Eating behavior in obese patients with and without type 2 diabetes mellitus
International Journal of Obesity (2002) 26, 848–853
ß 2002 Nature Publishing Group All rights reserved 0307–0565/02 $25.00
www.nature.com/ijo
PAPER
Eating behavior in obese patients with and without
type 2 diabetes mellitus
E Mannucci1, F Tesi1, V Ricca2, E Pierazzuoli1, E Barciulli2, S Moretti2, M Di Bernardo2, R Travaglini2,
S Carrara1, T Zucchi2, GF Placidi2 and CM Rotella1*
1
Chair of Metabolic Diseases, Section of Endocrinology, Department of Clinical Pathophysiology, University of Florence Medical
School, Florence, Italy; and 2Psychiatric Unit, Department of Neurologic and Psychiatric Sciences, University of Florence Medical
School, Florence, Italy
OBJECTIVE: Aim of this study was the assessment of the prevalence of eating disorders, and of eating disorder symptoms, in
obese patients with type 2 diabetes, compared to non-diabetic subjects.
DESIGN: Three samples of individuals were studied: a series of 156 (76 male, 80 female) overweight and obese type 2 diabetic
patients, aged 30 – 65 y, with a body mass index (BMI) > 28 kg=m2 (DM); a series of 192 (20 male, 172 female) obese
(BMI > 30 kg=m2) non-diabetic patients aged 30 – 65 y seeking treatment for weight loss (OC); and a non-clinical sample of 48
(22 male, 26 female) obese (BMI > 30 kg=m2) subjects aged 30 – 65 y selected from the lists of two general practices (OP). Eating
behavior was assessed using the Eating Disorder Examination (EDE 12.0D).
RESULTS: The prevalence of Binge Eating Disorder was lower than 5% in all the three samples. Median EDE scores in females
were significantly higher in OC (3.0) and OP (3.4) than in DM (1.7), while diabetic patients showed higher scores on Restraint
than both non-diabetic samples. Among diabetic patients, a significant correlation of EDE scores with HbA1c was observed.
CONCLUSIONS: Type 2 diabetes is unlikely to induce relevant eating disturbances in obese patients, apart from an increase in
restraint. Abnormalities of eating attitudes and behavior are associated with an impairment of metabolic control.
International Journal of Obesity (2002) 26, 848 – 853. doi:10.1038=sj.ijo.0801976
Keywords: eating disorders; behavior; diabetes; obesity; EDE 12.0D
Introduction
Several studies have been performed in order to evaluate the
possible role of diabetes mellitus as a risk factor for eating
disorders. Although some earlier studies performed with selfreported questionnaires reported a high prevalence of anorexia nervosa and bulimia nervosa in young diabetic females,1
controlled surveys with interview-based diagnoses always
failed to detect any significant difference in the prevalence
of eating disorders between type 1 diabetic individuals and
control subjects.2 – 7 However, different studies have shown
that minor eating disturbances in type 1 diabetes are frequent, and they are associated to remarkable impairment of
metabolic control.2,7 – 11
*Correspondence: CM Rotella, Sezione Malattie del Metabolismo e
Diabetologia, Dipartimento di Fisiopatologia Clinica, Viale Pieraccini 6,
50134 Firenze, Italy.
E-mail:
Received 31 July 2001; revised 5 December 2001;
accepted 13 December 2001
Few studies have been addressed to the assessment of
eating attitudes and behaviors in patients with type 2 diabetes.12 – 14 The typical age of onset of type 2 diabetes is much
higher than that of eating disorders; for this reason, this
disease is unlikely to represent a risk factor for the development of eating disorders. It should also be considered, however, that most type 2 diabetic patients are obese, and that
obesity is often associated with minor eating disturbances; in
fact, binge eating has been reported to be frequent among
type 2 diabetic patients.12 – 14 Dietary restraint could facilitate
the onset of minor eating disturbances, as proposed for type
1 diabetic patients.15 Furthermore, eating disorders could
facilitate, at least theoretically, the onset of type 2 diabetes;
in fact, disturbances of eating behavior could promote
weight gain, and excess weight is a well-known risk factor
for the development of type 2 diabetes. In the only controlled study published to date, newly diagnosed type 2
diabetic patients did not show any significant difference in
eating behavior when compared to matched controls,13 but
the sample could have been too small to detect minor
Eating behavior in obese type 2 diabetes
E Mannucci et al
849
differences. It should also be considered that abnormalities
of eating behavior can affect considerably body weight and
metabolic control in type 2 diabetes.14
The aim of this study was the assessment of the prevalence of eating disorders, and of eating disorder symptoms,
in overweight and obese patients with type 2 diabetes,
compared to non-diabetic obese subjects.
Methods
The study was designed as a cross-sectional, controlled
survey. Three samples of subjects were studied. The first
sample was overweight and obese patients with type 2
diabetes (DM): all patients with known type 2 diabetes,
aged 30 – 65 y, with a duration of diabetes of at least 1 y
and with body mass index (BMI) > 28 kg=m2 attending the
Outpatient Clinic of Metabolic Diseases and Diabetology of
the University of Florence for the treatment of diabetes
between 1 January 1998 and 31 August 1999, were asked to
participate to the study, unless they met one of the following
exclusion criteria:
physical illnesses known to interfere considerably with
eating behavior, such as untreated thyroid disorders and
gastrointestinal diseases;
known mental disorders, such as schizophrenia, substance or alcohol dependence, or mental retardation,
which could affect the ability to answer the interviewer’s
questions;
current medication with psychotropic drugs known to
interfere with eating behavior, such as anti-epileptics,
anti-depressants and anti-psychotics;
pregnancy or lactation.
Of the 161 patients invited, five refused to participate,
with a response rate of 96.9%. Non-participants did not
differ significantly from participants for any of the characteristics studied (data available upon request). The final
sample was therefore composed of 156 patients who provided their written informed consent, and whose characteristics are summarized in Table 1. Among the patients studied,
26 were treated with diet only, 83 with oral hypoglycemic
agents (21 with metformin, 15 with sulphonylureas, 11 with
acarbose and 36 with combinations of different drugs), 21
with combinations of insulin and hypoglycemic agents, and
26 with insulin only. Of the patients studied, 32 were
affected by diabetic retinopathy, 31 by diabetic neuropathy,
18 by coronary artery disease, 15 by nephropathy and 11 by
obstructive arteriopathy of the lower limbs; 80 patients were
not affected by any complication, 51 showed one complication, 20 were affected by two complications, and five by
three or more complications.
The second sample was obese non-diabetic subjects, clinical sample (OC): all non-diabetic patients aged 30 – 65 y,
with a BMI > (...truncated)
