Dietary Restriction: A Comparison of Vegetarians and Non-vegetarians

One type of vegetarians was at higher risk for disordered eating.

Vegetarianism is an increasingly popular lifestyle, and an estimated 1% of the US population now identify themselves as vegetarians. This seemed an ideal opportunity for psychologist Dr. Alix Timko and colleagues at the University of Pennsylvania and Towson University to investigate claims of a link between adherence to a vegetarian diet and the onset and maintenance of disordered eating (Appetite 2012; doi:10.1016/j.appet.2012.02.005). According to the authors, results of previous studies have been hampered by small samples of true vegetarians, lack of appropriate operational definitions of “vegetarianism,” and uncertainly about the appropriate assessments of eating behaviors for semi-vegetarians.

Three categories of ‘vegetarians’

Vegetarians are usually defined in three categories: ovo-vegetarians include eggs but no dairy products in their diets; lacto-vegetarians include dairy products but avoid eggs; and lacto-ovo vegetarians include both eggs and dairy products in their diets. Semi-vegetarians restrict the type of meat they consume, and vegans exclude all red meat, fish, poultry, dairy and other animal-origin foods, including eggs; they also generally avoid buying or using non-edible animal products such as leather.

To date, two studies have attempted to link vegetarianism and eating disorders. In the first, of 200 patients receiving treatment for anorexia nervosa (AN), just under half were considered to be vegetarians (Int J Eat Disord 1986; 5:539). In the second study, of 116 patients with AN, slightly more than half claimed they ate a vegetarian diet (defined as not eating red meat) (Med J Australia 1987; 147:540) but only 4 of the patients reported adhering to a vegetarian diet before the onset of their eating disorder.

Two studies were designed by the Pennsylvania team to investigate the link between vegetarianism and disordered eating behaviors. In the first, eating behaviors were assessed in the largest samples of confirmed true vegetarians and vegans surveyed to date, and were then compared with behaviors among semi-vegetarians and omnivores. The second study examined differences in restraint and disordered eating between semi-vegetarians and omnivores.

Study 1: True vegetarians and vegans

Of the 714 people recruited, a group of 486 (374 females and 111 males) made up the final study group. In addition to providing demographic data, participants were asked to complete a detailed questionnaire including whether or not they adhered to a vegetarian diet. They also completed several questionnaires, including the Self-Esteem Scale, the Eating Attitudes Test-26 (EAT-26), and the Food Frequency Questionnaire.

The study included 35 vegans, 111 true vegetarians, 75 semi-vegetarians, and 265 non-vegetarians. Unlike previous studies that reported concern about health as the number-one reason given for choosing a vegetarian diet, participants in this study indicated ethics as the primary reason for choosing a vegan, vegetarian, or semi-vegetarian diet. Only 8 participants indicated that concern about weight was the primary reason they began a meat-restricted diet.

Participants who were semi-vegetarians had the most disordered food-related behaviors and attitudes, as evidenced by higher levels of restraint, external eating, hedonic hunger, and avoidance of food cutes. Although not significantly different, semi-vegetarians had higher EAT-26 scores and diet scores than did the other groups.

Study 2: Semi-vegetarians and omnivores

The Pennsylvanian researchers then designed a follow-up study to assess weight- and eating-related behaviors of semi-vegetarians compared to omnivores in more detail. Since semi-vegetarians adhere to strict rules about eating certain types of food and even eliminate entire food groups from their diets, the authors hypothesized that this group would have higher-than-normal scores on traditional assessments of abnormal eating behaviors.

A group of 136 undergraduate females at the University of Pennsylvania completed an anonymous Web-site-based questionnaire in exchange for research participation credit. The final group included 117 respondents who either avoided only some meat (“red meat” or poultry only) and 74 who indicated they ate all types of meat. All participants provided current height and weight information as well as their ideal weights. They were also asked about perceived overweight in relation to their peers. All also completed the Restraint Scale and Eating Disorders Examination-Questionnaire.

Most respondents were in the low-to-average weight range, with a mean body mass index, or BMI, of 21.40 kg/m². There were no significant differences in self-reported BMIs or prevalence of dieting between semi-vegetarians and omnivores. A t-test indicated that semi-vegetarians scored significantly higher on the Restraint Scale than did omnivores.

Semi-vegetarians were at greatest risk

When the results of both studies are taken together, they indicate that semi-vegetarians are at the greatest risk of disordered eating behaviors. According to the authors, because of the current food environment in the US, where high-fat foods are less expensive than many more nutritious foods, semi-vegetarians may eat a diet low in meat products in an attempt to control their weight and to spend less on food. The authors also stress the importance of knowing the different types and degrees of vegetarianism to better identify individuals who may be at risk for disordered eating.