ArticlePDF Available

A 57-YEAR FOLLOW-UP INVESTIGATION AND REVIEW OF THE MINNESOTA STUDY ON HUMAN STARVATION AND ITS RELEVANCE TO EATING DISORDERS

Authors:
  • School of Medicine Stanford University

Abstract and Figures

Objective: This follow-up study reports on 19 of 36 male participants in the Minnesota Semi-Starvation Experiment. As systematic data were obtained for only 3 months of controlled nutritional rehabilitation following 6 months of semi-starvation, the follow-up aim was to reexamine the acute effects and inquire into possible long term physical and psychosocial effects from undergoing semi-starvation. The experiment has been a source of information for understanding eating disorders, particularly anorexia nervosa. Therefore, another aim was to examine the relevance of any starvation-induced symptomatic changes to eating disorders. Method: Semi-structured phone interviews were employed to explore 1) physical and psychological consequences of semi-starvation and nutritional rehabilitation 2) eating and weight changes during and since completion of the experiment, and 3) quality of the participants' lives following completion of the experiment. Results: Participants proceeded to lead interesting and productive lives, free of lifelong adverse effects. Personality differences, inferred from the Minnesota Multiphasic Personality Inventory, likely influenced the severity of the psychopathological reactions to starvation. Many participants reported maintaining a higher than normal weight and had abnormal eating habits for many months and even years before returning to "normal" state. Discussion: Reestablishment of normal body weight took significantly longer than suggested in the original experiment, and might therefore constitute a factor contributing to the extended course of illness and tendency to relapse in eating disorders. The preservation of energy and normal to high activity levels in the presence of signs of severe weight loss and starvation and body image disturbances seen in anorexia nervosa were not observed nor reported in the Minnesota Semi-starvation Experiment.
Content may be subject to copyright.
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 1 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
A 57-YEAR FOLLOW-UP INVESTIGATION AND
REVIEW OF THE MINNESOTA STUDY ON HUMAN
STARVATION AND ITS RELEVANCE TO EATING
DISORDERS
Elke D Eckert1*, Irving I Gottesman2, Susan E Swigart3, Regina C Casper4
Author Affiliations:
1* Elke D Eckert MD - Professor of Psychiatry, emeriti, University of Minnesota, Minneapolis, MN
55454, ecker001@umn.edu Corresponding author
2 Irving I. Gottesman PhD, FRC Psych Hon, FKC Hon, Senior Fellow and Professor, University of
Minnesota, Department of Psychology, Minneapolis, MN 55455, deceased
3 Susan Elizabeth Swigart MD, Child, Adolescent and Adult Psychiatrist, The Emily Program, St.
Louis Park, MN, Susan.Swigart@emilyprogram.com
4 Regina C Casper MD, Professor of Psychiatry, emerita, active, Stanford University School of
Medicine, Stanford, CA 94305, rcasper@stanford.edu
Abstract
Objective: This follow-up study reports on 19 of 36 male participants in the Minnesota Semi-
Starvation Experiment. As systematic data were obtained for only 3 months of controlled
nutritional rehabilitation following 6 months of semi-starvation, the follow-up aim was to re-
examine the acute effects and inquire into possible long term physical and psychosocial effects from
undergoing semi-starvation. The experiment has been a source of information for understanding
eating disorders, particularly anorexia nervosa. Therefore, another aim was to examine the
relevance of any starvation-induced symptomatic changes to eating disorders.
Method: Semi-structured phone interviews were employed to explore 1) physical and psychological
consequences of semi-starvation and nutritional rehabilitation 2) eating and weight changes during
and since completion of the experiment, and 3) quality of the participants’ lives following
completion of the experiment.
Results: Participants proceeded to lead interesting and productive lives, free of life-long adverse
effects. Personality differences, inferred from the Minnesota Multiphasic Personality Inventory,
likely influenced the severity of the psychopathological reactions to starvation. Many participants
reported maintaining a higher than normal weight and had abnormal eating habits for many months
and even years before returning to “normal” state.
Discussion: Reestablishment of normal body weight took significantly longer than suggested in the
original experiment, and might therefore constitute a factor contributing to the extended course of
illness and tendency to relapse in eating disorders. The preservation of energy and normal to high
activity levels in the presence of signs of severe weight loss and starvation and body image
disturbances seen in anorexia nervosa were not observed nor reported in the Minnesota Semi-
starvation Experiment.
