Physiology & Behavior, Vol. 50, pp. 101-108. © Pergamon Press plc, 1991. Printed in the U.S.A.
0031-9384/91 $3.00 + .00
Eating Behavior in Eating Disorders: Response to Preloads M A R I O N M. H E T H E R I N G T O N *l A N D B A R B A R A J. R O L L S t
*Department of Psychology, University of Dundee, Dundee, Scotland "pDepartment of Psychiatry, The Johns Hopkins University School of Medicine Meyer 207, 600 North Wolfe Street, Baltimore, MD 21205 Received 31 January 1990 HETHERINGTON, M. M. AND B. J. ROLLS. Eating behavior in eating disorders:Response to preloads. PHYSIOL BEHAV 50(1) 101-108, 1991.--Eating behavior in eating-disordered subjects was investigated by recording food intake and subjective ratings following three preloads differing in calories, weight and connotation. Subjects were patients with a DSM-IIIR diagnosis of anorexia nervosa or bulimia nervosa and nonpatient volunteers (normal-weight or overweight dieters, and normal-weight nondieters). After all preloads, anorectics ate significantly less than all other subjects except normal-weight dieters, and anorectics rated hunger and desire to eat consistently lower and fullness greater than all other subjects. When analysis of intake was adjusted for body weight, anorectics and normal-weight dieters still consumed significantly less than controls. Anorectics selected foods that were lower in fat and carbohydrate and ate a larger proportion of calories as protein than the other subjects. All groups decreased subsequent intake after the high-calorie preload except bulimics. This study demonstrates that the regulatory capacity of eatingdisordered individuals can be investigated and that aberrant eating behavior was observed. Anorexia nervosa Bulimia nervosa Dieting Restraint Eating disorders Food intake Satiety Human feeding Obesity Eating behavior
ANOREXIA and bulimia nervosa are both characterized by disturbed eating behaviors which may range from self-starvation to chaotic eating involving chronic cycles of bingeing and purging. Clinical reports have provided the first descriptions of abnormal appetite regulation and disordered eating (1, 20, 24, 31, 33). Despite the obvious importance of understanding the aberrant features of disturbed eating for the treatment of disordered eating, surprisingly few experimental studies have been done. Most studies that have investigated the appetite regulation of eating-disordered subjects suggest that these individuals are deficient in hunger and satiety detection (5, 6, 9, 16, 25, 28, 35). Research into dietary choices, energy intake, macronutrient selection and food preferences in these subjects has revealed no consistent patterns (18). In one of the earliest of these studies (32), anorectic patients ate significantly fewer calories and obtained less of their calories from carbohydrate than normal controis, whereas a later study (3) indicated that the source of calories from carbohydrate was normal in anorectics, but that energy derived from protein was higher and from fats was significantly lower than in controls. Both studies used diet records as the method of assessing food choices, whereas assessment of macronulrient selection and energy intake in a hospital setting revealed that anorectic patients selected a diet similar in composition to that of controls. In addition, it was demonstrated that energy intake per kilogram of body weight was not different between anorectics and controls (12). Using laboratory tests of food intake, Kissileff and his col-
Appetite
Hunger
leagues (21,39) assessed the caloric and macronutrient intake of test meals by bulimic subjects. They found that the macronutrient composition of intake for both "binge" meals and "nonbinge" meals was similar to that of the typical American diet. However, bulimic women ate significantly more than controls when instructed to binge. Rosen et al. (30) found that bulimics ate less than controls when the opportunity to purge was prevented. The present study was conducted to assess the ability of eating-disordered individuals to respond appropriately to a fixed quantity of food. The main aim of this study was to combine an investigation of the subjective experience of hunger, appetite and satiety with an assessment of some objective parameters of energy intake, macronutrient selection and food choice in eatingdisordered subjects. Dieting subjects were also tested. Dieters, though not meeting criteria for an eating disorder, chronically restrict food intake, may binge when dietary limits are crossed, and show counterregulation (13,14) in the laboratory. This population provides an intermediate group between nondieting, normal-weight controls and individuals with a clinically diagnosed eating disorder (15). Both normal-weight and overweight dieters were selected for this study to assess the relative contribution of dietary restriction and body weight to eating behavior. Three different preloads were selected for this study, varying in calories, amount of food and connotation (i.e., "high calorie" or "low calorie"). Connotation was determined in a pilot
~Requests for reprints should be addressed to Dr. Marion M. Hetherington, Department of Psychology, Park Place, The University, Dundee, DD1 4HN, Scotland,UK.
