Obesity,physical activity and the development of metabolic syndrome: the atherosclerosis risk in communities study

Abstract

Background

The objective of this study is to determine the impact of body weight and physical activity on the development of metabolic syndrome (MetS).

Design

and methods We used the public use data from the Atherosclerosis Risk in Communities study. From the baseline cohort, we identified, as the study population, 9359 individuals who did not have MetS and who completed the second follow-up examination in 1993-1995.

Results

In 6 years of follow-up, 1970 individuals (25%) developed MetS. Compared with normal weight group [body mass index (BMI) <25 kg/m²], the odds ratios [95% confidence interval (CI)] of incident MetS were 2.81 (95% CI: 2.50-3.17) and 5.24 (95% CI: 4.50-6.12) for the overweight (BMI: 25-30 kg/m²) and the obese groups (BMI ≥ 30 kg/m²), respectively. Compared with persons in the lowest quartile of leisure-time physical activity, the odds ratios (95% CI) of incident MetS were 0.80 (95% CI: 0.71-0.91) and 0.92 (95% CI: 0.81-1.04) for persons in the highest and the middle quartiles of leisure-time physical activity, respectively. Our results indicated that at any level of physical activity, there is a graded increase in the risk of incident MetS with an increase in BMI, in contrast to a lack of graded association between physical activity and the incidence of MetS in all categories of BMI.

Conclusion

Keywords

cohort study metabolic syndrome obesity physical activity

Introduction

Approximately 47 million adults in the US have metabolic syndrome (MetS), an important risk factor for the subsequent development of diabetes and cardiovascular disease (CVD)[1]. According to the current National Heart, Lung, and Blood Institute (NHLBI) and American Heart Association (AHA) definition, MetS is characterized by the presence of any three of the following: increased waist circumference (WC), increased triglycerides, low high-density lipoprotein (HDL), increased blood pressure (BP), and elevated fasting glucose levels [2, 3].

The management of MetS consists of a two-pronged approach of maintaining a healthy weight and increasing physical activity. Previous small studies have shown that physical activity may be protective against MetS independent of weight loss and change in body composition [4–6]. Some retrospective studies have shown an improvement in the components of MetS with moderate weight reduction [7]. In this study, we seek to determine the relative strength relationship between these two major determinants and the development of incident MetS in a population-based sample of middle-aged Americans.

Materials and methods

Study population

We obtained the public use data from the Atherosclerosis Risk in Communities (ARIC) study, a population-based cohort study of arteriosclerotic CVD sponsored by the NHLBI [8, 9], which was designed to look into the cause and natural history of atherosclerosis and its risk factors by race, sex, location and date. Population-based samples of 15 792 residents aged 45-64 years in four US communities in North Carolina, Mississippi, Minnesota, and Maryland were examined and reviewed by institutional review boards. These participants were re-examined every 3 years with the first screen (baseline) occurring in 1987-1989, and the third in 1993-1995. The response rate for this study was about 75-80%.

Of the baseline cohort of 15 792 individuals, we identified 9359 individuals who were free of MetS at baseline as our study population. We identified characteristics in 7854 individuals during their third cohort clinical examination, which formed the sample size for this study.

Classification of metabolic syndrome status

MetS is defined according to the revised Adult Treatment Panel III criteria, adopted by NHLBI and AHA, as the presence of three or more of the following components: (i) elevated BP: systolic BP ≥ 130 mmHg, diastolic BP ≥ 85 mmHg, or use of antihypertensive medications; (ii) elevated plasma glucose: fasting glucose ≥ 100 mg/dl (5.56 mmol/l), self-reported physician diagnosed diabetes, or current use of hypoglycemic agent or insulin; (iii) elevated serum triglycerides ≥ 150mg/dl (1.7mmol/l); (iv) decreased HDL-cholesterol: ≤ 40mg/dl (1.0mmol/l) for men and ≤ 50 mg/dl (1.3mmol/l) for women; and (v) greater WC: 102 cm (40 inch) for men and 88 cm (35 inch) for women [10, 11].

Obesity and physical activity

Baseline obesity was assessed using current Center for Disease Control body mass index (BMI) criteria: BMI less than 25 kg/m² as normal weight, 25-29 kg/m² as overweight, and ≥ 30 kg/m² as obese. The leisure-time physical activity index was obtained in the ARIC study through the application of Baecke Physical Activity Questionnaires. For this analysis, we focused on leisure-time physical activity, which captures time spent viewing television (negative score), bicycling, walking, and time spent walking and bicycling to and from work or shopping. These responses were given a leisure-time physical activity score, ranging from 1 (low activity) to 5 (high activity), with 0.25 unit of increments [8]. For this analysis, we further subdivided the leisure-time physical activity into three categories using the tertile distribution - lowest tertile (1.00-2.00), middle tertile (2.25-2.50), and upper tertile (≥ 2.75).

