Liver Cancer Mortality and Food Consumption in Serbia, 1991-2010: An Ecological Study.

AIM This paper investigates the correlation between liver cancer mortality and consumption of food-groups in Serbia. METHODS We conducted an ecological study. The study comprised the population of the Republic of Serbia (about 7.5 million inhabitants) during the period 1991-2010. This ecological study included the data on food consumption per capita which were obtained by the Household Budget Survey and mortality data for liver cancer made available by the National Statistical Office. Linear trend model was used to assess a trend of age-adjusted liver cancer mortality rates (per 100,000 persons) that were calculated by the method of direct standardization using the World Standard Population. Pearson correlation was performed to examine the association between liver cancer mortality and per capita food consumption quantified with a correlation coefficient (r value). RESULTS In Serbia, over the past two decades a significantly decreasing trend of liver cancer mortality rates has been observed (p<0.001). Liver cancer mortality was significantly (p<0.01) positively correlated with animal fat, beef, wine and spirits intake (r=0.713, 0.631, 0.632 and 0.745, respectively). A weakly positive correlation between milk consumption and mortality from liver cancer (r=0.559, p<0.05) was found only among women. The strongest correlation was found between spirits consumption and liver cancer mortality rates in women (r=0.851, p<0.01). A negative correlation between coffee consumption and age-adjusted liver cancer mortality rates was found (r=0.516, p<0.05) only for the eldest men (aged 65 years or older). CONCLUSIONS Correlations between liver cancer and dietary habits were observed and further effort is needed in order to investigate a possible causative association, using epidemiological analytical studies.