarded as secondary, explorative tests. The overall differences between tests were therefore subjected to adjustment for multiple tests (Bonferroni adjustment for 21 tests, P < 0.0023). If significant differences were found, we proceeded to evaluating if BMI and/or MetS prevalence (as two independent factors) and/or a BMI/MetS interaction term were driving the differences. We also evaluated if there were sex-based differences among BMI/MetS groups using a BMI/MetS group X sex interaction term. To examine the influence of obesity in more detail, we evaluated differences in lifestyle factors between subjects in the two obesity groups (with and without MetS). All above-mentioned analyses were adjusted for age, sex, education level, date of investigation, and study site (Uppsala/Malmö).

We used ordinal logistic regression models for survey answers that followed an ordinal scale. For continuous variables with a normal distribution, such as age, we used linear regression models. For continuous variables with a nonnormal distribution, such as drinks per week and years of smoking, with many observations being zero, we used Tobit censored regression (lower limit censoring at 0).

The lifestyle factors significantly related to MetS independently of BMI were entered as independent variables in a logistic regression model with MetS as the outcome. Age, sex, and BMI were used as confounders. Questions on working conditions were added to a separate model, as they were only answered by the proportion of subjects still at work. We investigated the five different MetS components separately in a similar manner and also used linear regression to study how lifestyle factors were related to the number of MetS components.

STATA 14 (StataCorp, College Station, TX) was used for all calculations.

Results

Participant characteristics are presented in Table 1. Of the 18,880 individuals in the study sample, 42% were normal weight, 42% were overweight, and 16% were obese. The total MetS prevalence in the study sample was 23%. The mean number of MetS criteria was 1.6 (standard deviation 1.2).

Table 1.

Age and Sex Distribution in Different Combinations of Body Mass Index and the Metabolic Syndrome

	Normal weight		Overweight		Obesity
Variable	MetS−	MetS+	MetS−	MetS+	MetS−	MetS+	P value
N	7552	385	5857	2101	1160	1825
Mean age [SD] (years)	58.9 [8.6]	63.4 [7.2]	60.0 [8.5]	62.6 [7.7]	58.8 [8.5]	61.0 [8.1]	<0.0001
Sex (% women)	68	62	46	49	59	48	<0.0001

P-values are given for overall differences between BMI/MetS groups.

BMI, body mass index; MetS, metabolic syndrome.

Lifestyle factors in relation to MetS prevalence

When the lifestyle factors were analyzed one by one, significant differences between BMI/MetS groups were found for leisure-time physical activity, time spent watching TV, smoking habits, irregular eating habits, sleep quality, general quality of life, perception of not having control over the important aspects of one's life, feeling nervous and stressed, education level, unemployment, hours at work per week, shift work, years with nighttime work, and exposure to conflicting demands at work (P < 0.0001 for all) (see Supplementary Tables S1–S5; Supplementary Data are available online at www.liebertpub.com/met). The respective frequencies in each BMI/MetS group for physical activity, TV watching, general quality of life, and education level are graphically represented in Fig. 1.

FIG. 1.

Alterations in lifestyle factors between BMI/MetS groups: leisure-time physical activity on a scale from 1 (sedentary) to 5 (athlete) (A), TV watching in minutes per day (B), general quality of life on a scale from 0 to 10 where 10 denotes excellent (C), and education level (≤9 years, 10–12 years, or >12 years which equals university level) (D). Graphs show proportions of answers in questionnaire. NW, normal weight (BMI <25 kg/m²); OW, overweight (BMI 25–30 kg/m²); OB, obesity (BMI >30 kg/m²); +/−, denotes presence or absence of MetS, respectively. MetS, metabolic syndrome; BMI, body mass index.

No significant differences between BMI/MetS groups were found for alcohol intake, time spent sitting and working with one's hands per day, time spent standing per day, or the psychosocial working conditions freedom to decide, creativity, opportunity to learn new things, repetitive work, and having enough time to perform tasks (see Supplementary Tables S1, S2, and S5).