Other covariates

Baseline demographic variables and risk factors for coronary heart disease (CHD) were assessed according to standardized protocols [12, 13]. These covariates include age, race, sex, education level, cigarette smoking status, prevalence of CHD and total cholesterol.

Statistical methods

The relationship between the incident MetS and obesity and physical activity was assessed using logistic regression models with adjustments for sample weights. The odds ratio and its 95% confidence interval (95% CI) were used to assess the strength of associations. All analyses were performed with SAS software (SAS institute Inc, Cary, North Carolina, USA)[14].

Results

Baseline characteristics

Baseline characteristics are presented in Table 1 for the entire study population and by 6-year incident MetS status. The mean age of the entire cohort at baseline was 53.5 years, with 56% female, and 21% African-American.

We used logistic regression models to estimate the odds ratios and the 95% CIs of 6-year incident MetS in relationship to baseline obesity and leisure-time physical activity. These results are presented in Table 2, from an unadjusted model (Model 1 in Table 2), and multivariable adjusted model (Model 2 in Table 2), and a model with both obesity and physical activity as independent variables (Model 3 in Table 2). As seen in Table 2, the incidence of MetS doubled in overweight groups and tripled in the obese group, with odds ratio of 2.81 (95% CI: 2.50-3.17) and 5.24 (95% CI: 4.50-6.12), respectively, from the unadjusted model. In individuals with a leisure-time physical activity in the upper tertile or in the middle tertile, the incidence of MetS was reduced with an odds ratio of 0.80 (95% CI: 0.71-0.91) and 0.92 (95% CI: 0.81-1.04), respectively from the unadjusted model. The baseline obesity and physical activity relationship with the incident MetS remain unchanged after adjusting for baseline covariables, including age, sex, race, education, current smoking, total cholesterol, and history of CHD. Furthermore, after additional adjustment for physical activity (Model 3), the baseline obesity and incident MetS relationship remained unchanged. There was significant graded increased risk of incident MetS with the increased level of obesity measures. However, after adjusting for obesity, the association between baseline physical activity and incident MetS was diminished towards a non-significant association, although the point estimates remain protective (odds ratio less than one with increased level of physical activity).

Table 1

Baseline characteristics (mean, SD, or proportion) of study population

	All (N=7854)		Incident MetS (N=1970)		No incident MetS (N=5884)
Variable	Mean	SD	Mean	SD	Mean	SD
Age (year)	53.6	5.7	53.7	5.62	53.5	5.73
BMI (kg/m²)	25.77	4.29	27.8	4.54	25.1	3.98
Race, Black (%)	21		23		20
Sex, male (%)	44		43		44
Education, less than high school (%)	17		22		15
Current smoking (%)	23.5		23.5		23.5
Hypertension (%)	17		25		15
Diabetes (%)	3.15		4.67		2.64
Fasting glucose (mg/dl)	98	19.8	101	24.0	96.9	18.0
LDL-cholesterol (mg/dl)	134	37.6	140	37.8	132	37.4
Triglycerides (mg/dl)	99	45.5	115	51.2	93	42.01
Total cholesterol (mg/dl)	211	39.0	214	40.3	209	38.5
History of CHD (%)	3.01		4.16		2.63	16.0
Leisure-Time Activity Score	2.43	0.56	2.39	0.56	2.44	0.56
Sports Activity Score	2.52	0.82	2.44	0.78	2.54	0.83

BMI, body mass index; CHD, coronary heart disease; LDL, low-density lipoprotein; MetS, metabolic syndrome; SD, standard deviation.

Table 2

Association between obesity categories, leisure-time physical activity tertiles, and the 6-year incidence of metabolic syndrome

		Model 1		Model 2		Model 3
	Incidence (cases/%)	OR	95% CI	OR	95% CI	OR	95% CI
BMI (kg/m²)
<25	516/3.98	1.00		1.00		1.00
25-30	650/28.63	2.81	2.50-3.17	2.96	2.61-3.35	2.95	2.60-3.35
≥ 30	804/42.63	5.24	4.50-6.12	5.51	4.69-6.48	5.49	4.67-6.46
Physical activity tertile
Low	641/27.1	1.00		1.00		1.00
Middle	975/25.36	0.92	0.81-1.04	0.95	0.84-1.09	0.99	0.86-1.13
Upper	354/21.55	0.80	0.71-0.91	0.85	0.75-0.97	0.90	0.78-1.03

Model 1: adjusted for sampling weight only. Model 2: adjusted for baseline age, race, sex, education, smoking, prevalent CHD, and total cholesterol, in addition to sampling weight. Model 3: adjusted for covariables in Model 2, and physical activity when modeling obesity and obesity when modeling physical activity, in addition to sampling weight. BMI, Body mass index; CHD, coronary heart disease; CI, confidence interval; OR, odds ratio.

