‘Food addiction’ may explain why patients overeat despite obesity-related disease

Obesity and its concomitant health problems — including diabetes, heart disease, osteoarthritis and certain types of cancer — are rampant worldwide. According to WHO, more than 500 million adults had obesity in 2008, and 2.8 million deaths each year can be attributed to the disease.

What drives people to continue overeating when the associated risks are so high? Perhaps the same thing that spurs a smoker to continue smoking or a drug addict to continue using.

Photo courtesy of Washington University School of Medicine.

“People can also have food addictions besides just cravings for certain foods,” Samuel Klein, MD, professor of medicine and nutritional science and director of the Center for Human Nutrition and of the Weight Management Program at Washington University School of Medicine in St. Louis, told Endocrine Today.

A complex mix of genes, neurons and environment are implicated. And increasingly, research indicates that some modern foods might be at least partially to blame. Much like drugs of abuse, such as methamphetamine and cocaine, highly palatable foods that are heavy in fat and sugar stimulate the reward centers in the brain, producing an addiction-like response in some people.

“The endogenous hormones and other homeostatic factors that regulate food intake and help maintain an overall energy balance can be completely overridden by external cues and environmental influences,” Klein said. “Reward mechanisms and learned social interactive behaviors are an important part of driving food intake, which make it very difficult for people to make long-term lifestyle changes in eating behaviors.”

Endocrine Today spoke to experts about the science of “food addiction” and its relationship to obesity. They address the parallels with drug abuse, which foods are the most dangerous, the hormonal drivers of appetite and how drug addiction research may play a role in treating food addiction and possibly preventing obesity.

Addicted to food

Food addiction is a controversial topic, with some researchers questioning whether the problem really exists.

“We know food addiction can be induced in rodent models, where behaviors and brain responses you would see with amphetamine and other drug addictions can be reproduced with sugar and certain foods,” Klein said.

Currently, the latest Diagnostic and Statistical Manual of Mental Disorders (DSM-5) lists no diagnosis for food addiction but does include “substance-related and addictive disorders” as a category. The Yale Food Addiction Scale, based on these criteria, was designed to assess food addiction in people. According to Klein, evidence of food addiction type of behavior can be found in as many as one-third of people with obesity.

“It’s not a medically accepted diagnosis, but it’s one that has relevance because it’s related to some of the same properties that define addiction of other substances, like drugs and alcohol, and even gambling addition,” Klein said.

Ashley Gearhardt, PhD, an assistant professor in the department of psychology at the University of Michigan, sees several parallels between these eating behaviors and drug abuse. “Food addiction at this point is more of an idea; we don’t know if it exists for sure,” she told Endocrine Today. However, there is a growing body of work that demonstrates that some highly processed foods “might highjack the [brain’s] reward system, so you’ll be really sensitive to any cues in the environment that tell you that these foods are available,” she said. “[You may] experience intense cravings or lose control of your eating when you start to consume some of these foods.”

Some researchers suggest that food addiction is a misnomer and that “eating addiction” better describes this addiction-like food consumption because the phrase encompasses the behavioral addiction to eating.

One cause of obesity epidemic

Regardless of the terminology, this type of disordered eating — and the food that fuels it — is a big driver of the current obesity epidemic. “Many food companies have learned what drives people to eat, and they take advantage of those behaviors to sell their products,” Klein said. “That begins with having foods that have a long shelf life and are in abundant supply, inexpensive and marketed by commercials designed to stimulate food purchase. ... You can’t eat just one chip.”

Highly processed foods, with their blend of fats and sugars, are cheap and easily accessible. Food manufacturers strip out nutrients in favor of fat and sugar, making these foods more appealing, according to Gearhardt.

“If you want to make a drug more addictive, you increase the dose of a really rewarding ingredient,” Gearhardt said. “Drugs of abuse become way more addicting when they rapidly hit the system. What we see is that in many foods, we have stripped out things like water and fiber and protein and refined them down so there’s a high dose of rewarding ingredients, like sugar, [that] are more rapidly absorbed into the system. They give you a much more rapid blood sugar spike.”

Furthermore, modern foods are extremely concentrated, according to Joseph Frascella, PhD, who is director of the division of clinical neuroscience at the NIH. For example, it takes eight oranges to make one glass of orange juice. Most people would be unable to eat eight oranges in one sitting, but nearly everyone could drink concentrated juice — and calories — in seconds. “It’s not like the satiety system can respond to say, ‘Whoa, you’ve consumed enough,’” he told Endocrine Today. “It’s already consumed.”