Keywords: semi-starvation, healthy volunteers, anorexia nervosa, eating disorders, MMPI
RESEARCH ARTICLE
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 2 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
The Minnesota Experiment on the Biology of
Human Starvation was carried out at the
University of Minnesota in 1944/45 (1). It
remains the most systematic, ethically
justified study of experimental semi-
starvation to date. The study was designed to
monitor the physiological and psychological
effects of severe and prolonged food
restriction in 36 healthy young volunteer
males, recruited from conscientious objectors
during World War II. The experiment
involved a 3 months baseline control period,
6 months of semi-starvation, and 3 months of
controlled nutritional rehabilitation. The
results were published in 1950 as “The
Biology of Human Starvation” (1)
Researchers and clinicians have long
regarded this wartime Minnesota semi-
starvation study abbreviated from here on as
the 'Keys study' as a source of information for
understanding eating disorders (2-4). Our
aims in conducting this essentially qualitative
follow-up of the experimental subjects after
57 years were 1) to re-examine the acute, and
inquire into possible long-term, physical and
psychosocial effects from undergoing semi-
starvation in these originally healthy
volunteers; 2) to compare the data obtained
from the participants to symptoms observed
in eating disorder patients; 3) to obtain
information on the participants' quality of life.
The follow-up interviews were conducted in
2002, when most men were in their late 70’s
or early 80’s.
1.1. The Semi-Starvation Experiment
Toward the end of World War II, Dr. Ancel
Keys, well known for his invention of the “K
ration”, and Dr. Josef Brozek collaborated
with others from the Laboratory of
Physiological Hygiene to design a scientific
study on human semi-starvation. Their goal
was to characterize psychological and
physical consequences of starvation and
explore “refeeding” strategies, thus
permitting data-informed rehabilitation of
food-deprived civilians and prisoners of war.
From over 100 volunteering conscientious
objectors in the Civilian Public Service
(CPS), 36 men were chosen to participate in
the study. The inclusion criteria specified
good physical and mental health, normal
weight range, unmarried, the ability to get
along well with others under difficult
conditions, and a genuine interest in relief and
rehabilitation work. Each applicant completed
the Minnesota Multiphasic Personality
Inventory (MMPI), which had just been
published (5).
The average age of the 36 men who served as
subjects was 25 (range 20-33) years.
Physically, they were similar to their draftee
peers, although slightly taller and relatively
thinner than the average adult male in the US
at the time. Unlike the draftees, 25/36 men
belonged to the historic peace churches that
rejected participation in combatant military
service. They also differed intellectually and
educationally from their peers in the army.
All had at least one year of college, and 18
had college degrees (range of education was
13-19 years). Their IQ scores were almost
identical with scores of a graduate class in
advanced statistics at Columbia University.
On the Army General Classification Test of
IQ the 36 men averaged 2 SD above the mean
for Selective Service inductees, equivalent to
an IQ of 130. Thus, the participants
represented healthy young men with an above
average IQ.
During the 3-month control period, the men
ate an individualized diet in order to achieve
their ideal body weight according to height
and activity level. The mean caloric intake
per day during the control period was 3492
calories. The 6 months semi-starvation phase
was designed to bring about a loss of 25% of
their ideal body weight. Average calorie
intake during this phase was 1570 calories.
Mimicking the European wartime conditions,
the diet included large amounts of cabbage
and potatoes with minimal amounts of meat
and dairy. Thirty-two men reached the goal of
a 25% weight loss. Four subjects were
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 3 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
excluded from the study because they did not
lose the expected weight. For the
rehabilitation phase, the remaining 32 men
were divided into 4 groups with food
differing in caloric content in order to
determine what effect the caloric content of
food had on the rate of rehabilitation. After 3
months of controlled re-feeding, none of the
men had reached their control weights. At
this time, the men were allowed to eat an
unrestricted diet and some of the men left the
study. Twelve men remained in the
experiment to be monitored for an additional
2 months. Twenty-one subjects were
reevaluated eight months after the start of
rehabilitation and 8 men were reevaluated at
one year.
