Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 
Print this page Email this page Users Online: 136

  Table of Contents  
ORIGINAL ARTICLE
Year : 2020  |  Volume : 13  |  Issue : 4  |  Page : 326-332  

Bidirectional association between central obesity and serum lipids (triglycerides and high-density lipoprotein-cholesterol): A community-based study on predictors of central obesity


Department of Preventive and Social Medicine, All India Institute of Hygiene and Public Health, Kolkata, West Bengal, India

Date of Submission03-Aug-2019
Date of Decision03-Aug-2019
Date of Acceptance15-Oct-2019
Date of Web Publication20-Jul-2020

Correspondence Address:
Sweta Suman
Department of Preventive and Social Medicine, All India Institute of Hygiene and Public Health, Kolkata 110 C.R. Avenue, Kolkata - 700 073, West Bengal
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/mjdrdypu.mjdrdypu_226_19

Rights and Permissions
  Abstract 


Background: Obesity is considered as one of the major noncommunicable diseases of the modern world. It is considered as a major risk factor for cardiovascular diseases, type 2 diabetes mellitus, and dyslipidemia. This study was planned to find the prevalence and predictors of central obesity and its relationship with serum lipids. Materials and Methods: This was a community-based observational cross-sectional study among 388 participants aged 18–49 years in an area of Kolkata from 2016 to 2018. Data collection was done using a structured questionnaire along with anthropometry, blood pressure measurement, and relevant blood tests (fasting blood sugar and lipid profile). Data were analyzed using Statistical Package for the Social Sciences Version 16.0, and logistic regression was done to determine the strength of association between central obesity and different risk factors. Linear trends between waist circumference (WC) and serum lipids were evaluated to explore the bidirectional relationship between the two. Results: Central obesity was present among 66% of study participants. Female gender, high per capita income, poor diet, sedentary lifestyle, decreasing high-density lipoprotein-cholesterol (HDL-C), and hypertension were found as predictors of central obesity. A statistically significant bidirectional association was found between serum lipids (triglycerides (TGs) and HDL-C) and WC. Conclusion: Central obesity was found as an important public health problem which is related to diet, lifestyle, gender, income, and other comorbidities. TG and HDL-C seemed to have bidirectional relationship with WC, and high WC may be considered as surrogate indicator for dyslipidemia (especially low HDL-C and high TGs level).

Keywords: Bidirectional association, central obesity, predictor of central obesity, serum lipids


How to cite this article:
Banerjee R, Dasgupta A, Suman S, Pan T, Burman J, V. Augustine AT. Bidirectional association between central obesity and serum lipids (triglycerides and high-density lipoprotein-cholesterol): A community-based study on predictors of central obesity. Med J DY Patil Vidyapeeth 2020;13:326-32

How to cite this URL:
Banerjee R, Dasgupta A, Suman S, Pan T, Burman J, V. Augustine AT. Bidirectional association between central obesity and serum lipids (triglycerides and high-density lipoprotein-cholesterol): A community-based study on predictors of central obesity. Med J DY Patil Vidyapeeth [serial online] 2020 [cited 2020 Dec 5];13:326-32. Available from: https://www.mjdrdypv.org/text.asp?2020/13/4/326/290168




  Introduction Top


Obesity may be defined as an abnormal growth of the adipose tissue due to an enlargement of fat cell size (hypertrophic obesity) or an increase in fat cell number (hyperplastic obesity) or a combination of both.[1] It may be assessed using different parameters such as body mass index (BMI), waist circumference (WC), and waist–hip ratio (WHR). Obesity is considered as one of the major noncommunicable diseases (NCDs) of the modern world. Some researchers called it as adiposity-based chronic disease.[2] According to the World Health Organization, obesity is one of the most common, yet among the most neglected, public health problem in both developed and developing countries.[3]

In 2016, more than 1.9 billion adults (39%), 18 years and older, were overweight, and among them, over 650 million (13%) were obese. The worldwide prevalence of obesity nearly tripled between 1975 and 2016.[4] The Asian Indians are known to have a lower BMI than Europeans.[5] However, for any given BMI, Asian Indians have greater WHR and abdominal fat than Europeans.[6],[7]

Central obesity or abdominal obesity which indicates intraabdominal adiposity is considered as a better parameter of obesity than BMI. It is defined by the International Diabetes Federation as WC 90 cm or more for males and 80 cm or more for females.[8] Central obesity is considered as a major risk factor for different NCDs such as cardiovascular diseases (CVD), hypertension, type 2 diabetes mellitus, dyslipidemia, different musculoskeletal and metabolic diseases, and few cancers. Central obesity is one of the most important factors of metabolic syndrome (MetS).[8]

With this background, this study was conducted to find the prevalence and predictors of central obesity and its bidirectional relationship with serum lipids among adults aged 18–49 years in an area of Kolkata.