Additionally, these foods are reinforcing. “It’s like sitting down with a box of cookies and before you know it, a quarter of the box is gone,” Frascella said. “Where are those satiety signals? They’ve either been dampered or not paid attention to because they are overshadowed by a stimulus that is far more salient.”

Hunger loop a complex system

Food consumption is regulated by a complex system of gut–brain interactions, including hunger and satiety signals, all of which are propelled by the hypothalamus. Other key players in hunger control include the neurotransmitter neuropeptide Y (NPY), which communicates with the hunger hormones ghrelin and leptin. Insulin and glucagon-like peptide-1 also are important, according to Paul J. Kenny, PhD, professor and chair of Pharmacology and Systems Therapeutics at Mount Sinai Hospital in New York.

When this delicate balance is disturbed, the result can be disordered eating. In an extreme example, people with Prader-Willi syndrome, a complex neurodevelopmental disorder, have an overwhelming desire to eat and are prone to morbid obesity without strict food control, according to Jennifer Miller, MD, associate professor of pediatric endocrinology at the University of Florida in Gainesville.

“They never feel full and will eat anything they can get their hands on,” Miller said. “They’ll eat garbage. They’ll eat pet food. They’ll eat food off the floor ... but their primary drive is always carbohydrates.”

Although rare, there have been documented cases of congenital leptin deficiency. Without leptin, also known as the satiety signal, these patients also experience hyperphagia.

Neurons that drive appetite

Many specific neurons in the brain are involved in appetite control. Bradford B. Lowell, MD, PhD, a neuroendocrinologist and professor of medicine at Harvard Medical School, and colleagues found that a small group of neurons in the arcuate nucleus, located at the base of the brain, express agouti-related peptide (AgRP), which has been shown in other studies to play a critical role in controlling hunger.

Other animal studies have demonstrated that turning off those neurons induces starvation. “Which is what you’d predict because they have lost that key component that detects the fasted state and then drives one to eat when they are activated,” Lowell said.

Lowell’s team moved in a different direction. “We took mice that are in the fed state, so their AgRP neurons are naturally off and they have no reason to eat ... and just forced those neurons on, experimentally and acutely,” Lowell said. They found that very quickly, the mice began eating voraciously despite being full.

In addition, Lowell and his team recently discovered that the hunger-inducing neurons that activate AgRP are located in the paraventricular nucleus, an area previously thought to control satiety.

This type of research will help Lowell assemble a hunger wiring diagram, which he hopes will untangle the hunger feedback loop and eventually reveal optimal targets for pharmacologic intervention in obesity, he said.

“Just like you have a wiring diagram of your house or apartment, your brain has a wiring diagram,” Lowell said. “But unlike your house, it’s much more complicated. It’s been extremely difficult to figure out what the wiring diagram is because the brain contains millions of neurons that make billions of connections. It doesn’t look like a wiring diagram; it looks like a bowl of spaghetti.”

Established effect on the brain

Highly rewarding foods have a demonstrated effect on the brain, especially the brain’s reward center, according to Kenny. Just like alcohol and drugs, highly palatable foods activate the reward system in the brain; continued compulsive use leads to a decrease in the reward function. “That decrease in reward function is probably contributing to the development of compulsive drug use,” he said.

In his animal research, Kenny found the same thing occurs in rats that have been fed highly palatable foods; they, too, developed deficits in their reward systems. Kenny and colleagues fed rats standard food and, in addition, provided them with high-calorie, cafeteria-style options. Over time, these animals required increasing amounts of palatable food to satisfy their craving.

To assess compulsive-like eating behavior, the researchers warned the rats that there would be punishment — a painful foot shock — if they continued to eat. Only the obese animals displayed compulsive-like eating; they continued to eat despite the threat of punishment, according to Kenny.

“We take that as a hallmark of some kind of emergence of compulsive drug use or at least motivation to obtain food that has reached a level that they’re discounting lots of important stuff in the environment,” Kenny said.

The underlying mechanism of this reward system deficiency seems to be similar to that seen in drug addicts as well: a decrease in dopamine D2 receptors. Kenny and colleagues reduced D2 levels in the rats and found that this hastened the development of reward deficits and compulsive food seeking.