To determine which of the baseline factors (obesity or physical activity) is a better predictor of MetS, we examined the interaction between obesity and physical activity measures, and calculated the odds ratios and their 95% CI for the nine different combinations of obesity categories (normal weight, overweight, and obese) and physical activity tertiles (lowest, middle, and upper tertiles). These data are presented in Table 3, and the point estimates (the incident odds ratios of MetS) are graphically presented in Fig. 1. Table 3 and Fig. 1 both indicated that with BMI increases, the odds ratio for developing MetS significantly increases at all levels of physical activity, especially among obese patients (with an odds ratio greater than 5). At any level of physical activity, an increase in BMI corresponds with a clear dose-response increase in odds ratio, and the 95% CIs within a physical activity level do not overlap. In contrast, at any given level of BMI, there are no statistically significant changes in the odds ratio at different levels of physical activity and the 95% CI overlaps, except for the normal BMI group where there is a graded increased odds ratio with the decrease in physical activity levels. However, the magnitude of odds ratios clearly indicated a much stronger association from BMI measures than from leisure-time physical activity measures.

Discussion

In this population-based data, we found that obesity was strongly associated with the development of incident MetS, and leisure-time physical activity was weakly associated with incident MetS. After adjusting for each obesity and leisure-time physical activity, our findings clearly suggest that obesity is strongly associated with the development of incident MetS whereas physical activity is weakly associated with it. Even when adjusting for potential confounders such as age, sex, race, level of education, total cholesterol, history of CHD, and smoking status, obesity was a strong predictor of incident MetS than physical activity. These results highlight the need to target obesity more than physical activity to prevent the development of MetS in the general population. Although previous studies have shown that both these strategies as a solution for betterment of community health [15], the clinical implication of our study is that instead of following the dual strategy of reducing obesity and engaging in regular physical activity, it may be more beneficial for patients to focus on weight loss. Some studies have also shown that an intervention like physical activity can change the parameters of MetS and can also decrease the inflammatory markers [16, 17].

Table 3

Joint effect of obesity categories and leisure-time physical activity tertiles on the 6-year incidence of metabolic syndrome

BMI (kg/m²)	Physical activity tertile	OR	95% CI
<25	Upper	1.00
25-30	Upper	3.13	2.55	3.856
≥30	Upper	5.64	4.27	7.44
<25	Middle	1.09	0.87	1.36
25-30	Middle	3.30	2.68	4.06
≥ 30	Middle	7.16	5.42	9.46
<25	Lowest	1.27	1.00	1.60
25-30	Lowest	3.36	2.70	4.17
≥ 30	Lowest	5.69	4.37	7.41

Adjusted for baseline age, race, sex, education, smoking, prevalent CHD, and total cholesterol. BMI, body mass index; CHD, coronary heart disease; CI, confidence interval; OR, odds ratio.

Fig. 1

Joint effect of obesity categories and leisure-time physical activity tertiles on the 6-year incidence of metabolic syndrome (MetS). BMI, body mass index; OR, odds ratio.

Our study population was free of MetS at baseline. Since the varied components, which comprise the clinical definition of MetS have been closely linked to each other, it makes logical sense that WC and BMI would be highly correlated, which our study confirms. It is worth noting that we also performed the entire analysis excluding persons with baseline WC greater than the MetS-increased WC [WC ≥ 102 cm (40 inch) for men and ≥ 88 cm (35 inch for women)]. In this scenario, the pattern of associations from both obesity and leisure-time physical activity did not change meaningfully (data not shown). Therefore, the strong association between obesity and MetS observed in this study does not seem to be affected by our inclusion of individuals who have a large WC but did not otherwise meet the MetS criteria.

