Background: A striking difference in fish consumption and lung cancer mortality (LCM) exists among populations worldwide. This study investigated the relation between fish consumption and LCM at the population level.
Methods: Sex-specific LCM data, mostly around 1993 and fish consumption data for 10 periods 1961-1994 in 36 countries were obtained from WHO and FAO, respectively.
Results: A significant inverse correlation exists between log fish consumption and LCM rate in 9 out of the 10 time periods (r = -0.34 to r = -0.46, P = 0.044 to P = 0.005). After adjusting for smoking and other confounders, log fish consumption (% of total energy [% E]) was inversely and significantly associated with LCM rate (per 100 000 per year) in all 10 time periods (beta = -26.3 to beta = -36.7; P = 0.0039 to P < 0.0001). The stratified analysis showed that this inverse relation was significant only in countries with above median level of smoking (>2437 cigarettes/adult/year) or animal fat minus fish fat consumption (22.4% E). An increase in fish consumption by 1% E was calculated to reduce mean male LCM rate of the populations examined in the age class of 45-74 years by 8.4%. In women, no significant relation between fish consumption and LCM could be established.
Conclusions: Fish consumption is associated with a reduced risk from LCM, but this possible protective effect is clear-cut only in men and in countries with high levels of cigarette smoking or animal fat consumption.