Dietary Habits and Diabetes Mellitus Prevalence in Men and Women: National Health Survey in Serbia ^†

Abstract

This study assessed the relationship between dietary habits and diabetes mellitus prevalence in Serbia. The consumption of dairy products (OR = 1.15, 95% CI = 1.03–1.30; p = 0.017) and fruits (OR = 1.31, 95% CI = 1.19–1.44; p < 0.001) was associated with diabetes mellitus in males, whereas the consumption of margarine still reached statistical significance (OR = 5.33, 95% CI = 1.00–28.72; p = 0.050). The use of margarine (OR = 4.69, 95% CI = 1.35–16.32; p = 0.015) and lard (OR = 2.38, 95% CI = 1.10–5.12; p = 0.027) was associated with a higher prevalence of diabetes mellitus in females. A lower prevalence of diabetes mellitus in females was associated with bread consumption, pertaining to half-white bread (p = 0.027) and integral bread (p = 0.011).

1. Introduction

Diabetes mellitus is a critical public health problem worldwide, and although it can be modified, it is a major risk factor for cardiovascular diseases, stroke, and cancer [1,2]. According to World Health Organization 2014 estimates, the prevalence of diabetes was 422 million, with 1.5 million deaths directly caused by diabetes [3]. In 2014, 8.5% of adults aged 18 years and older had diabetes worldwide, particularly in low- and middle-income countries. In addition to cardiovascular diseases and cancer, diabetes mellitus is now an emergence public health problem in Serbia [2,4].

The prevalence of diabetes mellitus is attributed to the growth and aging of the population, as well as risk behaviors such as overweight, harmful use of alcohol, smoking tobacco, insufficient physical activity, long-term stress exposure, hereditary factors, etc. [5]. Among the behavioral factors, dietary habits are of particular importance in the etiology of diabetes mellitus [5,6].

Numerous studies have shown that dietary habits may be associated with diabetes, but the results are inconsistent [6,7]. In addition, the relationship between diabetes and certain foods and dietary factors remains complex and unclear. This study assessed the relationship between dietary habits and diabetes mellitus prevalence in Serbia.

2. Materials and Methods

This research was designed as a secondary analysis of databases provided by the National Health Survey conducted in Serbia in 2013 [8]. Data for this cross-sectional study were obtained from the National Survey of the Population of Serbia in 2013, which was carried out by the Ministry of Health of Serbia and the Institute of Public Health of Serbia “Dr Milan Jovanović Batut”. The research was carried in the period from 7 October 2013 to 30 December 2013.The study population included adults ≥ 15 years old who were permanent residents in the Republic of Serbia. The 2013 National Health Survey was performed in line with the EUROSTAT recommendations for the European Health Interview Survey.

We performed univariable logistic regression, where the independent variable was the diet, and the results are presented as odds ratios (ORs) with 95% confidence intervals (CIs). Associations between diet and diabetes mellitus prevalence were evaluated separately among men and women. Statistical significance was considered when p < 0.050. All statistical analyses were performed using the Statistical Package for Social Sciences software (v. 20.0, SPSS Inc., Chicago, IL, USA).

3. Results

The study sample consisted of 14,623 participants, 6759 (46.2%) men and 7864 (53.8%) women (

Table 1. Characteristics of study respondents by age, Republic of Serbia, 2013.

		Males (n = 6759)	Females (n = 7864)	p
Variables		No. (%)	No. (%)
Age groups (years)
	15–24	843 (12.5)	879 (11.2)
	25–34	937 (13.9)	1024 (13.0)
	35–44	1068 (15.8)	1123 (14.3)
	45–54	1055 (15.6)	1273 (16.2)
	55–64	1328 (19.6)	1553 (19.7)
	65–74	875 (12.9)	1080 (13.7)
	75–84	589 (8.7)	796 (10.1)
	85+	64 (0.9)	136 (1.7)	0.000
Place of residence
	Urban	3372 (5.1)	4502 (57.2)
	Rural	3037 (44.9)	3362 (42.8)	0.008
Education level
	Low	1673 (24.8)	2954 (37.6)
	Middle	3972 (58.8)	3728 (47.4)
	High	1114 (16.5)	1182 (15.0)	0.000

Abbreviations: p, probability value according to univariate regression analysis.

). About half of study participants had an urban place of residence and a moderate education level.

The prevalence of diabetes mellitus was 8.4%, significantly more frequent in women (8.9%) than in men (8.0%), p = 0.049 (

Table 2. Prevalence of diabetes mellitus, by sex.