lO1
102
HETHERINGTON AND ROLLS
TABLE 1 SUBJECT CHARACTERISTICS
Group
Age
% IBW
EAT
EI:R
EI:DIS
EI:PH
GHQ
Anorexics (n = 10) Bulimics (n = 10) Normals (n = 10) Nwt. dieters (n= 11) Owt. dieters (n = 10) ANOVA F(4,46)
22.7 z 1.4
76.1 - 2.6
42.6 __. 5.5
15.2 ± 1.2
6.7 --_ 1.3
5.5 ± 1.2
9.5 ± 2.5
21.7 --- 1.8
105.6 - 3.4
50.9 -- 5.8
16.2 + 0.7
13.2 --- 0.8
6.5 + 1.0
7.4 --- 2.2
26.3 --+ 1.3
91.7 ± 2.5
8.0 • 1.2
6.1 --- 0.8
4.8 --- 1.1
4.1 _ 0.6
1.3 --- 0.7
23.9 ± 1.2
93.2 --- 3.2
18.8 --- 5.1
13.8 --+ 1.2
7.2 ± 0.9
5.4 __+ 1.0
1.6 ± 1.2
26.2 ± 1.8
139.0 --- 8.0
22.5 --- 4.5
14.0 - 1.2
11.0 --- 0.9
5.3 ± 0.8
3.6 --- 1.7
NS
F = 28.5 p
0031-9384/91 $3.00 + .00
Eating Behavior in Eating Disorders: Response to Preloads M A R I O N M. H E T H E R I N G T O N *l A N D B A R B A R A J. R O L L S t
*Department of Psychology, University of Dundee, Dundee, Scotland "pDepartment of Psychiatry, The Johns Hopkins University School of Medicine Meyer 207, 600 North Wolfe Street, Baltimore, MD 21205 Received 31 January 1990 HETHERINGTON, M. M. AND B. J. ROLLS. Eating behavior in eating disorders:Response to preloads. PHYSIOL BEHAV 50(1) 101-108, 1991.--Eating behavior in eating-disordered subjects was investigated by recording food intake and subjective ratings following three preloads differing in calories, weight and connotation. Subjects were patients with a DSM-IIIR diagnosis of anorexia nervosa or bulimia nervosa and nonpatient volunteers (normal-weight or overweight dieters, and normal-weight nondieters). After all preloads, anorectics ate significantly less than all other subjects except normal-weight dieters, and anorectics rated hunger and desire to eat consistently lower and fullness greater than all other subjects. When analysis of intake was adjusted for body weight, anorectics and normal-weight dieters still consumed significantly less than controls. Anorectics selected foods that were lower in fat and carbohydrate and ate a larger proportion of calories as protein than the other subjects. All groups decreased subsequent intake after the high-calorie preload except bulimics. This study demonstrates that the regulatory capacity of eatingdisordered individuals can be investigated and that aberrant eating behavior was observed. Anorexia nervosa Bulimia nervosa Dieting Restraint Eating disorders Food intake Satiety Human feeding Obesity Eating behavior
ANOREXIA and bulimia nervosa are both characterized by disturbed eating behaviors which may range from self-starvation to chaotic eating involving chronic cycles of bingeing and purging. Clinical reports have provided the first descriptions of abnormal appetite regulation and disordered eating (1, 20, 24, 31, 33). Despite the obvious importance of understanding the aberrant features of disturbed eating for the treatment of disordered eating, surprisingly few experimental studies have been done. Most studies that have investigated the appetite regulation of eating-disordered subjects suggest that these individuals are deficient in hunger and satiety detection (5, 6, 9, 16, 25, 28, 35). Research into dietary choices, energy intake, macronutrient selection and food preferences in these subjects has revealed no consistent patterns (18). In one of the earliest of these studies (32), anorectic patients ate significantly fewer calories and obtained less of their calories from carbohydrate than normal controis, whereas a later study (3) indicated that the source of calories from carbohydrate was normal in anorectics, but that energy derived from protein was higher and from fats was significantly lower than in controls. Both studies used diet records as the method of assessing food choices, whereas assessment of macronulrient selection and energy intake in a hospital setting revealed that anorectic patients selected a diet similar in composition to that of controls. In addition, it was demonstrated that energy intake per kilogram of body weight was not different between anorectics and controls (12). Using laboratory tests of food intake, Kissileff and his col-
Appetite
Hunger
leagues (21,39) assessed the caloric and macronutrient intake of test meals by bulimic subjects. They found that the macronutrient composition of intake for both "binge" meals and "nonbinge" meals was similar to that of the typical American diet. However, bulimic women ate significantly more than controls when instructed to binge. Rosen et al. (30) found that bulimics ate less than controls when the opportunity to purge was prevented. The present study was conducted to assess the ability of eating-disordered individuals to respond appropriately to a fixed quantity of food. The main aim of this study was to combine an investigation of the subjective experience of hunger, appetite and satiety with an assessment of some objective parameters of energy intake, macronutrient selection and food choice in eatingdisordered subjects. Dieting subjects were also tested. Dieters, though not meeting criteria for an eating disorder, chronically restrict food intake, may binge when dietary limits are crossed, and show counterregulation (13,14) in the laboratory. This population provides an intermediate group between nondieting, normal-weight controls and individuals with a clinically diagnosed eating disorder (15). Both normal-weight and overweight dieters were selected for this study to assess the relative contribution of dietary restriction and body weight to eating behavior. Three different preloads were selected for this study, varying in calories, amount of food and connotation (i.e., "high calorie" or "low calorie"). Connotation was determined in a pilot
~Requests for reprints should be addressed to Dr. Marion M. Hetherington, Department of Psychology, Park Place, The University, Dundee, DD1 4HN, Scotland,UK.
lO1
102
HETHERINGTON AND ROLLS
TABLE 1 SUBJECT CHARACTERISTICS
Group
Age
% IBW
EAT
EI:R
EI:DIS
EI:PH
GHQ
Anorexics (n = 10) Bulimics (n = 10) Normals (n = 10) Nwt. dieters (n= 11) Owt. dieters (n = 10) ANOVA F(4,46)
22.7 z 1.4
76.1 - 2.6
42.6 __. 5.5
15.2 ± 1.2
6.7 --_ 1.3
5.5 ± 1.2
9.5 ± 2.5
21.7 --- 1.8
105.6 - 3.4
50.9 -- 5.8
16.2 + 0.7
13.2 --- 0.8
6.5 + 1.0
7.4 --- 2.2
26.3 --+ 1.3
91.7 ± 2.5
8.0 • 1.2
6.1 --- 0.8
4.8 --- 1.1
4.1 _ 0.6
1.3 --- 0.7
23.9 ± 1.2
93.2 --- 3.2
18.8 --- 5.1
13.8 --+ 1.2
7.2 ± 0.9
5.4 __+ 1.0
1.6 ± 1.2
26.2 ± 1.8
139.0 --- 8.0
22.5 --- 4.5
14.0 - 1.2
11.0 --- 0.9
5.3 ± 0.8
3.6 --- 1.7
NS
F = 28.5 p