Most of the lifestyle factors that exhibited significant differences between BMI/MetS groups were related to both BMI and MetS when both BMI and MetS were included in the models together with the confounders previously described. Thus, independently of the BMI level, individuals with MetS were less physically active, spent more time per day watching TV, had been a smoker longer, had more irregular eating habits, had a lower education level, were more often unemployed, and were more exposed to conflicting demands at work (P < 0.0001 for all). They also had lower sleep quality (P < 0.01), were more prone to feeling nervous and stressed (P < 0.0001), more often perceived that they did not have control over the important aspects of their lives (P < 0.0001), and had a lower self-rated general quality of life (P < 0.0001). We did not find any significant interactions between BMI and MetS.

Individuals that worked many hours per week were more likely to have MetS (P = 0.006). Current or previous shift work and more years with nighttime work were more common in individuals with higher BMI regardless of MetS (P < 0.0001 for both).

We found significant interactions between BMI/MetS group and sex for alcohol intake and quality of life. Although alcohol intake showed no significant overall differences between BMI/MetS groups (P = 0.047, Supplementary Table S2), the number of drinks per week increased with higher BMI and prevalent MetS in men, but decreased in women. The negative association between MetS (and BMI) and general quality of life was more pronounced in women than in men (p = 0.0008).

The lifestyle factors that differed between BMI/MetS groups were entered as independent variables in a logistic regression model with MetS prevalence as outcome, also including age, sex, and BMI, in the 12,247 subjects with complete data on all lifestyle variables. More years of smoking, less leisure-time physical activity, more TV watching, lower education level, and a poor general quality of life remained significantly related to MetS, independently of each other and independently of BMI (P < 0.0001 for all variables, Table 2). Sleep quality, irregular eating habits, and stress indices were not significant in this model (Table 2). Substituting BMI for fat mass (measured by bioimpedance), fat-free mass, waist circumference, or waist/hip ratio did not affect what variables were significantly related to MetS, except that smoking was no longer significant when using the waist/hip ratio.

Table 2.

Relationships Between Prevalent Metabolic Syndrome and Multiple Lifestyle Factors, Adjusted for Age, Sex, and Body Mass Index, in 12,247 Subjects with Complete Data for All Variables

Variable	OR	OR 95% CI	P value
Leisure-time physical activity	0.87	0.826–0.920	<0.0001
TV watching	1.00	1.001–1.002	<0.0001
Years of smoking	1.01	1.009–1.019	<0.0001
Irregular eating habits	0.93	0.833–1.033	0.17
Sleep quality	0.97	0.921–1.019	0.23
General quality of life	0.96	0.931–0.986	0.004
Perception of not having control	0.96	0.916–1.014	0.16
Feeling nervous and stressed	1.01	0.947–1.068	0.86
Education level	0.92	0.861–0.976	0.007
Age	1.04	1.032–1.045	<0.0001
Female sex	0.81	0.734–0.896	<0.0001
BMI	1.36	1.341–1.382	<0.0001

OR, odds ratio; CI, confidence interval.

As many of the subjects had reached retirement age, the unemployment variable was only present for approximately half the sample. We therefore added unemployment, hours at work per week, and conflicting demands at work in a separate model. None of the work-related variables was significant in this model.

Lifestyle factors in relation to MetS criteria

Using number of MetS criteria variable as outcome generated very similar results to using the binary MetS prevalence variable (see Supplementary Table S6 for details).

The same lifestyle factors were also tested for interdependence against the five individual MetS criteria in separate logistic regression models with age, sex, and BMI as confounders (see Supplementary Tables S7–S11 for details). The blood pressure criterion was related to more TV watching and lower education level. The glucose criterion was related to smoking and more TV watching. The triglyceride criterion was related to smoking, less leisure-time physical activity, more TV watching, lower education level, and a poor general quality of life. The waist circumference criterion was related to smoking, less leisure-time physical activity, more TV watching, and a poor general quality of life. Finally, the HDL criterion was related to less leisure-time physical activity, less standing, lower education level, and a poor general quality of life.