		Diabetes Mellitus
		Absent	Present
		No. (%)	No. (%)	p
Sex	Males	6199 (92.0)	536 (8.0)
	Females	7139 (91.1)	695 (8.9)	0.049
Total		13338 (91.6)	1231 (8.4)

Abbreviations: p, probability, χ²-test.

).

The consumption of dairy products (OR = 1.15, 95% CI = 1.03–1.30; p = 0.017) and fruits (OR = 1.31, 95% CI = 1.19–1.44; p < 0.001) was associated with diabetes mellitus in males, whereas the consumption of margarine still reached statistical significance (OR = 5.33, 95% CI = 1.00–28.72; p = 0.050) (

Table 3. The association of dietary habits with diabetes mellitus prevalence, in Serbia, by sex: National Health Survey.

Prevalence of Diabetes Mellitus
	Males				Females
		OR	(95% CI)	p Value	OR	(95% CI)	p Value
Dietary items
Dairy		1.15	(1.03–1.30)	0.017	1.07	(0.97–1.19)	0.190
Bread
	White	1.89	(0.66–5.38)	0.234	0.58	(0.21–1.59)	0.291
	Half-white	1.11	(0.38–3.21)	0.846	0.32	(0.12–0.88)	0.027
	Integral	0.43	(0.15–1.26)	0.125	0.27	(0.10–0.74)	0.011
	All types	1.56	(0.54–4.51)	0.413	0.41	(0.15–1.13)	0.085
	None	1.00 *			1.00 *
Fat
	Lard	2.00	(0.77–5.18)	0.153	2.38	(1.10–5.12)	0.027
	Margarine	5.33	(1.00–28.72)	0.050	4.69	(1.35–16.32)	0.015
	Vegetable oil	1.37	(0.54–3.49)	0.513	1.51	(0.71–3.20)	0.287
	None	1.00 *
Fish		1.00	(0.99–1.24)	0.866	0.96	(0.82–1.13)	0.599
Fruits		1.31	(1.19–1.44)	0.000	0.97	(0.88–1.03)	0.259
Vegetables		1.10	(0.99–1.24)	0.083	1.01	(0.91–1.12)	0.837

Abbreviations: OR, odds ratio; 95%CI, confidence interval; p, probability value according to univariate regression analysis; * reference category.

). The consumption of margarine was associated with a higher prevalence of diabetes mellitus in females (OR = 4.69, 95% CI = 1.35–16.32; p = 0.015), as well as the use of lard (OR = 2.38, 95% CI = 1.10–5.12; p = 0.027). A lower prevalence of diabetes mellitus in females was associated with bread consumption, including half-white bread (OR = 0.32, 95% CI = 0.12–0.88; p = 0.027) and integral bread (OR = 0.27, 95% CI = 0.10–0.74; p = 0.011). No association was found between vegetable and fish intakes and the prevalence of diabetes mellitus in either males or females.

4. Discussion

One of the main findings in this study was the high prevalence of diabetes mellitus in Serbia in both sexes. The consumption of dairy products and fruits was associated with diabetes mellitus in males, whereas the consumption of margarine still reached statistical significance. The consumption of lard and margarine was associated with a higher prevalence of diabetes mellitus in females. A lower prevalence of diabetes mellitus in females was associated with the consumption of half-white and integral bread. No association was found between vegetable and fish intakes and the prevalence of diabetes mellitus in either males or females.

The association between the consumption of food and diabetes mellitus has been found to vary in significance and magnitude across countries [9]. In Latin America, the per capita consumption of milk showed a strong positive correlation with the incidence of childhood type 1 diabetes mellitus [10]. Studies exploring the role of other dairy products as a heterogeneous food group (including specific low-fat and high-fat dairy foods, such as butter) in diabetes mellitus mortality have not yielded consistent results [11].

Our results are consistent with findings from previous meta-analyses investigating the associations of high-fat versus low-fat diets and fatty acid intakes with the incidence of diabetes mellitus [6]. Current dietary guidelines on the prevention of diabetes mellitus recommend a diet low in total fat and animal fat, and high in vegetable fat [12]. Moreover, dose–response relationships have not yet been examined for the majority of these associations. In addition, substantial differences according to sex could be attributed to different exposures to lifestyle-related risk factors, such as smoking habits and obesity.

This study has some limitations. In addition to the well-known shortcomings of cross-sectional studies, such as ‘ecological fallacy’, there was a lack of control of confounding variables, and some of the collinearity and correlations may not be linear.

5. Conclusions

Dietary habits are of considerable importance for the prevalence of diabetes mellitus in Serbia. Consequently, adherence to a healthy dietary pattern at the population and individual level should be encouraged. Further epidemiological analytical studies are needed to investigate possible causative associations.