Dopamine plays an equally important role in driving overeating in people, according to research by Gene-Jack Wang, MD, a senior clinician and clinical director of the Laboratory of Neuroimaging at the National Institute on Alcohol Abuse and Alcoholism, and colleagues.

In one study, Wang and colleagues used PET to evaluate dopamine levels in food-deprived healthy volunteers. Participants were allowed to see and smell highly palatable foods, called “cue stimulation,” but were not allowed to eat them. The results showed that dopamine levels rose just by seeing and smelling enticing food.

“That’s why we think that if obese persons are under the cue stimulation that it might induce their desire for food,” Wang told Endocrine Today.

Wang also found that people with morbid obesity have lower levels of dopamine D2 receptors compared with people of normal body weight. “Overeating is related to dopamine,” he said.

It is unclear whether people are born with lower levels of dopamine D2 receptors or whether consistent overeating leads to a drop in the levels, according to Wang. “It may not be only one thing; it could be both things,” he said.

Low dopamine levels may be difficult to overcome. Wang studied the effects of stopping drug use in methamphetamine patients. After 6 months of no drug use, their dopamine levels did not change. “That is not good news,” he said.

Genetic basis of food cravings

There are genetic reasons for craving high-fat foods, according to research presented at ObesityWeek 2015. Goldstoneand colleagues found that when they looked at images of high-calorie foods, adults with a variant near the FTO gene had more activation in the orbitofrontal cortex region of the brain.

Using functional MRI scans, the researchers analyzed the interaction of an FTO-associated single nucleotide polymorphism (SNP) and dopamine receptor D2 SNP on the anticipation of food reward. In addition, the researchers measured blood oxygen level dependent (BOLD) signaling in the nucleus accumbens, caudate, anterior insula, amygdala and orbitofrontal cortex.

The participants with the FTO gene variant had a larger BOLD signal response to high-energy foods in the orbitofrontal cortex and a larger high-energy food appeal rating. Participants who had the dopamine D2 receptor variant had more reward system responses to low-energy foods, which were seen primarily in the caudate and nucleus accumbens regions.

Participants who had the FTO variant gene but not the dopamine D2 receptor variant had increased reward responses to the high-calorie food images. Those who were positive for the dopamine D2 receptor variant had greater responses to high-energy foods, but only when they also were negative for the FTO variant.

In other research on FTO, Clare Llewellyn, PhD, a lecturer in behavioral obesity research at the Health Behavior Research Center at the University College London, and colleagues found that the FTO gene was linked to satiety sensitivity. At the time of this research, FTO was the only known genetic variant associated with obesity; currently, there are 97 known variants, according to Llewellyn.

Further study showed that, in more than 2,000 British children, satiety sensitivity also was associated with 28 other genetic variants of obesity, according to Llewellyn.

“We also took FTO out of the score to show that it wasn’t just FTO explaining all the relationship,” Llewellyn told Endocrine Today. “It was actually the other variants as well ... now it seems really demonstrative that many genetic risk variants for obesity appear to be associated with appetite.”

Knowing that satiety sensitivity has such a strong genetic basis — heritability is in the order of about 70% — it may be possible to change physiologic sensations, Llewellyn said. “Can you change your actual sensitivity to satiety so you can actually feel fuller than you did before, or is it possible to train people to pay more attention to it?” she asked.

Research by Karra and colleagues found a link between FTO, ghrelin and impaired brain food-cue responsivity. The researchers found that participants with the FTO “obesity risk” gene had higher postprandial concentrations of ghrelin — the only known hormone that stimulates food intake — which altered their appetite and food intake.

A study in older adults by Benedict and colleagues studied how FTO not only increases circulating ghrelin, but also reduced the level of satiety-enhancing leptin, creating a shift in that delicate endocrine balance, which led to weight gain.

Treating obesity, food addiction

As the subject of food addiction becomes more prominent, research has been turning to potential treatments.

Currently, strategies for treating someone who scores high for food addiction on the Yale Food Addiction Scale should be similar to options for treating other addictions, according to Klein.

“This means understanding that ‘food-addiction obesity’ is a chronic disease, as are substance use disorders, and requires a lifelong effort by the patient to control food intake. It is particularly important to identify the specific triggers that cause an individual to overeat so that treatment strategies can be developed that are designed to prevent the exposure to those triggers,” he said. “This requires developing a very structured lifestyle when it comes to food, just like an alcoholic who needs to be away from situations where they’re exposed to alcohol.”

Klein admits that the concept is easier to grasp than the practice. “The difference, of course, is that unlike with drinking or drugs, you can’t stop eating completely.”

