Mone, I, Kraja B, Hyska G, Burazeri G. Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania (Original research). SEEJPH 2021, posted: 02 April 2021. DOI : 10.11576/seejph-4297 P a g e 1 | 9 ORIGINAL RESEARCH Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania Iris Mone1,2, Bledar Kraja1,2, Jolanda Hyska1, Genc Burazeri1,3 1 University of Medicine, Tirana, Albania; 2 University Hospital Center Mother Theresa, Tirana, Albania; 3 Department of International Health, School for Public Health and Primary Care (CAPHRI), Fac- ulty of Health, Medicine and Life Sciences, Maastricht University, Maastricht, The Netherlands. Corresponding author: Iris Mone, MD, PhD Address: Rr. “Dibrës”, No. 371, Tirana, Albania Telephone: +355692149301; E-mail: iris_mone@yahoo.com Acknowledgments: Genc Burazeri was a recipient of an Irma Milstein International Fellowship at the Hebrew University–Hadassah School of Public Health and Community Medicine, Jerusa- lem, Israel, which provided support for the study. Mone, I, Kraja B, Hyska G, Burazeri G. Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania (Original research). SEEJPH 2021, posted: 02 April 2021. DOI : 10.11576/seejph-4297 P a g e 2 | 9 Abstract Aim: In light of the controversial evidence regarding health effects of chocolate intake, we aimed to assess its association with body mass index (BMI) among adult individuals in Albania, a tran- sitional post-communist country in South Eastern Europe which has traditionally employed a Med- iterranean dietary pattern. Methods: A cross-sectional study was conducted in 2003-2006 involving a population-based sam- ple of 737 Tirana residents aged 35-74 years (469 men, 268 women; overall response: 70%). Of these, 565 individuals (373 men and 192 women) provided data on chocolate intake and anthro- pometrics (77% of the sample). A 105-item food frequency questionnaire, including chocolate consumption, was administered to all individuals. Nine categories were used to assess the average frequency of intake of each food item in the past 12 months. In the analysis, chocolate intake was dichotomized into: consumption of <1/month vs. ≥1/month. A physical examination included measurement of weight and height. Furthermore, information on socio-demographic characteris- tics and classical risk factors was collected. Multivariable-adjusted general linear model was used to calculate the mean BMI values by chocolate intake groupings. Results: Upon simultaneous adjustment for socio-demographic characteristics, classical risk fac- tors and nutritional factors, there was an inverse association between BMI and chocolate intake in both sexes (sex-pooled mean BMI: 26.1 among participants who consumed chocolate <1/month vs. 27.0 in those with an intake of ≥1/month; P<0.001). Conclusions: This study points to a beneficial effect of moderate chocolate intake on lowering BMI, which deserves further vigorous investigation and replication in prospective studies in Al- bania and other populations. Keywords: Albania, body mass index, chocolate, cross-sectional study, epidemiology. Mone, I, Kraja B, Hyska G, Burazeri G. Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania (Original research). SEEJPH 2021, posted: 02 April 2021. DOI : 10.11576/seejph-4297 P a g e 3 | 9 Introduction Chocolate is a typical sweet food that evokes ambivalent feelings: pleasure as the result of its taste, flavor and appearance, and concerns as the results of its content with high sugar and calorie (1). Therefore, previous epidemi- ological studies have observed healthy and unhealthy effects of chocolate intake. Several studies have reported a positive link between frequent chocolate intake and a lower body mass index (BMI) (2,3), or a re- duction in the risk of cardio-metabolic disor- ders (4) and diabetes (5). Furthermore, choc- olate consumption has been linked to benefi- cial effects on human health and diseases (6) including