Differences in lifestyle factors between individuals with obesity with or without MetS

When comparing the two groups with obesity, those with prevalent MetS exhibited lower physical activity and more TV watching per day (P < 0.0001 for both), had smoked for more years (P = 0.008), more often had irregular eating habits (P = 0.03), and had lower education level (P = 0.04).

In a logistic regression model, only physical activity and TV watching of the above mentioned lifestyle factors were independently different between subjects with obesity with and without MetS (P < 0.01 for both).

Discussion

MetS prevalence in the EpiHealth cohort was associated with less physical activity, more TV watching, more years of smoking, lower education level, and experiencing a poor general quality of life. The same lifestyle factors, with the addition of unemployment, were also associated with the total number of MetS criteria and related in different combinations to the individual MetS criteria.

Our results confirm previous findings on lifestyle factors of importance for metabolic health,^{4,5,7

–11,13
–15,17,18,20} but in contrast to previous studies, we were able to show in a large population-based sample that these highly significant lifestyle alterations were characteristics of MetS independently of BMI and independently of each other.

Our results confirm results from previous studies that the respective amounts of leisure-time physical activity and sedentary behavior, here measured as time spent watching TV, differ significantly between individuals with obesity and without and with MetS.^28

–33 Our addition to this field of research is that we measured a large number of lifestyle factors of importance for metabolic health in a large number of individuals, but we found that physical activity and TV watching were the only factors that determined metabolic health among individuals with obesity.

We found marked sex differences with respect to alcohol intake and MetS. Gender and geographical differences in the effect of alcohol intake on MetS have indeed been suggested in recent meta-analyses.^5,16 Low and moderate alcohol consumption is generally regarded as protective of MetS while higher consumption increases the MetS risk, at least in men.^{5,16,28,33,34} We note, however, that alcohol consumption in the present sample was rather low (median two drinks/week, 95th percentile at six drinks/week), and we were therefore not able to evaluate the possible effects of a very high alcohol intake.

When evaluating large datasets, as the present, it is of importance to distinguish between statistical significance and clinical significance. In the present analysis, we found that each additional year of smoking increased the MetS risk by 1%; reporting a high physical activity was associated with a 24%–41% decreased risk compared to being sedentary; watching TV more than 4 hr per day was associated with a 44%–95% increased risk compared with non-TV watchers; university education was associated with a 12% reduced risk compared with those with 9 years or less of education; reporting a high general quality of life was associated with a 29%–35% reduced risk of having MetS compared with those reporting a low general quality of life. We therefore believe that the identified lifestyle factors are not only highly statistically significant but also indeed related to MetS in a clinically significant manner.

MetS prevalence in the current sample was 23%. This level is in accordance with several other studies in northern European populations using the NCEP/ATP III-criteria,^15,35
–37 although it is lower than the levels reported from similar age groups in the United States³⁸ or levels using other MetS definitions.³⁵

Strengths and limitations

Among the strengths of this study is that we assessed the association of prevalent MetS with a large number of lifestyle factors in a large, population-based sample. We furthermore investigated the potential interplay between BMI and MetS prevalence regarding these lifestyle factors, as well as their interdependence. The main limitation of our study is the cross-sectional, observational nature that prevents assessment of directions of causality. In addition, most of our subjects were of northern European descent and all were between 45 and 75 years of age, which may limit the generalizability of our results to other age and/or ethnic groups.

Conclusions

MetS was more common in individuals who were less physically active, spent more time watching TV, had been smoking longer, had lower education level, and experienced a poor general quality of life. These lifestyle factors were associated with MetS independently of BMI and independently of each other. In individuals with obesity, those with MetS were less physically active and spent more time watching TV. Future longitudinal studies are required to establish the causal and temporal relationships between these lifestyle factors and MetS development.