Bariatric surgery, which has benefits for obesity-related comorbidities as well as obesity itself, may also prove helpful for treating food addiction. Using the Yale Food Addiction Scale and other measures, Klein and colleagues identified patients with high scores for food addiction who then underwent weight-loss procedures. After surgery, 93% of these patients experienced remission of food addiction and several eating behaviors associated with the condition.

Drug addiction literature may reveal other potential treatment options.

“Sharing of knowledge from one area could inform the other,” Frascella said. “We might develop some anti-obesity medications, and these might be effective in helping to treat and prevent substance abuse.”

Lawson and colleagues found that oxytocin may be helpful for appetite control. In a randomized, placebo-controlled crossover study, the researchers tested single-dose intranasal oxytocin in 25 fasting healthy men offered double portions of a meal. The results showed that oxytocin lowered caloric intake, with a particular effect on the consumption of fatty foods. There also was an increase in the level of anorexigenic hormone cholecystokinin with no effect on appetite or appetite-regulating hormones. Oxytocin use also improved insulin sensitivity.

In three animal models, Steensland and colleagues found that neurokinin 1 (NK1) receptor is involved in appetite and reward-related behaviors. Further, the researchers found that the NK1 antagonist ezlopitant decreased the consumption and seeking of sucrose and ethanol in animals.

Future research in this area must look at the hunger system as a whole, not just small sections, according to Kenny. “The greatest challenge for us in the field at the moment is understanding which systems may be involved in deciding which food you eat and how much you eat of it,” he said. “I really believe in the holistic approach.

“If you try to ... break everything down to one cluster of neurons, it’s great,” Kenny said. “You can understand them very well, but you have to be able to step back and look at everything and see how does all this fit together and have a novel therapeutic strategy that we can take that we wouldn’t think of otherwise when you consider the system as a whole.” – by Colleen Owens with additional reporting by Jill Rollet

• References:
• Benedict C, et al. Diabetes. 2014;doi:10.2337/db14-0470.
• Goldstone AP. OR #2065. Presented at: ObesityWeek; Nov. 2-6, 2015; Los Angeles.
• Hebebrand J, et al. Neurosci Biobehav Rev. 2014;doi:10.1016/j.neubiorev.2014.08.016.
• Karra E. J Clin Invest. 2013;doi:10.1172/JCI44403.
• Lawson EA, et al. Obesity. 2015;doi:10.1002/oby.21069.
• Llewellyn C. JAMA Pediatr. 2014;doi:10.1001/jamapediatrics.2013.4944.
• Steensland P, et al. PLoS One. 2010;doi:10.1371/journal.pone.0012527.

• For more information:
• Joseph Frascella, PhD, can be reached at National Institute on Drug Abuse, 6001 Executive Boulevard, Bethesda, MD 20892; email: jfrascel@nida.nih.gov.
• Ashley Gearhardt, PhD, can be reached at the University of Michigan, 2268 East Hall, 530 Church St., Ann Arbor, MI 48109-1043; phone: 415-265-9484; email: agearhar@umich.edu.
• Paul J. Kenny, PhD, can be reached at Mount Sinai Hospital, Hess CSM Building, Floor 9, Room 117, 1470 Madison Ave., New York, NY 10029; phone: 212-824-8970; email: paul.kenny@mssm.edu.
• Samuel Klein, MD, can be reached at Center for Human Nutrition, Washington University School of Medicine, 660 South Euclid Avenue Saint Louis, MO 63110; phone: 314-362-8708; email: sklein@dom.wustl.edu.
• Clare Llewellyn, PhD, can be reached at the University College London, 1-19 Torrington Place, Room 217, London WC1E6BT; phone: +44 20 7679 1263; email: c.llewellyn@ucl.ac.uk.
• Bradford B. Lowell, MD, PhD, can be reached at the Beth Israel Deaconess Medical Center, Center for Life Sciences, Rm-703 3 Blackfan Circle, Boston, MA 02115; phone: 617-735-3366; email: blowell@bidmc.harvard.edu.
• Jennifer Miller, MD, can be reached at the University of Florida, 1600 SW Archer Road, Gainesville, FL 32610; phone: 352-294-8226; email: millejl@peds.ufl.edu.
• Gene-Jack Wang, MD, can be reached at the National Institute on Alcohol Abuse and Alcoholism, 10 Center Drive, Rm B2L304, Bethesda, MD 20892-1013; phone: 301-496-5012; email: gene-jack.wang@nih.gov.