cardiovascular health (7,8) blood pressure and vascular function (9). However, a cross-sectional study reported a positive association between chocolate intake and BMI in a dose-response pattern (10). Hence, according to this report, the positive relationship between chocolate intake and a lower body mass was evident only among participants with preexisting serious obesity- related illness (10). Also, a meta-analysis of 10 observational studies concluded that the evidences on the association between choco- late intake and CVD risk need to be confirm in further studies (11). In light of the controversial evidence regard- ing health effects of chocolate intake, we aimed to assess its association with BMI among adult individuals of both sexes in Al- bania, a transitional post-communist country in South Eastern Europe which has tradition- ally employed a Mediterranean dietary pat- tern. Methods A cross-sectional study involving a repre- sentative sample of 35-74-year-old residents of Tirana, the Albanian capital, was con- ducted in 2003-2006 (12). Study population and sampling The sample consisted of an age-and-sex- stratified random sample from the adult pop- ulation of the Tirana municipality, as regis- tered in the Albanian census of April 2001. We sampled a total of 1200 individuals, 720 men and 480 women (12). Of the estimated 1046 subjects (644 men and 402 women) who met the eligibility criteria (12), 737 indi- viduals participated in the study (469 men, 268 women; overall response: 70%). Data collection A semi-quantitative food frequency question- naire (FFQ), consisting of 105 food items in- cluding chocolate consumption, was admin- istered to all individuals (13). Participants were asked to indicate how often, on average, they had eaten specified amounts of each food item in the past 12 months. Nine catego- ries were used to assess the average fre- quency of intake of each food item: <1/month, 1-3/month, 1/week, 2-4/week, 5- 6/week, 1/day, 2-3/day, 4-5/day, and >6/day. In the analysis, chocolate intake was dichot- omized into: consumption of <1/month vs. ≥1/month. Microdiet, Version 2 (Downlee Systems Limited, UK, 2005) was employed to calcu- late for each food item the daily calorie in- take. The respective values for all 105 food items were added up in order to get a sum- mary score for each participant (13) for the total daily calorie intake expressed in Kcal, protein, fat and carbohydrate (in the analysis, all expressed as percentage of total calorie in- take) and SFA, MUFA, PUFA and trans fatty acid intake (in the analysis, all expressed as g/daily calories*100). In addition, information on socio-demo- graphic characteristics (age and educational level) and classical risk factors (physical ex- ercise and alcohol intake) was collected for each participant. Furthermore, a physical ex- amination included measurement of weight and height (based on which we calculated the Mone, I, Kraja B, Hyska G, Burazeri G. Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania (Original research). SEEJPH 2021, posted: 02 April 2021. DOI : 10.11576/seejph-4297 P a g e 4 | 9 body mass index [BMI]: kg/m2) and waist and hip circumferences (based on which we calculated the waist-to-hip ratio [W/H]) (12). The study was approved by the Albanian Committee of Medical Ethics. Participants gave written consent after being informed about the aims and procedures of the study. Statistical analysis The statistical analysis included 565 individ- uals (373 men and 192 women) for whom data on chocolate intake and anthropometric measurements were available (565/737=77% of the overall sample of study participants). General linear model was used to assess the association between chocolate consumption and socio-demographic characteristics (age and education) and behavioral factors (exer- cise, alcohol intake, BMI, W/H and nutri- ents), separately in men and women. Age-ad- justed