2.6: Theories of the Causes of Obesity

Guest Lecturer: Albert Stunkard, MD

[In a 1986 lecture⁷, Dr. Stunkard, a pioneer in obesity research, reflected on his research of the preceding 30+ years—before today’s era of identifying specific genes. This is an excerpt from that lecture.]

Our understanding of the causes of obesity has progressed since the days when obesity was considered “the presenting symptom of a basic personality disorder” and “a particular way of handling one’s poor relationship with oneself.” Current approaches are far more modest. They start from a finding so simple that it was largely overlooked in the days of overarching theory. Despite its simplicity and uncontroversial nature, this finding has been intensively scrutinized until it was firmly established: obesity runs in families. Thirty years ago, this finding would have hardly seemed worth mentioning. Today it is a starting point for research.

Why does obesity run in families? Is it due to genetic influences or to environmental influences?

Environmental Influences

Until recently, most of the evidence favored environmental influences. This early evidence was uncovered in an unexpected way, in one of the surprises that makes research such a delight. We had started out to assess the extent of mental illness of obese people in the Midtown Manhattan Study, “Mental Health in the Metropolis.” Then we got sidetracked by a factor we had not considered—socioeconomic status.

Social class showed a striking relationship to obesity. Obesity was six times more common among lower class women than among upper class women. Furthermore, the Midtown data enabled us to go beyond simple correlations and infer causation.

The Midtown Study had assessed not only the social class of the respondents themselves, but also their social class of origin—the social class of their fathers when they were eight years old. Social class of origin strongly predicted obesity. A person’s social class of origin was almost as strongly related to obesity as was his or her own social class. The social class into which you are born, particularly if you are a woman, strongly determines whether or not you will be obese.

Among men, the relationship is less clear: sometimes, as among women, the lower the socioeconomic status, the greater the prevalence of obesity; sometimes it is the reverse. Among children (of both sexes) usually no relationship is found. Only as they grow into adulthood do the distinctive relationships develop.

Whatever the nature of the social forces that determine obesity or thinness, they have a pervasive influence. Even pet animals seem susceptible. A report in the English veterinary literature revealed that 44% of the dogs of people “of obese physique” were obese as compared with 25% of the dogs of those of normal weight!

Genetic Influence

Evidence of a genetic influence on human obesity is of very recent origin. It was obtained by the use of the classic method of assessing genetic influences on humans, twin studies. The rationale of twin studies is simple: identical twins share both a common environment and common genes. Fraternal (nonidentical) twins share a common environment but only half their genes.

The differences in correlations in weight between pairs of identical twins and between pairs of fraternal twins were calculated. Since the environments of twins of either type are presumably similar, any difference between the correlations of the different kinds of twin should be due to genetic influences. The data showed a strong genetic influence extending across the entire spectrum from thin to fat.

Conclusions from the Twin Studies

The research thus shows that both genes and environment play roles in determining human obesity. Genetic influences may largely determine whether or not an individual becomes obese. Given this genetic vulnerability, it is environmental influences that determine how obese the person becomes.

These twin studies have three limitations, however. First, they may overestimate the influence of heredity. Second, they are confined to men. Third, because they are confined to men who could pass a physical examination for entry into the armed forces, they effectively excluded those with childhood-onset obesity.

Adoption Studies

These limitations made it desirable to supplement the twin studies with other methods, and we have done that with two adoption studies. The rationale of adoption studies is straightforward. Adoptees are compared with both their biologic and their adoptive parents for obesity. A relationship between biologic parents and adoptees bespeaks a genetic influence; a relationship between adoptive parents and adoptees indicates an environmental influence.

The first of our adoption studies used the Danish Adoption Register, which had been employed successfully in previous studies on schizophrenia and on alcoholism. We obtained height and weight data for 3,590 adoptees who were living in Copenhagen at the time of the study, and whose average age was 42 years.

We next ascertained the heights and weights of both the biologic and adoptive parents, and compared them with those of the adoptees. This comparison of the relationship between biologic parents and adoptees indicated a genetic influence.

