Association of macronutrient intake patterns with being overweight in a population-based random sample of men in France

Abstract

Aim

Diet is considered an important modifiable factor in the overweight. The role of macronutrients in obesity has been examined in general in selected populations, but the results of these studies are mixed, depending on the potential confounders and adjustments for other macronutrients. For this reason, we examined the association between macronutrient intake patterns and being overweight in a population-based representative sample of middle-aged (55.1 ± 6.1 years) men (n = 966), using various adjustment modalities.

Methods

The study subjects kept 3-day food-intake records, and the standard cardiovascular risk factors were assessed. Weight, height and waist circumference (WC) were also measured.

Results

Carbohydrate intake was negatively associated and fat intake was positively associated with body mass index (BMI) and WC in regression models adjusted for energy intake and other factors, including age, smoking and physical activity. However, with mutual adjustments for other energy-yielding nutrients, the negative association of carbohydrate intake with WC remained significant, whereas the associations between fat intake and measures of obesity did not. Adjusted odds ratios (95% confidence interval) comparing the highest and lowest quartiles of carbohydrate intake were 0.50 (0.25–0.97) for obesity (BMI > 29.9) and 0.41 (0.23–0.73) for abdominal obesity (WC > 101.9 cm).

Conclusion

Consistent negative associations between carbohydrate intake and BMI and WC were seen in this random representative sample of the general male population. The associations between fat intake and these measures of being overweight were attenuated on adjusting for carbohydrate intake. Thus, the balance of carbohydrate-to-fat intake is an important element in obesity in a general male population, and should be highlighted in dietary guidelines.

Résumé

Objectif

L’alimentation est considérée comme un important facteur de risque modifiable de surcharge pondérale. Le rôle des macronutriments sur l’obésité a essentiellement été étudié dans des populations sélectionnées et l’hétérogénéité des résultats rapportés est en partie liée à la variabilité des facteurs de confusion pris en compte dans les analyses. Nous avons étudié l’association entre des comportements alimentaires basés sur l’équilibre en macronutriments et la surcharge pondérale dans une population aléatoire d’hommes (n = 966), âgés en moyenne de 55,1 ± 6,1 ans.

Méthodes

L’apport en nutriments a été estimé selon un enregistrement prospectif de l’alimentation de trois jours. Différentes mesures anthropométriques (taille, poids, tour de taille) ont permis de construire des indicateurs de surcharge pondérale : indice de masse corporelle (IMC) et tour de taille (TT).

Résultats

Après la prise en compte de l’apport énergétique, l’âge, les consommations de tabac et d’alcool et l’activité physique, la consommation de glucides était associée négativement et celle des lipides positivement à l’IMC et au TT. L’association entre les glucides et la surcharge pondérale restait significative après l’ajustement supplémentaire sur les autres macronutriments énergétiques. L’association avec les lipides n’était plus significative. Entre le quartile supérieur et le quartile inférieur de l’apport en glucides, les OR ajustés (intervalle de confiance à 95 %) sont de 0,50 [0,25–0,97] pour l’obésité (IMC > 29,9 km/m²) et 0,41 [0,23–0,73] pour l’obésité abdominale (TT > 101,9 cm).

Conclusion

Si la relation entre consommation de lipides et obésité était atténuée par un ajustement sur l’apport en glucides, la relation entre consommation d’hydrates de carbone et obésité persistait, quel que soit l’ajustement considéré. Pour un même apport et dépense énergétique, l’équilibre entre consommation de glucides et de lipides apparaît comme un paramètre important de la relation alimentation–obésité chez l’homme.

Introduction

The prevalence of people who are overweight or obese is increasing globally. Estimates indicate that approximately 30% of adults in the USA and as much as 10–25% adults in European countries are obese [1], [2]. Overweight and obesity are risk factors for chronic diseases such as diabetes, cardiovascular disease (CVD), certain cancers and joint disorders [3]. In surveys, being overweight or obese is often assessed by body mass index (BMI), while abdominal obesity is usually determined by waist circumference (WC). WC is considered a proxy for intraabdominal visceral and subcutaneous fat mass [4], and correlates strongly with diabetes and CVD [3], [5], [6].

It is widely recognized that being overweight or obese usually results from an imbalance between energy input and output. Although diet is a modifiable risk factor, the relationship between diet and obesity, and particularly the role of macronutrients, remains controversial. Over the past decade, research has focused on the macronutrient composition of diets and weight loss [7], [8], [9], and has suggested that the proportion of calories consumed as carbohydrates, protein and fats may be implicated in the aetiology and management of the overweight. The findings pertaining to the role of each of these macronutrients in weight loss, however, remain mixed [8], [9]. Similarly, the literature on the associations between carbohydrate, protein and lipid intakes, and indicators of overweight and obesity such as BMI and WC, in the general population is equivocal [10], [11], [12], [13]. This may be partly related to differences in study design and in the adjustments for potential confounders (such as lifestyle factors and medical history) in the statistical models used across studies.

To our knowledge, few large-scale population-based studies have assessed the independent contribution of various macronutrients to being overweight [13], [14], [15], [16] after accounting for several potential confounders such as age, education, and lifestyle factors such as smoking and physical activity, as well as a previous history of hypertension, diabetes or dyslipidaemia. Such findings could prove useful when formulating guidelines and recommendations for the general public as regards to carbohydrate, fat and protein intakes. Thus, our goal was to examine the association between energy-adjusted macronutrient intakes, using the residual method [17], and measures of overweight and obesity (BMI and WC) in a randomly selected population-based representative sample of men in France, after accounting for various individual characteristics and lifestyle factors.