A new studybased on data from the Framingham Offspring Cohort found that levels of omega-3 fatty acids in the blood are as good a predictor of mortality from any cause as smoking.

The study was performed by the Hospital del Mar Medical Research Institute (IMIM), The Fatty Acid Research Institute (FARI), and several universities in the United States and Canada. The researchers found that omega-3 levels in red blood cells are very good mortality risk predictors, concluding that "Having higher levels of these acids in the blood, as a result of regularly including oily fish in the diet, increases life expectancy by almost five years,” while being a regular smoker takes 4.7 years off of life expectancy.

Related: Omega-3 Index, Smoking Equal Predictors of CVD Omega-3 Revisited: The Most Important Benefit You’ve Never Heard Of Carlson Takes Steps Towards Plastic Neutrality

The study analyzed data on blood fatty acid levels in 2,240 people over the age of 65, who were monitored for an average of eleven years. The aim was to validate which fatty acids function as good predictors of mortality. The results: Four types of fatty acids, including omega-3, fulfill this role. Two of the fatty acids are saturated fats, traditionally associated with cardiovascular risk, the levels of which cannot be modified by diet. “This reaffirms what we have been seeing lately," says Dr. Aleix Sala-Vila, a postdoctoral researcher in the IMIM's Cardiovascular Risk and Nutrition Research Group and author of the study, in a press release. "Not all saturated fatty acids are necessarily bad."

Dr. Sala-Vila suggested that these results may contribute to the personalization of dietary recommendations, based on blood concentration of different fatty acids. "What we have found is not insignificant,” said Dr. Sala-Vila. “It reinforces the idea that small changes in diet in the right direction can have a much more powerful effect than we think, and it is never too late or too early to make these changes.”

Next steps for the researchers include analyzing the same indicators in similar but non-European population groups, to find out how widely the results can be applied.