  Materials and Methods Top


This study was a part of a large community-based observational cross-sectional study conducted among 388 permanent residents aged 18–49 years of the ward number: 66 under Kolkata Municipal Corporation of West Bengal over 2 years from November 2016 to October 2018.

Sample size calculation

Central obesity is a component of MetS. Based on the prevalence of MetS among adults aged between 20 and 40 years in a previous study which was 20.61%,[9] sample size was calculated using the formula given below:

N = (Zα/2) 2 pq/L2

[where N = sample size

Zα/2 for confidence level 95% =1.96;

p = Prevalence = 20.61; q = (100-p) = 79.39

suppose relative error is 20%

L = 20% of P = 20/100 × 20.61 = 4.122]

N = {(1.96) 2 × 20.61 × 79.39}/(4.122) 2

N = 369.947 ≈ 370

We assumed nonresponse rate would be 10% of N, i.e., (10/100 × 370) =37

Hence, total sample size calculated was = 370 + 37 = 407 [Figure 1].
Figure 1: Sampling technique

Click here to view


Inclusion criteria

  1. Adults aged 18–49 years living in ward no: 66 permanently for the past 6 months were included.


Exclusion criteria

  1. Critically ill patients
  2. Pregnant/lactating women
  3. Those who did not give informed written consent.


Study tools

  1. A predesigned, pretested, structured schedule in Bengali (Local language)
  2. Sphygmomanometer and stethoscope
  3. Nonstretchable measuring tape and portable analog weighing machine
  4. Cotton, vial, disposable syringe, tourniquet, and spirit.


Study technique

Informed written consent of the participants was obtained before starting the interview. After the interview was over, the participant was requested to attend a camp organized in a nearby club on the following Sunday for blood test in empty stomach and for anthropometric measurement. Blood collection was done for fasting blood sugar (FBS) and lipid profile following standard operating procedure (SOP) after 12 h of overnight fasting. Anthropometric and blood pressure measurements were done in accordance with SOP after allowing the participants to take rest for at least 10 min. Two days/week were spent for data collection, and 5–6 participants were interviewed each day.

Operational definition

  1. Central obesity: WC ≥90 cm (male) and WC ≥80 cm (female) (for Asian)[8]
  2. Physical activity was classified by the International physical activity questionnaire-short form[10]
  3. Diet score: It has ten components [Table 1]
  4. Table 1: Diet score

    Click here to view


    Diet score was kept linear during analysis. Higher score indicated satisfactory diet.

    (During interview, the participants were explained how to measure the consumption of salt, oil, and sugar. The average monthly consumptions of salt, oil, and sugar were divided by 30 days to obtain the daily consumption of the family, and then the daily consumption was divided by the number of family members [where anyone above 12 years was considered as one unit family member, and those below 12 years were considered as half unit. Children aged 1 year or less were considered as zero units]).

  5. Hypertension: Systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg or on treatment for hypertension[12]
  6. Diabetes: On medication for diabetes or FBS ≥126 mg/dl[13]
  7. Dyslipidemia: On medication for dyslipidemia or any one or more of the followings:[14]


    • Total cholesterol ≥200 mg/dl
    • Low-density lipoprotein ≥130 mg/dl
    • High-density lipoprotein-cholesterol (HDL-C) <40 mg/dl (male) and <50 mg/dl (female)
    • Triglyceride (TG) ≥150 mg/dl


Ethical clearance was obtained from the institutional ethical committee. Data were analyzed using Statistical Package for the Social Sciences (SPSS Inc., Chicago, IL, Version 16.0 for Windows) software, and descriptive statistics were calculated as frequency and percentage. As the data were not distributed normally, median (interquartile range [IQR]) was calculated instead of the mean (standard deviation). Univariate and multivariable binary logistic regression analyses were performed to find the strength of association between central obesity and different risk factors, and P < 0.05 was considered statistically significant throughout the analysis.