Several groups have proposed different definitions for MetS diagnosis criteria, including the WHO [18], the European Group for the Study of Insulin Resistance [19], US Adult Treatment Panel III of the National Cholesterol Education Program [2, 3], the American Association of Clinical Endocrinology [20], the International Diabetes Federation [21], and the AHA/NHLBI [8, 9]. We used the least stringent definition of MetS as characterized by Adult Treatment Panel III. However, the ARIC study is comprehensive, with a bi-racial cohort, large sample size and long enough follow-up time to be a representative study population. One may argue that the more subjective Baecke Physical Activity Questionnaire is not as reliable and valid as compared with more objective measures, such as energy expenditure (kJ/min), oxygen consumption (l/min or ml/min per kg), or metabolic activity relative to resting conditions (MetS). In contrast, the Baecke Questionnaire has been validated by other studies [22] and it is important to note that in the ARIC study, leisure-time physical activity assessed by the Baecke Questionnaire has been consistently shown to be predictive of cardiovascular events [8, 9].

In this middle-aged US population-based sample, the 6-year incidence of MetS was 25%, which is comparable with the MetS prevalence of 24% among adults in the US [1]. As MetS can result in type 2 diabetes and arteriosclerotic CVD [10, 11, 23, 24], which is the leading cause of death in the US and other developed countries, this study contains many implications for public health officials concerned about the obesity epidemic.

If physicians are able to identify groups with MetS as a possible precursor to diabetes or CVD, they can act to prevent this progression to disease at an earlier stage [25]. Clinicians can advise these patients that managing weight is a priority for an individual with MetS. Studies have shown that favorable metabolic changes are observed in obese patients with mild reduction in weight. This is attributed to sympathetic nervous system activation leading to preferential reduction in visceral fat mass [26]. Methods to prevent weight gain include assessing diet using a daily journal or speaking with a nutritionist about guidelines for healthy eating [12]. Physicians can also counsel patients about their attitudes about lifestyle choices that are causing weight gain and how to improve these behaviors [12]. Stone [12] describes several important strategies based on the Obesity Guidelines Panel from the National Institutes of Health report including (i) multistep strategies work better than a single approach and (ii) motivation increases as reasonable goals are set.

The major strengths of this study are a large number of African-American participants, detailed dietary habits and atherosclerotic disease pattern, which can be projected into US population. The other strengths are simple design, prospective study, large number of patients with potentially confounding variables, and extensive follow-up time. The limitations are use of self-reported physical activity, which can lead to bias. The other limitations are lack of Hispanic representation and a limited number of questions in the study regarding physical activity.

Conclusion

Obesity remains a national concern, affecting 65% of adults and 17% of children in the US and is growing to be a worldwide issue [13, 27]. This population-based data indicates that body weight is a more important determinant than leisure-time physical activity in the development of MetS. Thus, assessing obesity by BMI and treating overweight and obese patients are important goals in curbing this crisis [28–31]. Reducing body weight by an achievable amount of 5-10% would significantly improve the prognosis for patients with MetS [13]. As obesity is an important indicator for the development of MetS, targeting this should remain a focus for public health officials.

Footnotes

Acknowledgements

The ARIC Study is conducted and supported by the National Heart, Lung, and Blood Institute (NHLBI) in collaboration with the ARIC Investigators. The authors thank the NHLBI for providing the public use data for this investigation. The authors thank the staff and participants of the ARIC study for their important contributions. We also thank Laurie Schwing, MLS, for advice, guidance and editorial support for this project. The authors did not receive any source of funding for this project from anyone. There is no conflict of interest to report.

References

1 Ford

. Prevalence of the metabolic syndrome in US populations. Endocrinol Metab Clin North Am 2004; 33:333–350.

2 Executive summary of the 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). JAMA 2001; 285:2486–2497.

3 National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, 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.

4 Ivy

. Role of exercise training in the prevention and treatment of insulin resistance and non-insulin-dependent diabetes mellitus. Sports Med 1997; 24:321–336.

5 Rice

Janssen

Hudson

Ross

. Effects of aerobic or resistance exercise and/or diet on glucose tolerance and plasma insulin levels in obese men. Diabetes Care 1999; 22:684–691.

6 Ross

Dagnone

Jones

Smith

Paddags

Hudson

Janssen

. Reduction in obesity and related comorbid conditions after diet-induced weight loss or exercise-induced weight loss in men: a randomized, controlled trial. Ann Intern Med 2000; 133:92–103.

7 Case

Jones

Nelson

O'Brian Smith

Ballantyne

. Impact of weight loss on the metabolic syndrome. Diabetes Obes Metab 2002; 4:407–414.

8 ARIC investigators. The Atherosclerosis Risk in the Communities (ARIC) Study: design and objectives. Am J Epidemiol 1989; 129:687–702.