mean values and their respective 95% confidence intervals (95%CIs) were calcu- lated for each covariate by the two categories of chocolate intake (<1/month vs. ≥1/month). Subsequently, multivariable-adjusted (foot- note to Table 2) mean BMI values and their respective 95%CIs were calculated by the two categories of chocolate intake (<1/month vs. ≥1/month), separately in men and women. SPSS (Statistical Package for Social Sci- ences, version 19.0), was used for all the sta- tistical analyses. Results Mean age was significantly higher among women who reported chocolate intake at least once per month compared with their counter- parts who consumed chocolate <1/month (56 years vs. 50 years, respectively; P<0.001) – a finding which was not evident in men (Table 1, upper panel). Mean educational level was not significantly different between the two groups distinguished by frequency of choco- late intake in either sex. A similar finding was evident for alcohol intake, notwithstanding the particularly low consumption of alcohol among women. Conversely, in both sexes, participants who reported chocolate con- sumption of ≥1/month were more physically active than individuals who reported a choc- olate intake of <1/month (in men: 201 kcal vs. 87 kcal, respectively, P<0.001; in women: 164 kcal vs. 95 kcal, respectively, P<0.001). Furthermore, mean BMI was considerably lower among participants who consumed chocolate ≥1/month than those who con- sumed chocolate <1/month (in men: 26 vs. 28, respectively, P<0.001; in women: 25 vs. 27, respectively; P<0.001). In men only, mean W/H was significantly lower among participants who reported a chocolate intake ≥1/month compared with individuals who consumed chocolate <1/month (0.93 vs. 0.95, respectively; P<0.001). In both sexes, the total calorie intake was sig- nificantly higher among individuals who con- sumed chocolate ≥1/month than those who consumed chocolate <1/month (Table 1, lower panel). In men only, a higher chocolate intake was related to a lower protein (as per- centage of total calories), whereas in women only a higher chocolate consumption was as- sociated with higher total fats. In men, a higher chocolate intake was related to higher total carbohydrates, whereas in women there was evidence of the opposite. In both sexes, PUFA level (g/daily calories) was signifi- cantly higher among participants who con- sumed chocolate ≥1/month than those who consumed chocolate <1/month. In women only, MUFA level was higher in participants with a higher chocolate consumption. On the other hand, in men only, there was evidence of a higher level of trans fatty acids in those who consumed chocolate ≥1/month com- pared to those who consumed chocolate <1/month. Mone, I, Kraja B, Hyska G, Burazeri G. Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania (Original research). SEEJPH 2021, posted: 02 April 2021. DOI : 10.11576/seejph-4297 P a g e 5 | 9 Table 1. Association of chocolate intake with socio-demographic characteristics, conven- tional risk factors and nutrient intake in a population-based sample of Albanian adults; age-adjusted mean values from general linear models Upper panel: Socio-demo- graphic and classical risk factors Men (N=373) Women (N=192) N* Mean 95%CI P N* Mean 95%CI P Age (years): <1/month ≥1/month 146 227 52.1 51.8 50.5-53.7 50.5-53.1 0.742 74 118 55.7 50.0 53.3-58.0 48.2-51.9 <0.001 Education (years): <1/month ≥1/month 146 226 11.8 11.2 11.3-12.4 10.7-11.6 0.060 72 118 10.9 10.6 10.1-11.8 9.9-11.2 0.514 Alcohol intake (drinks/week): <1/month ≥1/month 146 226 3.6 4.4 2.4-4.8 3.4-5.4 0.320 73 118 0.8 0.6 0.2-1.3 0.2-1.1 0.657 Physical exercise (kcal): <1/month ≥1/month 146 225 86.8 201.5 69.5-104.1 187.5-215.4 <0.001 71 117 94.7 164.4 76.1-113.4 150.0-178.7 <0.001 BMI: <1/month ≥1/month 146 226 28.0 25.8 27.5-28.5 25.4-26.2 <0.001 74 118 27.1 24.7 26.3-27.9 24.0-25.3 <0.001 W/H: <1/month ≥1/month 146 226 