The men were required to continue or enroll
in educational programs at the University and
work 15 hours each week in study associated
jobs. They were also encouraged to continue
recreational and social activities according to
their ability. Required exercise included
walking 22 miles/week and 30 minutes/week
of 10% grade walking on a treadmill. They
were closely monitored throughout all phases
of the study with frequent physical and
psychological testing.
The Follow-up:
METHODS
2.1. Samples and Procedures:
The follow-up investigation was approved by
the Institutional Review Committee of the
University of Minnesota. “The Biology of
Human Starvation” (1) listed the names of the
men, but did not link them to their case
numbers. One participant, contacted by the
principal investigator, provided the addresses.
Of the 36 participants, six participants could
not be located and eleven were deceased. Of
the 19 participants who were contacted, all
agreed to participate in the follow-up. Our
sample includes two out of four men who did
not complete the study. Since one of the
eleven deceased participants had made a
detailed videotaped interview about his
experience in the study at the University of
Colorado in 1990 (6), it was possible to
include information about his extensive
changes in eating behavior and weight.
2.2. Interviews and Data Review:
After obtaining informed consent, two of the
investigators (EE and SS) completed
interviews of the participants by telephone
about their experiences during and following
the original experiment. On average,
interviews of participants lasted about 90
minutes. The interview was guided by a semi-
structured questionnaire containing open-
ended and specific questions and a checklist.
Brief notes were added as the interview
progressed. Each participant was questioned
about the following topics: the personal
history preceding participation in the original
experiment including their psychological
adjustment, physical health, and their weight
and eating patterns. Participants were queried
about their personal experiences and any
psychological, physical, cognitive, and
behavioral changes during the experiment.
Each participant was systematically
questioned about the following topics
experienced during starvation: 1. energy
level, lethargy, and fatigue. 2. food and
eating-related changes 3. non-food related
changes such as collecting things, smoking
changes, chewing gum, and 4. cognitive
changes. Each participant was asked whether
or not they recalled feeling depressed versus
non-depressed, and/or irritable. We
specifically focused on changes in eating
behavior and weight which occurred during
and after the experiment in each participant.
Several participants shared personal notes
with us that they had kept during the course
of the experiment. We explored the men’s
personal and professional lives following the
experiment, and any abnormal eating and
weight patterns and related physical and
psychological issues.
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 4 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
RESULTS
3.1. Demographics and Career Outcomes:
The mean age of the 19 out of 36 participants
at follow-up interview was 79 years with a
range from 75 to 83 years. All men were
college graduates, six had attained a Ph.D.
and one an M.A. Six were college professors,
four teachers, two ministers, an architect, a
lawyer, an engineer, and a social worker. All
felt they had led useful and interesting lives.
Six attributed aspects of their vocational and
avocational interests to their participation in
the study. Ten indicated that their
perceptions and perspectives regarding food
were permanently altered by the study
experience. All, but one, indicated they would
participate again.
After the experiment ten men went on to help
with world-wide relief efforts. One went on to
give lectures on world hunger and his
experience with starvation. Three men wrote
books and received various awards, and six
men travelled and worked in various parts of
the world.
Max Kampelman is the best known of the
study participants. He went on to a
successful career in law, teaching, and
politics, and eventually became a
distinguished career diplomat, culminating in
being awarded the Presidential Citizen’s
Medal in 1989 and the Presidential Medal of
Freedom in 1999. His autobiography
includes a description of his participation in
the semi-starvation study (7). He entered the
study with the same pacifist ideals as the
others as a 24 year-old law student. Despite
suffering the distressing physical
consequences of semi-starvation including
significant edema, lethargy, mild depression
and irritability, he was able to finish law
school and successfully took courses toward
an MA in political science during the
experiment.
3.2. Physical Changes:
Decades after participation, men recalled the
prominent physical sensations and changes
during semi-starvation described in the
original study. They all recalled a gradual
decrease in strength, coordination, and
endurance, a sense of lethargy paralleled by a
general curtailment of self-initiated
spontaneous activities. Twelve men recalled
they most dreaded the Johnson treadmill test,
which involved walking on a treadmill until
they felt they literally could not continue; two
fell with total exhaustion on completion of
the test. They avoided stairs and extra
exertion whenever possible. One man stated
“exercising was the hardest thing we did”.
None of the men interviewed reported periods
of increased activity.
Although the Keys report states that physical
symptoms gradually improved and the men
were generally back to a normal physical
state at 8 months, two participants stated that
it took longer, in one case up to two years, to
regain their former strength and stamina.
