Sugar and addiction: a review of the evidence
This article is a summary of a review paper published in the European Journal of Nutrition, titled ‘Sugar addiction: the state of the science,’ written by three British authors from Cambridge University working in the fields of psychiatry and metabolic science: Margaret Westwater, Paul Fletcher and Hisham Ziauddeen.
The idea that sugar is an addictive substance that creates neurobiological changes similar to those in drug addiction is popular. The authors of this paper reviewed human and animal neuroscience studies on sugar and compared it with studies on drug addiction to find out if sugar could indeed be acting like a drug that could contribute to overeating and obesity.
What is drug addiction?
The accepted model of drug addiction is described in the DSM-IV (Diagnostic and Statistical Manual of Mental Disorders) and is characterised as:
Is food addictive?
The notion of food addiction (FA) arose out of the drug addiction model because there are similarities, such as people reporting a loss of control, withdrawal and cravings for ‘problem foods’. The problem foods originally described in the 1950s were corn, milk, eggs and potatoes, but the modern view is highly processed foods rich in sugar and fat are most likely to be addictive, although only two studies have ever examined the addictive potential of foods. However, the drug model is not a neat fit for food and this ‘food is a drug’ model has been challenged. Animal studies demonstrate that while food and drug consumption share some common neurobiology, important differences remain.
Research into food addiction
The 25-itemYale Food Addiction Scale (YFAS), and the more recently updated YFAS 2.0, attempt to translate the DSM-IV substance dependence model to food, using the following criteria:
These criteria are applied to 5 food categories: sweets (e.g. ice cream), starches (French fries), salty snacks (pretzels), fatty foods (pizzas) and sugary drinks, and says FA is when these criteria are met and there is clinically significant impairment. FA is described as mild (2-3 symptoms), moderate (4-5 symptoms) or severe (6 or more). Preliminary validation of YFAS 2.0 estimates 15.8% of people meet criteria for FA and 11.9% met criteria for severe FA, and overweight or obese people reported more FA symptoms. This tool has been questioned as a tool to measure addiction as it does not clearly or rigorously address withdrawal symptoms and tolerance related to food.
Dietary carbohydrates or sugar was not significantly associated with food addiction scores, which does not suggest a strong effect of sugar in addictive-like eating in humans
Food addiction vs Binge Eating Disorder (BED)
There is significant overlap between FA and eating disorders, specifically BED and Bulimia Nervosa, which casts doubt on the existence of FA as a stand-alone phenotype. The authors say current measurement of FA through the YFAS yields significant heterogeneity. It may be that the YFAS is measuring components of other existing syndromes rather than FA as a distinct syndrome.
Sugar and addiction
Few studies have examined sugar addiction in humans and most evidence is from animal studies. It is a challenge to translate the results to humans as we don’t eat sugar by itself but rather in mixed foods. Animal research shows there are neurobiological differences between drugs and sugar, for example the dopaminergic response to sugar (and other foods) reduces after repeated exposure and with predictive cues (such as smells), while it does not with drugs. Addictive-like behaviours such as bingeing occur only in animal studies where sugar is given intermittently rather than ad-libitum and likely due to pleasant taste rather than neurobiological effects. In humans the research does not show increased dosage of sugar having increased potency and therefore increased addictive potential. In summary, addictive-like consumption of sugar is different to drug addiction in both neurobiology and behaviour.
If not addiction, then what?
While little evidence supports sugar as an addictive substance, it may be that sweetness or palatability is central to addictive-like eating. The glycemic index (GI) or glycemic load (GL) may play a role in the food addiction to sweet foods, although no mechanistic link has been suggested, and the authors suggest this as a future direction for research.
Perhaps aspects of sugar or sweet foods are a craving rather than an addiction. Cravings for palatable foods like chocolate differ from drug cravings in their intensity, frequency and duration. Food cravings are short lived and subside whereas drug cravings do not. It is also suggested that the combination of liking and restraint combine to make thoughts of liked foods (such as chocolate) more pre-occupying and this is experienced as a craving. Or put simply, not allowing yourself to eat something highly pleasurable makes you want it even more.
The authors say that most evidence relating to sugar addiction is limited to animal studies and “far from convincing”. They say the idea that sugar addiction is a primary causal mechanism for weight gain fails to capture the complexity of obesity and eating disorders, and may hamper more coordinated and appropriate responses.