9 National Heart Lung and Blood Institute. ARIC manuals of operation: no. 1, general description and study management; no. 2, cohort component procedures; no. 3, surveillance component procedures; no. 4, pulmonary function assessment; no. 5, electrocardiography; no. 6, ultrasound assessment; no. 7, blood collection and processing; no. 8, lipid and lipoprotein determinations; no. 9, hemostasis determinations; no. 10, clinical chemistry determinations; no. 11, sitting blood pressure and postural changes in blood pressure and heart rate; no. 12, quality assurance. Chapel Hill, NC: ARIC Coordinating Center, School of Public Health, University of North Carolina; 1987.

10.

10 Isomaa

Almgren

Tuomi

Forsen

Lahti

Nissen

Taskinen

. Cardiovascular morbidity and mortality associated with the metabolic syndrome. Diabetes Care 2001; 24:683–689.

11.

11 Lakka

Laaksonen

Lakka

Niskanen

Kumpusalo

Tuomilehto

Salonen

. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA 2002; 288:2709–2716.

12.

12 Stone

. Focus on lifestyle change and the metabolic syndrome. Endocrinol Metab Clin N Am 2004; 33:493–508.

13.

13 Kushner

. Obesity management. Gastroenterol Clin N Am 2004; 36:191–210.

14.

14 The SAS System for Windows. Cary, NC: SAS Institute Inc.; 2007.

15.

15 Ingelsson

Ärnlöv

Sundström

Risérus

Michaëlsson

Byberg

. Relative importance and conjoint effects of obesity and physical inactivity for the development of insulin resistance. Eur J Cardiovas Prev Rehabil 2009; 16:28–33.

16.

16 Balkau

Vierron

Vernay

Born

Arondel

Petrella

Ducimetiere

. The impact of 3-year changes in lifestyle habits on metabolic syndrome parameters: the DESIR Study. Eur J Cardiovas Prev Rehabil 2006; 13:334–340.

17.

17 Pitsavos

Panagiotakos

Chrysohoou

Kavouras

Stefanadis

. The associations between physical activity, inflammation, and coagulation markers, in people with metabolic syndrome: the ATTICA study. Eur J Cardiovas Prev Rehabil 2005; 12:151–158.

18.

18 Alberti

Zimmet

. Definition, diagnosis and classification of diabetes mellitus and its complications, part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 1998; 15:539–553.

19.

19 Balkau

Charles

. Comments on the provisional report from the WHO consultation. European Group for the Study of Insulin Resistance (EGIR). Diabet Med 1999; 16:442–443.

20.

20 Einhorn

Reaven

Cobin

Ford

Ganda

Handelsman

. American College of Endocrinology position statement on the insulin resistance syndrome. Endocr Pract 2003; 9:237–252.

21.

21 International Diabetes federation. Worldwide definition of the metabolic syndrome. Available at: http://www.idf.org/webdata/docs/IDF_Meta-syndrome_definition.pdf. [Access 27 October 2006]

22.

22 Baecke

Burema

Frijters

. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr 1982; 36:936–942.

23.

23 Grundy

Cleeman

Daniels

Donato

Eckel

Franklin

. Diagnosis and management of the metabolic syndrome. An American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Executive Summary. Circulation 2005; 112:2735–2752.

24.

24 Resnick

Jones

Ruotolo

Jain

Henderson

Howard

Strong Heart

Study

. Insulin resistance, the metabolic syndrome, and risk of incident cardiovascular disease in nondiabetic American Indians: the Strong Heart Study. Diabetes Care 2003; 26:861–867.

25.

25 Stolar

. Metabolic syndrome: controversial but useful. CleveClin J Med 2007; 74:199–208.

26.

26 Busetto

. Visceral obesity and metabolic syndrome: effects of weight loss. Nutr Metabol Cardiovas Dis 2001; 11:195–204.

27.

27 Cameron

Shaw

Zimmet

. The metabolic syndrome: prevalence in worldwide populations. Endocrinol Metab Clin North Am 2004; 33:351–375.

28.

28 Flegal

Carroll

Kuczmarski

Johnson

. Overweight and obesity in the United States: prevalence and trends, 1960-1994. Int J Obes 1998; 22:39–47.

29.

29 Ogden

Johnson

Carroll

Curtin

McDowell

Talak

Flegal

. Prevalence of overweight and obesity in the United States, 1999-2004. JAMA 2006; 295:1549–1555.

30.

30 Thom

Haase

Rosamond

Howard

Rumsfeld

Manolio

. Heart disease and stroke statistics-2006 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 2006; 113:e85–e151.

31.

31 Silventoinen

Sans

Tolonen

Monterde

Kuulasmaa

Kesteloot

Tuomilehto

. WHO MONICA Project. Trends in obesity and energy supply in the WHO MONICA Project. Int J Obes 2004; 28:710–718.