  Results Top


In this study, we have found that more than half of the study participants were male (54.6%), and most of the participants (43.0%) belonged to 29–39 years age group, followed by 40–49 years age group (39.2%) [Table 2]. The median (IQR) age of the participants was 36 (32–44) years. Majority of the study participants were married (71.6%), and 24.2% had per capita income (PCI) more than 8750 rupees. Almost half of the participants were engaged in moderate physical activity and 1/3rd in the low physical activity. The median (IQR) of diet score was 4.0 (3.0–5.0), and more than 2/3rd of the participants had diet score of 4 or less. About 82% of the study participants were suffering from dyslipidemia and 32.7% from hypertension, whereas only 19.6% had diabetes. WC was high for 66% of the study participants [Table 3]. During univariate binary logistic regression to find the correlates of high WC (central obesity), it was found that female (odds ratio [OR] =3.52) in comparison to male, married individuals (OR = 2.98) in comparison to others, individuals with PCI more than 8750 rupees in comparison to others (OR: 2.13), individuals with low (OR = 3.97) and moderate (OR = 4.42) physical activities in comparison to those with high physical activity, participants with diabetes (OR = 2.35) and hypertension (OR = 2.76) in comparison to those with normal status, participants with decreasing HDL-C (OR = 1.03) and diet score (OR = 1.14), and increasing age (OR = 1.04) were at higher risk of having central obesity and that was statistically significant.
Table 2: Background characteristics of the study participants (n=388)

Click here to view
Table 3: High waist circumference and its predictors: Univariate and multivariable logistic regressions (n=388)*

Click here to view


During multivariable logistic regression by ENTER method to build the final model of risk prediction of central obesity, it was found that there were augmentations of factors such as female gender (adjusted OR [AOR] = 5.99 vs. OR = 3.52), married individuals (AOR = 3.06 vs. OR = 2.98), PCI > 8750 rupees (AOR = 3.04 vs. OR = 2.13), poor diet (AOR = 1.21 vs. OR = 1.14), low (AOR = 4.78 vs. OR = 3.97) and moderate (AOR = 5.18 vs. OR = 4.42) physical activities, decreasing HDL-C (AOR = 1.04 vs. OR = 1.03), and attenuation of only one factor, i.e., hypertension (AOR = 2.18 vs. OR = 2.76) when adjusted with all other independent variables, and all of those associations were statistically significant. Increasing age and presence of diabetes lost their statistical significance when adjusted with other independent variables during multivariable logistic regression.

The goodness of fit of the model was judged by the Hosmer and Lemeshow test which was not significant (0.451) and Omnibus test of model coefficient which was significant (0.000). This model can define the outcome variable, i.e., central obesity here by 36.8% (Nagelkerke R2 = 0.368).

There was a statistically significant difference in serum TG across quintiles of WC (P for linear trend 0.002) and WC across quintiles of TG (P for linear trend 0.011) [Figure 2]]. With increase in the level of TGs, there is proportional increase in WC, and with increase in WC, there is also proportional increase in the level of TGs at different levels, and the relationship is statistically significant also. This implies that there exists a bidirectional association between the two [Figure 2]. There was a statistically significant difference in WC across quintiles of serum HDL-C (P for linear trend 0.005) and serum HDL-C across quintiles of WC (P for linear trend < 0.001). With increase in WC, there is reduction in the level of HDL-C, and with increase in the level HDL-C, there is reduction in WC, and this relationship is also statistically significant. Thus, we can conclude that there exists a bidirectional association between the two [Figure 3].
Figure 2: Bidirectional relationship between waist circumference and mean serum triglycerides. There is a statistically significant difference in serum TG across quintiles of waist circumference (P for linear trend 0.002) and waist circumference across quintiles of TG (P for linear trend 0.011)

Click here to view
Figure 3: Bidirectional relationship between waist circumference and mean high-density lipoprotein-cholesterol. There is a statistically significant difference in waist circumference across quintiles of serum high-density lipoprotein-cholesterol (P for linear trend 0.005) and serum high-density lipoprotein-cholesterol across quintiles of waist circumference (P for linear trend < 0.001)

Click here to view



  Discussion Top


In this study, 66% of the study participants had central obesity. It almost matches with the findings of a study conducted by Undavalli et al.[15] in Andhra Pradesh, India, where the prevalence was 71.2%. In Indian Council of Medical Research (ICMR) INDIAB-3 study,[16] the prevalence of abdominal obesity in urban area of Chandigarh, Tamil Nadu, Jharkhand, and Maharashtra was 46.6%, 37.4%, 37.2%, and 26.7%, respectively. This difference may be due to different study population.

