Maths can make food ‘tastier and healthier’
New and “comprehensive” mathematical models and software that predicts the growth of microorganisms were developed as part of a recently completed Danish project focusing on salt reduction in seafood products.
The four-year research initiative, ‘Developing seafood products with improved health value, food quality and food safety’, concentrated on lowering the salt content of “lightly preserved” fish that has been smoked or marinated. The research was carried out within the Danish Green Development and Demonstration Program (GUDP) in close cooperation between food manufacturer Royal Greenland Seafood and the National Food Institute (DTU Food) at the Technical University of Denmark (DTU).
‘This is of great practical importance’
Researchers were able to develop algorithms that, they say, will enable food developers to cut salt while also compensating for the “lost effect of salt on preservation” through the “targeted change of other product properties”.
“This is of great practical importance in product development,” Professor Paw Dalgaard of DTU Food suggested. “In this way, the use of predictive models and software contributes to faster, better and cheaper product development, as it would be costly and time consuming to obtain the same information from product trials."
Within the GUDP project, DTU Food developed new predictive models for psychrotolerant Clostridium botulinum and Pseudomonas bacteria. These models are particularly relevant for seafood products, as salt is used in lightly preserved seafood products to prevent the growth of psychrotolerant Clostridium botulinum.
The researchers suggested that the new models can be used together with existing and popular predictive models for Listeria monocytogenes and lactic acid bacteria.
Professor Dalgaard also developed the Food Spoilage and Safety Predictor (FSSP) program, which contains mathematical models that can predict the growth of both pathogenic and spoilage microorganisms. More than 10,000 people from companies, institutions and authorities in more than 100 different countries currently use FSSP. Launched in 1999, it is freely available and version 4.5 is due to be introduced in 2018.
“We see predictive models and software as tools to contribute to effective product development,” Professor Dalgaard told FoodNavigator. “Clearly, our models and software is used together with product development techniques and… we believe the predictive models and software contributes to faster, better and cheaper product development, as it would be costly and time consuming to obtain the same information from product trials."
Optimising taste and improving health
Utilising the predictive software, product developers were also able to improve the taste profile of fish products, Professor Dalgaard noted.
“The relatively high salt content of many ‘old-fashioned‘ lightly preserved seafood products was justified by a need to reduce/inhibit growth of critical bacteria and often the salt content was higher than what was optimal from a sensory point of view. Therefore, by reducing salt the effect on both taste and healthiness of the newly developed products could be improved.”
Danish dietary advice suggests increasing fish consumption but also reducing the amount of salt people eat in order to reduce the risk of cardiovascular disease. According to the researchers, it is estimated that an average three gram reduction in salt consumption per day per person would result in an annual saving of €130-260m for Denmark’s health spending. Lowering salt content is therefore one of the claims associated with the Nordic Keyhole nutritional labelling scheme.
During the project period, seafood processor Royal Greenland developed 37 new lightly preserved seafood products with the Nordic Keyhole-label, including 25 smoked or marinated products.
In 2017, four new products of pasteurized lumpfish roe were launched with less than 3% salt compared to the normal salt content of approx. 4.5%, corresponding to a reduction of more than 30%.
The group also launched eight new Nordic Keyhole-labeled products of cold-water prawns in brine. The salt content of these newly developed brined products is below 1.5%, compared to 2-3% previously. This represents a reduction of 25-50% salt.
"In practice, the models mean that Royal Greenland has become much more innovative and effective in the development of recipes for low-salt seafood products without compromising either taste or food safety.”
Extending the approach
The FSSP software and the predictive models it uses could also be applicable for other product categories, Professor Dalgaard revealed.
“The approach that we have used for lightly preserved seafood will be equally valid for lightly preserved meat product including cured meat. In fact, our predictive food microbiology models for growth and growth boundary of Listeria monocytogenes and psychrotolerant lactic acid bacteria as included in the FSSP-software have already been validated for meat products in extensive validation studies.”
He also believes that the approach could be applicable for other reformulation efforts. “The approach we have used for salt reduction in lightly preserved seafood can definitely find application for other reformulation efforts.
"The developed predictive growth models include the effect of 9-12 different product characteristics and storage conditions. This allows the models to predict how reduced growth inhibition form lowering the content of one ingredients or additive can be compensated by changes in one or more likely several other product characteristics and storage conditions. As one example we have previously documented how the preserving effect of benzoic acid can be replaces by other factors in brined shrimps,” he noted.
