| Abstract | Objectives: To assess the association between the ability to identify odors and anthropometric characteristics, as well as dietary preferences and diet composition in a large population-based sample.
Materials and Methods: Examinees from the city of Split, and two settlements on the Island of Korcula (Smokvica and Blato) were included in this cross-sectional study. Anthropometric indices included in the analyses were BMI, waist circumference, waist-to-hip ratio, brachial width and neck circumference. Dietary preferences were assessed for bacon, pork chops, broccoli, ice cream, milk chocolate, cakes, sweet foods, salty foods, bitter foods, spicy foods, vegetables and fruit on a Likert scale. Mediterranean dietary score was calculated to obtain the overall dietary pattern. The statistical analysis was performed using a Chi-square test, Kruskal-Wallis test, ANOVA test, Spearman Rank correlation test, and finally, a multivariate binary logistic regression analysis, where the outcome was a BMI≥30 kg/m2, and covariates included age, gender, education, material status, chronic diseases, smoking, alcohol intake, physical activity, Mediterranean diet score and odor identification ability.
Results: There was a negative correlation between the ability of correct odour identification and age (Rho=-0.377, P<0.001), as well as with BMI (Rho=-0.140, P<0.001), waist circumference (Rho=-0.210, P<0.001), waist-to-hip ratio (Rho=-0.305, P<0.001), brachial width (Rho=-0.174, P<0.001) and neck circumference (Rho=-0.222, P<0.001). Additionally, all anthropometric characteristics were positively and significantly correlated with age. Results indicated a positive correlation between correct odor identification and the liking of bacon (Rho=0.070, P<0.002), ice-cream (Rho=0.079, P<0.001), milk chocolate (Rho=0.088, P<0.001), cakes (Rho=0.062, P=0.005), sweet foods (Rho=0.099, P<0.001), salty foods (Rho=0.149, P<0.001), spicy foods (Rho=0.079, P<0.001) and vegetables (Rho=0.051, P=0.021), whereas all these food preferences were negatively correlated with age. In a logistic regression model the association of obesity and odor identification abilities were attenuated (OR=1.03; 95% CI 0.97-1.09), while only age, gender, education, material status, the presence of chronic diseases and ex-smoking status remained to be associated with obesity.
Conclusion: We found that age and all of the anthropometric indices were associated with olfactory function in a bivariate analysis. But, after controlling for confounding factors, namely age and gender, this association for BMI disappeared. |
| Abstract (croatian) | Ciljevi: Utvrditi postoji li povezanost između sposobnosti prepoznavanja mirisa i antropometrijskih obilježja, kao i prehrambenih preferencija i sastava prehrane u velikom uzorku iz opće populacije.
Materijali i metode: U ovo presječno istraživanje uključeni su ispitanici iz grada Splita te dva naselja na otoku Korčuli (Smokvica i Blato). Antropometrijska obilježja uključivala su indeks tjelesne mase (ITM), opseg struka, omjer opsega struka i kukova, širina lakta i opseg vrata. Prehrambene preferencije su procijenjene za slaninu, svinjetinu, brokule, sladoled, mliječnu čokoladu, kolače, slatku hranu, slanu hranu, goru hranu, začinjenu hranu, povrće i voće koristeći Likertovu ljestvici. Indeks mediteranske prehrane izračunat je kako bi se dobio opći obrazac prehrane. Statistička analiza provedena je korištenjem hi-kvadrat testa, Kruskal-Wallisov testa, ANOVA testa, Spearman Rank korelacijskog testa i binarne logističke regresije, gdje je ishod bio BMI ≥30 kg/m2, a kovarijate su uključivale dob, spol, obrazovanje, materijalni status, kronične bolesti, pušenje, unos alkohola, tjelesnu aktivnost, mediteransku prehranu i sposobnost identifikacije mirisa.
Rezultati: Pronađena je negativna korelacija između sposobnosti točne identifikacije mirisa i dobi (R =-0,377; P <0,001), kao i ITM-a (R=-0,140; P<0,001), opsega struka (R =-0,210; P<0,001), omjera opsega struka i kukova (R=-0,305; P<0,001), širine lakta (R=-0,174; P<0,001) i opseg vrata (R=-0,222; P <0,001). Osim toga, sve antropometrijske karakteristike pozitivno su i značajno bile povezane s dobi. Rezultati su ukazali na pozitivnu korelaciju između sposobnosti točne identifikacije mirisa i preferencije za slaninu (R=0,070; P <0,002), sladoled (R=0,079;P <0,001), mliječnu čokoladu (R=0,088; P <0,001), kolače (R=0,062; P=0,005), slatku hranu (R=0,099; P<0,001), slanu hranu (R=0,149; P<0,001), ljutu hranu (R=0,079; P<0,001), povrće (R=0,051; P=0,021), dok su sve te preferencije bile negativno povezane s dobi. U multivarijatnom logističkom regresijskom modelu povezanost između sposobnosti točne identifikacije mirisa i pretilosti je atenuirana (OR=1,03, 95% CI 0,97-1,09), dok su samo dob, spol, obrazovanje, materijalni status, prisutnost kroničnih bolesti i status bivšeg pušača bili statistički značajno povezan s pretilosti.
Zaključak: Utvrđeno je da su dob i sva antropometrijska obilježja ispitanika bili povezani s osjetom mirisa u bivarijatnoj analizi. No, nakon što se isključio učinak varijabli zabune, posebice dobi i spola, ova je povezanost za ITM je nestala. |