During multivariable logistic regression analysis, female gender, married status, high PCI, unhealthy diet, low physical activity, decreasing HDL-C, and hypertension were found to be the predictors of central obesity after adjusting with other variables. Dabou et al.[17] in their study in Cameroon had found that the prevalence of obesity and central obesity appeared to be significantly higher in women than in men. Men appeared to be less affected by obesity (OR = 0.09, 95% P = 0.0000) and central obesity (OR = 0.1, P = 0.0000), compared to women, just like the current study. In ICMR INDIAB-3 study,[16] it was observed that during multiple logistic regression analysis, the odds of having abdominal obesity were significantly associated with increase in age, female gender, hypertension, diabetes, high socioeconomic status, and with physical inactivity. In our study also, we had a similar type of findings except for diabetes and increasing age. Moreover, in our study, decreasing HDL-C was come out to be an independent risk factor for central obesity.

Girdhar et al.[18] in their study in North India had found that during logistic regression analysis, the odds of having obesity was 3.80 in 50–60 years age group and 2.67 and 1.68 among 40–49 and 30–39 years age groups in comparison to 20–29 years age group, though the last one was not statistically significant, it meant that it was increasing with age just like the current study, though during multivariable regression, it lost its statistical significance in our study. Homemakers with odds of 2.21 (not statistically significant) and married individuals with odds of 5.08 were more vulnerable for obesity just like the present study. Individuals with high (OR = 4.64) and high-middle (OR = 1.99) socioeconomic status were at more risk for obesity than individuals with low and low-middle background just like our study.

Mondal and Mukhopadhyay[19] in a study had found insignificant correlation between serum lipids and WC. However, in the current study, we have found significant bidirectional relationship between WC and serum lipids (TGs and HDL-C). In this study, we have found that with an increase in the level of TG, WC is also increasing, and the reverse is also true that with increase in WC, TG level is increasing. The same is true for HDL-C and WC also, here with decrease in HDL-C, WC is increasing, and with increase in HDL-C, WC is decreasing. We all know that there are a number of common risk factors for high WC (central obesity) and dyslipidemia (high TG and low HDL-C), for example, sedentary lifestyle, poor dietary pattern, and stress, and both of these entities (i.e., central obesity and dyslipidemia) may lead to dire consequences such as CVD in future. Patients who have one disease state should be encouraged to be vigilant about the signs/symptoms of the other disease state. There may be a number of physiological mechanisms for this bidirectional relationship, but further research is needed.

Limitation

Temporal relationship between the risk factors and central obesity could not be assessed by this study, as it was cross-sectional in design. Information regarding addiction, diet, physical activity, and income was collected using a schedule. Hence, a chance of recall bias was present in the study. In spite of all precautions during data collection, there always remained a scope for deliberate fabrication by the respondents on certain information regarding diet, addiction, income, and physical activity.


  Conclusion Top


In spite of the fact that obesity is a proximate predictor of many NCD including cancer, it is still a neglected public health issue. Mass awareness is still lacking regarding obesity and its consequences. Primordial and primary levels of prevention in the form of very strong, effective, and heart-reaching behavioral change communication are required at individual, at-risk, family, and community levels to generate awareness regarding control of obesity.

Apparently, WC, when considered for assessment of obesity, proves to be faster, more feasible, and more convenient than other methods of obesity assessment. TG level and HDL-C seemed to have bidirectional relationship with WC, and this finding generates further intensive research to establish this relationship to the extent that high WC may be considered as surrogate indicator for dyslipidemia (especially abnormal HDL-C and TGs).

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Häger A. Adipose tissue cellularity in childhood in relation to the development of obesity. Br Med Bull 1981;37:287-90.  Back to cited text no. 1
    
2.
Mechanick JI, Hurley DL, Garvey WT. Adiposity-based chronic disease as a new diagnostic term: the American Association of Clinical endocrinologists and American college of endocrinology position statement. Endocr Pract 2017;23:372-8.  Back to cited text no. 2
    
3.
Obesity: Preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser 2000;894:i-xii, 1-253.  Back to cited text no. 3
    
4.
World Health Organization. Obesity and Overweight. World Health Organization; February, 2018. Available from: http://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight. [Last accessed on 2019 Mar 28].  Back to cited text no. 4
    