Footnotes

Author Disclosure Statement

J.S. serves on the advisory board for the fitness company Itrim. E.I. is an advisor and consultant for Precision Wellness, Inc. No potential competing financial interests exist for the other authors.

References

Lind

, Jakobsson

, Lithell

, et al. Relation of serum calcium concentration to metabolic risk factors for cardiovascular disease. BMJ, 1988; 297:960–963.

Reaven

. Banting lecture 1988. Role of insulin resistance in human disease. Diabetes, 1988; 37:1595–1607.

Sundström

, Risérus

, Byberg

, et al. Clinical value of the metabolic syndrome for long term prediction of total and cardiovascular mortality: Prospective, population based cohort study. BMJ, 2006; 332:878–882.

Aljohani

. Metabolic syndrome: Risk factors among adults in Kingdom of Saudi Arabia. J Family Community Med, 2014; 21:170–175.

Alkerwi

, Boutsen

, Vaillant

, et al. Alcohol consumption and the prevalence of metabolic syndrome: A meta-analysis of observational studies. Atherosclerosis, 2009; 204:624–635.

Bergmann

, Gyntelberg

, Faber

. The appraisal of chronic stress and the development of the metabolic syndrome: A systematic review of prospective cohort studies. Endocr Connect, 2014; 3:R55–R80.

Cena

, Fonte

, Turconi

. Relationship between smoking and metabolic syndrome. Nutr Rev, 2011; 69:745–753.

Chiolero

, Faeh

, Paccaud

, et al. Consequences of smoking for body weight, body fat distribution, and insulin resistance. Am J Clin Nutr, 2008; 87:801–809.

Churilla

, Zoeller

. Physical activity: Physical activity and the metabolic syndrome: A review of the evidence. Am J Lifestyle Med, 2008; 2:118–125.

10.

Edwardson

, Gorely

, Davies

, et al. Association of sedentary behaviour with metabolic syndrome: A meta-analysis. PLoS One, 2012; 7:e34916.

11.

, Xi

, Xue

, et al. Association between leisure time physical activity and metabolic syndrome: A meta-analysis of prospective cohort studies. Endocrine, 2014; 46:231–240.

12.

Ngo

, Paquet

, Howard

, et al. Area-level socioeconomic characteristics, prevalence and trajectories of cardiometabolic risk. Int J Environ Res Public Health, 2014; 11:830–848.

13.

Park

, Oh

, Cho

S-I

, et al. The metabolic syndrome and associated lifestyle factors among South Korean adults. Int J Epidemiol, 2004; 33:328–336.

14.

Petersen

, Nielsen

, Bauman

, et al. Joint association of physical activity in leisure and total sitting time with metabolic syndrome amongst 15,235 Danish adults: A cross-sectional study. Prev Med, 2014; 69:5–7.

15.

Qader

, Shakir

, Nyberg

, et al. Sociodemographic risk factors of metabolic syndrome in middle-aged women: Results from a population-based study of Swedish women, The Women's Health in the Lund Area (WHILA) study. Climacteric, 2008; 11:475–482.

16.

Sun

, Ren

, Liu

, et al. Alcohol consumption and risk of metabolic syndrome: A meta-analysis of prospective studies. Clin Nutr, 2014; 33:596–602.

17.

Ford

, Li

. Metabolic syndrome and health-related quality of life among U.S. adults. Ann Epidemiol, 2008; 18:165–171.

18.

Miettola

, Niskanen

, Viinamäki

, et al. Metabolic syndrome is associated with impaired health-related quality of life: Lapinlahti 2005 study. Qual Life Res, 2008; 17:1055–1062.

19.

Sierra-Johnson

, Undén

A-L

, Linestrand

, et al. Eating meals irregularly: A novel environmental risk factor for the metabolic syndrome. Obesity, 2008; 16:1302–1307.

20.

Slagter

, van Vliet-Ostaptchouk

, van Beek

, et al. Health-related quality of life in relation to obesity grade, type 2 diabetes, metabolic syndrome and inflammation. PLoS One, 2015; 10:e0140599.