Disclosures: Frascella, Miller and Wang report no relevant financial disclosures. Kenny reports being co-founder and shareholder in Eolas Therapeutics. Klein reports being a shareholder of AspireBariatrics and having served on the scientific advisory boards for Takeda and NovoNordisk. Lowell reports a basic science sponsored research agreement with Pfizer.

Should there be a clinical diagnosis of food addiction?

There may well be need for a clinical diagnosis of food addiction in the future, but not yet.

Despite its popularity and appeal, the idea of food addiction is very poorly defined. That may not be the impression that one gets from both the scientific literature and the lay media, which can convey the sense that this is a condition that we know and understand and is just awaiting recognition. There are two very closely related ideas that are often conflated: One is that certain foods (the candidates are highly palatable foods, rich in fat and sugar) are addictive; the other is that certain people have problems with eating that strongly resemble addiction to drugs like cocaine and alcohol. However, there is really no evidence to say that any foods are addictive in humans, and the syndrome of food addiction has not been rigorously defined and validated. By the latter I mean that we do not have a good definition of food addiction, one that distinguishes it from other eating disorders in terms of its features and its course (and treatment, when we get to it). This is something we would require for a clinical diagnosis.

So if this is the case, why do I say “not yet” instead of “no”? This is because it is possible that there may be a small group of people who do have problems with eating that are very much like a substance addiction. Indeed, given what we know about the breadth of symptoms experienced by people with eating disorders, the idea of food addiction is not inconceivable. As with drug addiction, it may be the combination of certain foods and certain individuals with particular vulnerabilities that results in a food addiction. If we could come to an idea of food addiction in this specific sense, then we should seriously think about recognizing it as a clinical diagnosis. Unfortunately, so far no one is really looking into this, and recognizing what we have at the moment as a clinical diagnosis is not going to help us identify or help these individuals.

Thus, no clinical diagnosis is necessary yet.

Hisham Ziauddeen, MRCPsych, PhD, is a clinical senior research associate at the University of Cambridge in the United Kingdom and an honorary consultant psychiatrist at the Wellcome Trust-MRC Institute of Metabolic Science, Cambridge Biomedical Campus. Disclosure: Ziauddeen reports no relevant financial disclosures.

Focus should be on addictive behavior, not food.

The term “food addiction” is commonly used to describe an addictive-like eating behavior. We can almost immediately think of a particular person as being “food addicted,” most likely an individual with obesity. Nevertheless, as with any other diagnosis of a mental disorder — including any defined behavioral — it is important to delineate the respective diagnostic criteria.

The criteria for food addiction were first proposed by Gearhardt and colleagues in 2009. This led to the development of the original Yale Food Addiction Scale, a questionnaire based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) criteria for substance dependence; a revised version was introduced in 2014 to accommodate the changes made in DSM-5 for substance use disorders. However, the mere delineation of criteria for a novel diagnosis is not sufficient for reaching a formally accepted diagnosis. Instead, both empirical evidence and a consensus within the psychiatric community and, in particular, within working groups responsible for defining specific criteria for the disorders within a specific diagnostic category — such as within the context of the newly defined DSM-5 category “Substance-Related and Addictive Disorders” — are required prior to the creation of a novel diagnosis.

The major criticism about the creation of a food addiction diagnosis that is based on criteria used for diagnosing substance use disorders is related to the fact there is no clear-cut evidence in humans to indicate that particular foods (eg, specific micronutrients) cause a chemical-like addiction similar to what is seen with drugs, such as nicotine, alcohol and cannabis. Quite to the contrary, most subjects with extreme obesity eat a host of different foods; it could even be argued that obesity would most likely not be common in our societies if only a very limited amount of food products were available. The novel inclusion of addictive disorders on top of substance use disorders in DSM-5 (the term “addiction” had been avoided in previous versions because of problems related to a precise operationalization of the term “addiction”) theoretically now entails the possibility to create a more broadly defined diagnosis of “eating addiction.” This would entail a shift away from substance use disorders to an addictive behavior.

Johannes Hebebrand, MD, is a psychiatrist specializing in eating behaviors and eating disorders in the Department of Child and Adolescent Psychiatry, LVR-Klinikum Essen, University of Duisburg-Essen in Essen, Germany. Disclosure: Hebebrand reports no relevant financial disclosures.