Not surprisingly, this finding confirmed the results of the twin study. What was a surprise was the absence of any apparent effect of the early family environment. The weight class of the adoptees was not related to the weight classifications of their adoptive parents. Whatever the influence their adoptive parents may have had when the adoptees were children, there were no traces of it when they reached middle age. The dire consequences of faulty childhood eating habits, so often blamed for the developments of obesity, have been, to say the least, exaggerated.

Both twin and adoption studies thus show that heredity influences human obesity.

The Role of Genetics

What role then does genetics play in human obesity? Is human obesity determined at conception, like hair color and eye color, and can people do nothing to alter their fate? As far as obesity is concerned, are genes destiny?

The answer is “No!” Obesity is not determined at conception. What is inherited is a vulnerability, which requires a suitable environment in order to be manifested.

Our current understanding of the genetic vulnerability to obesity provides no basis for personal despair or therapeutic surrender. Countless formerly obese people have shown how courage and effort can overcome even the most unpromising heritage. With increased understanding of genetic influences on human obesity, we shall be able to identify people at increased risk and help them to cope in a more informed manner. Understanding their greater vulnerability to obesity can be a prelude to better control.

The Environmental Challenge

The environmental circumstances that contribute to obesity in the United States today are probably extensive. Americans consume a diet high in calories and expend few calories in physical activity. Our diet is high enough in fat to produce obesity in thin rats, and our labor-saving devices are so efficient that we become fat on a diet that now has 1,000 fewer calories than it did at the turn of the century.

The issue in human obesity is not one of heredity or environment, or heredity versus environment, but of heredity and environment, or genetic vulnerability and environmental challenge. This issue affects all of psychiatry, indeed all of medicine. It may well set the agenda for medical research for the foreseeable future.

The Set Point Theory

Another theory whereby genetic and environmental influences both may exert their effect is the set point theory of obesity. Arguments for this theory begin with the observation that the body weight of animals is regulated. Not only does their body weight remain at a relatively constant level when circumstances remain constant, but it is also defended at that level when circumstances change; it responds to environmental challenges.

The remarkable ability of animals to regulate body weight is exemplified by animals that migrate and hibernate. They not only regulate body weight at a constant level under ordinary circumstances, but also perform extraordinary feats of anticipatory regulation. Before its 600 mile flight across the Caribbean Sea, the ruby-throated hummingbird overeats to produce fat stores of precisely the amount necessary for this long flight. Too much fat and it could not get off the ground; too little and it would fall into the sea short of its goal. Similarly, the golden-mantled ground squirrel overeats to produce the exact amount of energy stores it needs for its months of hibernation.

For many years, evidence of the regulation of body weight was confined to animals of normal weight. Then a number of investigations revealed that the body weight of obese animals. is also regulated. One way of explaining their obesity is that their body weight is being regulated at an elevated level by an elevation of their set point.

Studies of hoarding behavior support the existence of regulation. When food intake is restricted and rats lose a certain percentage of their body weight, they begin to hoard food. The weight at which hoarding begins can be interpreted as a body-weight set point, and rats will hoard enormous amounts of food as their weight falls below this level. Hoarding behavior occurs in obese rats as well as in normal-weight rats when their body weight falls below a certain critical level.

Evidence for the existence of a body weight set point in humans is limited to two studies of men of normal weight. In the landmark Minnesota starvation study, volunteers lost 25% of their body weight, whereas in the Vermont prison study, men gained 50 or more pounds. In each study, when subjects were allowed to control their own food intake, their body weight rapidly returned to normal.

Normal weight people can apparently regulate their body weight. We have no information about the regulation of body weight of obese humans, and the constraints upon experimentation with human subjects make it unlikely that further experiments of this type will be carried out. As a result, much of what we learn about the regulation of body weight in humans must be extrapolated from research with animals.

Albert Stunkard, MD was a Professor of Psychiatry at the University of Pennsylvania School of Medicine. He was the first to describe binge eating, and published more than 400 articles, including his studies of the night-eating syndrome.⁸ He died at age 92 in 2014.