5.
WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004;363:157-63.  Back to cited text no. 5
    
6.
Raji A, Seely EW, Arky RA, Simonson DC. Body fat distribution and insulin resistance in healthy Asian Indians and Caucasians. J Clin Endocrinol Metab 2001;86:5366-71.  Back to cited text no. 6
    
7.
Park YW, Zhu S, Palaniappan L, Heshka S, Carnethon MR, Heymsfield SB. The metabolic syndrome: Prevalence and associated risk factor findings in the US population from the third national health and nutrition examination survey, 1988-1994. Arch Intern Med 2003;163:427-36.  Back to cited text no. 7
    
8.
Balkau B, Charles MA. Comment on the provisional report from the WHO consultation. European group for the study of insulin resistance (EGIR) Diabet Med 1999;16:442-3.  Back to cited text no. 8
    
9.
Sawant A, Mankeshwar R, Shah S, Raghavan R, Dhongde G, Raje H, et al. Prevalence of metabolic syndrome in urban India. Cholesterol 2011;2011:1-7.  Back to cited text no. 9
    
10.
International Physical Activity Questionnaire; August, 2002. Available from: http://uacc.arizona.edu/sites/default/files/ipaq_english_telephone_short.pdf. [Last accessed on 2018 Jan 30].  Back to cited text no. 10
    
11.
Harikrishnan S, Sarma S, Sanjay G, Jeemon P, Krishnan MN, Venugopal K, et al. Prevalence of metabolic syndrome and its risk factors in Kerala, South India: Analysis of a community based cross-sectional study. PLoS One 2018;13:e0192372.  Back to cited text no. 11
    
12.
Armstrong C; Joint National Committee. JNC8 guidelines for the management of hypertension in adults. Am Fam Physician 2014;90:503-4.  Back to cited text no. 12
    
13.
Report of a WHO/IDF Consultation 2006. Definition and Diagnosis of Diabetes Mellitus and Intermediate Hyperglycemia. Summary of Technical Report and Recommendations; 2006. Available from: https://www.who.int/diabetes/publications/Definition%20and%20diagnosis%20of%20diabetes_new.pdf. [Last accessed on 2017 Sep 08].  Back to cited text no. 13
    
14.
Jellinger PS, Handelsman Y, Rosenblit PD, Bloomgarden ZT, Fonseca VA, Garber AJ, et al. American association of clinical endocrinologists and American college of endocrinology guidelines for management of dyslipidemia and prevention of cardiovascular disease. Endocr Pract 2017;23:1-87.  Back to cited text no. 14
    
15.
Undavalli VK, Ponnaganti SC, Narni H. Prevalence of generalized and abdominal obesity: India's big problem. Int J Community Med Public Health 2018;5:1311-6.  Back to cited text no. 15
    
16.
Pradeepa R, Anjana RM, Joshi SR, Bhansali A, Deepa M, Joshi PP, et al. Prevalence of generalized and abdominal obesity in urban and rural India – The ICMR-INDIAB study (Phase-I) [ICMR INDIAB-3]. Indian J Med Res 2015;142:139-50.  Back to cited text no. 16
[PUBMED]  [Full text]  
17.
Dabou S, Telefo PB, Sama LF. Evaluation of dietary habits and lifestyle on the prevalence of metabolic syndrome and obesity in undergraduate university students in Cameroon: A cross sectional study. J Metab Syndr 2018;7:236.  Back to cited text no. 17
    
18.
Girdhar S, Sharma S, Chaudhary A, Bansal P, Satija M. An epidemiological study of overweight and obesity among women in an urban area of North India. Indian J Community Med 2016;41:154-7.  Back to cited text no. 18
[PUBMED]  [Full text]  
19.
Mondal S, Mukhopadhyay SK. Effect of central obesity on lipid profile in healthy young adults. Med J DY Patil Vidyapeeth 2018;11:152-7.  Back to cited text no. 19
  [Full text]  


    Figures

  [Figure 1], [Figure 2], [Figure 3]
 
 
    Tables

  [Table 1], [Table 2], [Table 3]



 

Top
   
 
  Search
 
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

 
  In this article
   Abstract
  Introduction
   Materials and Me...
  Results
  Discussion
  Conclusion
   References
   Article Figures
   Article Tables

 Article Access Statistics
    Viewed256    
    Printed30    
    Emailed0    
    PDF Downloaded35    
    Comments [Add]    

Recommend this journal