0.95 0.93 0.94-0.95 0.92-0.94 <0.001 74 118 0.87 0.86 0.86-0.88 0.85-0.87 0.139 Lower panel: Nutrients Men Women N* Mean 95%CI P N Mean 95%CI P Total calorie intake (kcal): <1/month ≥1/month 146 227 2909 3186 2824-2996 3117-3255 <0.001 74 118 2431 2711 2333-2529 2634-2788 <0.001 Total proteins (% of calories): <1/month ≥1/month 146 227 17.9 16.8 17.6-18.2 16.6-17.1 <0.001 74 118 17.7 17.6 17.4-18.0 17.4-17.9 0.670 Total fats (% of calories): <1/month ≥1/month 147 227 35.1 35.5 34.5-35.6 35.0-35.9 0.281 74 118 38.3 39.8 37.6-39.0 39.3-40.4 0.001 Total carbohydrates (% of calo- ries): <1/month ≥1/month 146 227 47.4 48.2 46.8-48.1 47.7-48.7 0.053 74 118 46.1 44.7 45.3-46.9 44.1-45.4 0.013 SFA (g/daily calories*100): <1/month ≥1/month 146 227 1.34 1.32 1.32-1.36 1.30-1.34 0.231 74 118 1.43 1.47 1.40-1.46 1.45-1.49 0.038 MUFA (g/daily calories*100): <1/month ≥1/month 146 227 1.48 1.50 1.45-1.51 1.48-1.52 0.395 74 118 1.65 1.71 1.60-1.69 1.68-1.75 0.018 PUFA (g/daily calories*100): <1/month ≥1/month 146 227 0.76 0.83 0.74-0.78 0.81-0.85 <0.001 74 118 0.86 0.94 0.83-0.90 0.91-0.96 0.002 Trans (g/daily calories*100): <1/month ≥1/month 146 227 0.022 0.024 0.021-0.023 0.023-0.025 0.003 74 118 0.025 0.025 0.024-0.027 0.023-0.026 0.378 * Discrepancies in the totals are due to missing covariate values. Mone, I, Kraja B, Hyska G, Burazeri G. Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania (Original research). SEEJPH 2021, posted: 02 April 2021. DOI : 10.11576/seejph-4297 P a g e 6 | 9 In crude/unadjusted models (Table 2, model 1), mean BMI was substantially lower among participants who consumed chocolate ≥1/month compared to those who consumed chocolate <1/month (sex-pooled mean BMI: 25.3 vs. 27.5, respectively; P<0.001). Adjust- ment for age (model 2) did not affect the find- ings (sex-pooled mean BMI: 25.2 vs. 27.5, respectively; P<0.001). Table 2. Association of chocolate intake with BMI; unadjusted and multivariable-adjusted mean BMI values by sex Model Men (N=373) Women (N=192) Overall (N=565)* Mean 95%CI P Mean 95%CI P Mean 95%CI P Model 1†: <1/month ≥1/month 27.98 25.79 27.47- 28.49 25.38- 26.21 <0.001 26.94 24.77 26.12- 27.76 24.12- 25.42 <0.001 27.46 25.28 27.02- 27.91 24.92- 25.64 <0.001 Model 2‡: <1/month ≥1/month 27.99 25.79 27.48- 28.50 25.38- 26.20 <0.001 27.10 24.67 26.27- 27.94 24.02- 25.33 <0.001 27.52 25.23 27.08- 27.97 24.89- 25.61 <0.001 Model 3¶: <1/month ≥1/month 27.00 26.41 26.58- 27.42 26.08- 26.74 0.046 26.47 24.86 25.83- 27.11 24.39- 25.34 <0.001 27.04 26.13 26.66- 27.41 25.84- 26.43 <0.001 * Adjusted for sex. † Model 1: crude/unadjusted models. ‡ Model 2: adjusted for age. ¶ Model 3: adjusted for age, education, exercise, alcohol intake and W/H, total calorie intake, protein, fat and carbohy- drate (all expressed as percentage of total calorie intake) and SFA, MUFA, PUFA and trans fatty acid intake (all expressed as g/daily calories*100). Upon simultaneous adjustment for all covari- ates (model 3), the inverse association be- tween BMI and chocolate intake was attenu- ated but nevertheless remained statistically significant in both sexes (sex-pooled mean BMI: 26.1 vs. 27.0, respectively; P<0.001). Discussion We found a strong inverse relationship be- tween frequency of chocolate intake and BMI, which was consistent in both sexes and persisted upon adjustment for a wide array of socio-demographic characteristics and be- havioral factors including nutrient intake as assessed by a detailed FFQ. The results of the present study are compara- ble with a previous study conducted by Golomb et al. (2), which examined the cross- sectional relationship of chocolate intake and BMI among American adults. They reported that chocolate consumption frequency is linked to lower BMI in unadjusted model (P=0.008) and in adjusted models adding age, sex, activity, saturated fats, fruit and vegeta- Mone, I, Kraja B, Hyska G, Burazeri G. Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania (Original research). SEEJPH 2021, posted: 02 April 2021. DOI : 10.11576/seejph-4297 P a g e 7 | 9 bles, mood and calories (P=0.001). Also, an- other study found that high chocolate con- sumption was associated with lower BMI, body fat and waist circumference in young people regardless of different confounders (age, sex, total energy, saturated fats, fruit and vegetable, and physical activity) (3). One possible explanation for these findings is due to the fact that the caloric components as well as the other biologically active compo- nents of the food could influence BMI. Hence, the observed inverse association be- tween chocolate consumption and BMI may relate to the effects of other biologically ac- tive components of chocolate such as fla- vanols including catechin, epicatechin, and procyanidins which have a variety of benefi- cial physiologic actions (6). Flovanols promote the release of nitric oxide which has been shown to increase oxidation of fatty acids and glucose in skeletal muscle, inhibits fat synthesis in adipose tissue, and stimulates lipolysis in adipocytes (6). The Netherlands Cohort Study after 14 years of follow-up reported that women with the high- est dietary intake of total flavonols had sig- nificantly lower increases in BMI than women with the lowest intake, over time (0.41 and 0.91, respectively; P<0.05), sug- gesting a favorable effect of dietary flavanols intake on maintaining of body weight (14). Additionally, animal studies have shown that dietary flovanols intake may possibly reduce weight gain through effects of epicatechin and catechin on target tissues (15,16). Epicat- echin improves the mitochondrial content, structure and function as well as capillarity of skeletal muscle (15), whereas catechin in- creases energy expenditure, decreases fatty acid synthase levels in adipose tissue and in- hibits adipocyte differentiation (16). Another explanation for the observed find- ings may relate to the effect of chocolate con- sumption on appetite and satiety. Massolt et al. have shown that chocolate eating and smelling both could reduce appetite (17) whereas in a randomized, controlled study, Tey et al. demonstrated that chocolate con- sumption could decrease satiety (18). These findings suggest that chocolate consumption may aid for weight maintenance as a result of early termination of food intake. The main advantages of this study are its community-based design which included men and women from general population of Albania and use of a validated questionnaire for assessment of nutrient intake and physical activity. The FFQ we used for measurement of dietary patterns was customized to the Albanian con- text and previously validated in a small sam- ple of Albanian adults of both sexes. Seemingly, there is no plausible reason to as- sume differential reporting among partici- pants distinguished by socioeconomic char- acteristics or BMI groupings. Nevertheless, we cannot entirely exclude the possibility of information bias. Our study has other limita- tions. The response rate raises the possibility of selection bias. Male non-respondents, in addition to being older than participants, were more likely to be retired; however, ex- clusion of retirees from the analysis did not affect the magnitude of the association. In women, respondents and non-respondents were more similar. If non-response among thinner individuals was associated with higher chocolate intake, this could attenuate the findings. Conversely, if obese individuals who did not respond tended to employ a higher chocolate consumption, this could spuriously strengthen the findings, but would be unlikely to rule out the entire association. Also, importantly, the data included in this analysis are old. In conclusion, our findings point to a benefi- cial effect of moderate chocolate intake on lowering the body mass, which deserves fur- ther vigorous investigation and replication in Mone, I, Kraja B, Hyska G, Burazeri G. Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania (Original research). SEEJPH 2021, posted: 02 April 2021. DOI : 10.11576/seejph-4297 P a g e 8 | 9 prospective studies in Albania and other pop- ulations. Conflicts of interest: None declared. References 1. Roeline G. Kuijer, Jessica A. Boyce. Chocolate cake. Guilt or celebration? Associations with healthy eating atti- tudes, perceived behavioural control, intentions and weight-loss. Appetite 2014;74:48–54. 2. Golomb BA, Koperski S, White HL. Association between more frequent chocolate consumption and lower body mass index. Arch Intern Med 2012;172:519-21. 3. Magdalena Cuenca-Garc, Jonatan R. Ruiz, Francisco B. Ortega, Manuel J. Castillo. Association between choco- late consumption and fatness in Eu- ropean adolescents. Nutrition 2014;30:236–9. 4. Buitrago-Lopez A, Sanderson J, Johnson L, Warnakula S, Wood A, Di Angelantonio E, Franco OH. Chocolate consumption and cardi- ometabolic disorders: systematic re- view and meta-analysis. BMJ 2011;343:d4488. 5. Greenberg JA. Chocolate intake and diabetes risk. Clin Nutr 2015;34:129- 133. DOI: 10.1016/ j.clnu.2014.02.005. 6. Katz DL, Doughty K, Ali A. Cocoa and chocolate in human health and disease. Antioxid Redox Signal 2011;15:2779e811. 7. Corti R, Flammer AJ, Hollenberg NK, Lüscher TF. Cocoa and Cardio- vascular Health. Circulation 2009;119:1433-41. 8. Fernández-Murgaa, J.J. Tarínb, M.A. García-Perezc, A. Canoa. The impact of chocolate on cardiovascular health. Maturitas 2011;69:312–21. 9. Sudano I, Flammer AJ, Roas S, Enseleit F, Ruschitzka F, Corti R, Noll G. Cocoa, Blood Pressure, and Vascular Function. Curr Hypertens Rep 2012;14:279-84. 10. Greenberg JA, Brian Buijsse B. Ha- bitual Chocolate Consumption May Increase Body Weight in a Dose-Re- sponse Manner. PLoS ONE 2013;8:e70271. 11. Zhizhong Zhang, Gelin Xu, Xinfeng Liu. Chocolate intake reduces risk of cardiovascular disease: Evidence from 10 observational studies. Int J Cardiol 2013;168:5448–77. 12. Burazeri G, Kark JD. Hostility and acute coronary syndrome in a transi- tional post-communist Muslim coun- try: a population-based study in Ti- rana, Albania. Eur J Public Health 2011;21:469-76. 13. Mone I, Bulo A. Total fats, saturated fatty acids, processed foods and acute coronary syndrome in transi- tional Albania. Mat Soc Med. 2012;24:91-3. 14. Hughes LA, Arts IC, Ambergen T, Brants HA, Dagnelie PC, Goldbohm RA, et al. Higher dietary flavone, flavonol, and catechin intakes are as- sociated with less of an increase in BMI over time in women: a longitu- dinal analysis from the Netherlands Cohort Study. Am J Clin Nutr 2008;88:1341–52. 15. Nogueira L, Ramirez-Sanchez I, Per- kins GA, et al. (-)-Epicatechin en- hances fatigue resistance and oxida- tive capacity in mouse muscle. J Physiol 2011;589(pt 18):4615-31. 16. Wolfram S, Raederstorff D, Wang Y, et al. TEAVIGO (epigallocatechin Mone, I, Kraja B, Hyska G, Burazeri G. Chocolate intake is associated with a lower body mass index in adult men and women in transitional Albania (Original research). SEEJPH 2021, posted: 02 April 2021. DOI : 10.11576/seejph-4297 P a g e 9 | 9 © 2021 Mone et al; This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. gallate) supplementation prevents obesity in rodents by reducing adi- pose tissue mass. Ann Nutr Metab 2005;49:54–63. 17. Massolt ET, van Haard PM, Rehfeld JF, Posthuma EF, van der Veer E, and Schweitzer DH. Appetite sup- pression through smelling of dark chocolate correlates with changes in ghrelin in young women. Regul Pept 2010;161:81–6. 18. Siew Ling Tey, Rachel C Brown, Andrew R Gray, Alexandra W Chisholm, Conor M Delahunty. Long-term consumption of high en- ergy-dense snack foods on sensory- specific satiety and intake. Am J Clin Nutr 2012;95:1038–47. ____________________________________________________________________________