21.

Lind

, Elmståhl

, Bergman

, et al. EpiHealth: A large population-based cohort study for investigation of gene-lifestyle interactions in the pathogenesis of common diseases. Eur J Epidemiol, 2013; 28:189–197.

22.

Brochu

, Tchernof

, Dionne

, et al. What are the physical characteristics associated with a normal metabolic profile despite a high level of obesity in postmenopausal women?. J Clin Endocrinol Metab, 2001; 86:1020–1025.

23.

Karelis

, Brochu

, Rabasa-Lhoret

. Can we identify metabolically healthy but obese individuals (MHO)?. Diabetes Metab, 2004; 30:569–572.

24.

Meigs

, Wilson

, Fox

, et al. Body mass index, metabolic syndrome, and risk of type 2 diabetes or cardiovascular disease. J Clin Endocrinol Metab, 2006; 91:2906–2912.

25.

McAuley

, Blair

. Obesity paradoxes. J Sports Sci, 2011; 29:773–782.

26.

National Cholesterol Education Program (NCEP) Expert Panel on Detection E., and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). Final report. Circulation, 2002; 106:3143–3421.

27.

Grundy

, Brewer

Jr ., Cleeman

, et al. Definition of metabolic syndrome: Report of the National Heart, Lung, and Blood Institute/American Heart Association conference on scientific issues related to definition. Circulation, 2004; 109:433–438.

28.

Bradshaw

, Monda

, Stevens

. Metabolic syndrome in healthy obese, overweight, and normal weight individuals: The atherosclerosis risk in communities study. Obesity (Silver Spring), 2013; 21:203–209.

29.

Camhi

, Crouter

, Hayman

, et al. Lifestyle behaviors in metabolically healthy and unhealthy overweight and obese women: A preliminary study. PLoS One, 2015; 10:e0138548.

30.

Camhi

, Waring

, Sisson

, et al. Physical activity and screen time in metabolically healthy obese phenotypes in adolescents and adults. J Obes, 2013; 2013:984613.

31.

Khan

, Wang

, Karvonen-Gutierrez

, et al. Progression from metabolically benign to at-risk obesity in perimenopausal women: A longitudinal analysis of study of women across the nation (SWAN). J Clin Endocrinol Metab, 2014; 99:2516–2525.

32.

Phillips

, Dillon

, Harrington

, et al. Defining metabolically healthy obesity: Role of dietary and lifestyle factors. PLoS One, 2013; 8:e76188.

33.

Gutierrez-Repiso

, Soriguer

, Rojo-Martinez

, et al. Variable patterns of obesity and cardiometabolic phenotypes and their association with lifestyle factors in the Di@bet.es study. Nutr Metab Cardiovasc Dis, 2014; 24:947–955.

34.

Wildman

, Muntner

, Reynolds

, et al. The obese without cardiometabolic risk factor clustering and the normal weight with cardiometabolic risk factor clustering: Prevalence and correlates of 2 phenotypes among the US population (NHANES 1999–2004). Arch Intern Med, 2008; 168:1617–1624.

35.

Cameron

, Shaw

, Zimmet

. The metabolic syndrome: Prevalence in worldwide populations. Endocrinol Metab Clin North Am, 2004; 33:351–375, table of contents.

36.

Mørkedal

, Vatten

, Romundstad

, et al. Risk of myocardial infarction and heart failure among metabolically healthy but obese individuals: HUNT (Nord-Trondelag Health Study), Norway. J Am Coll Cardiol, 2014; 63:1071–1078.

37.

Lind

, Siegbahn

, Ingelsson

, et al. A detailed cardiovascular characterization of obesity without the metabolic syndrome. Arterioscler Thromb Vasc Biol, 2011; 31:e27–e34.

38.

Aguilar

, Bhuket

, Torres

, et al. Prevalence of the metabolic syndrome in the United States, 2003–2012. JAMA, 2015; 313:1973–1974.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

0.02 MB

0.04 MB

0.03 MB