3.3. Eating Related Behavior During Semi-
starvation:
The men confirmed the Keys report that food
and eating and thoughts about and
preoccupation with food became the earliest
and most prominent focus and main topic of
conversation. Five men recalled collecting
recipes or cookbooks, ten reported distress
about the waste of food, seven enjoyed the
vicarious pleasure of watching others eat,
while others tried to avoid watching others
eat. Some reported dreaming about eating
forbidden foods and waking up feeling guilty.
During mealtimes, three reported creating odd
concoctions with their food, while others
added spices or additional water to their soup.
Some ate more rapidly, while others dawdled
over food and licked their dishes in order to
get the last morsel of food. Ten drastically
increased their consumption of coffee (up to
15 cups per day) and water in an attempt to
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 5 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
fight hunger. Fourteen men interviewed
reported chewing gum, although not all to
excess, and one man stated “we all chewed
gum”. The Keys researchers reported that 4
participants chewed gum compulsively, up to
40-60 packs a day (8). Four men interviewed
stated they began smoking.
One of our participants wrote in his personal
diary: “Several of us got talking about how
people play with their food…men paw over
the food, shove it from one side of the plate to
the other, spend gobs of time making
sandwiches, continue to jump up and down
fixing this and getting that.”
Four men recalled subverting their desire for
food by developing other habits, such as
collecting and even hoarding items they did
not particularly need, such as books or
trinkets. One man wrote: “I’ve been
wondering just why I want to buy books so
much lately. I’ve always liked to look at
books, but now I seem to want to go farther
and actually acquire them…I find I’d like to
buy clothes, too, some things that I really
don’t need. Not being able to buy food
whenever we want to may have something to
do with this”.
One man recognized and noted in his diary
that he could control his hunger feelings
somewhat by diversion activities: “My
hunger seems to have some relation to how
occupied I may be…I am hungry when I’m
not busy at something I am really interested
in.”
3.4. Eating Related Behavior During
Rehabilitation:
Many men remembered feeling a loss of
control over eating during the early re-feeding
period. Four men stated that they felt like
eating more or less continuously for a long
time. Six of nineteen men (32%) reported
binge eating (defined as eating large amounts
of food in a short period of time along with a
feeling of loss of control over eating),
particularly during the initial re-feeding
period when dietary restrictions were lifted.
Table 1. Eating Changes Resulting from Semi-starvation in
19 Follow-up Participants
Participants who admitted to binge-eating
N
o During semi-starvation
0
o During rehabilitation
6
Participants who admitted to “overeating” (not
binge-eating) during rehabilitation
11
Participants who vomited (involuntarily) during
rehabilitation after “overeating”
2
Participants hospitalized for gastric distension
after “overeating”
1
None of these men reported binge eating
before the experiment. One man overate so
much that he developed gastric distension and
had to be hospitalized. In addition, 11 (55%)
reported overeating either because they did
not feel satiated with normal meals or
because they continued to eat despite feeling
full, and two reported involuntarily vomiting
after eating large amounts of food.
When the prevalence of these abnormal
eating behaviors was recognized, the Keys
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 6 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
researchers reinstated some control over
access to food, except on weekends, when
they described a pattern of “weekend
gorging”. They estimated that the men
initially ate from 50- 200% more food on
weekends than during controlled re-feeding.
Eight months after the start of rehabilitation,
the Keys researchers obtained data on 14
participants and indicated that 10 of the 14
reported eating normal amounts of food and
showed less interest in food, although four
still thought a lot about food and continued to
eat large quantities.
At follow-up, the men did not recall for how
long the abnormal eating lasted. Interestingly,
they readily recalled their weights at various
times during and after the experiment.
3.5. Weight Changes
We calculated the body mass index (BMI)
using the weights and heights of the original
participants during all phases of the
experiment. During the control period, the
mean BMI of the 32 participants who
completed the experiment was 21.9 (SD 1.7),
and the range was 18.4-25.4. At the end of
the semi-starvation period the 32 men
remaining in the study had reached the goal
of 25% weight loss to a mean BMI of 16.4
(SD 0.9) and a range of 14.9-18.6.
Table 2. BMI of Semi-Starvation Subjects
Phase/Wk
N
Mean
SD
C-12
32
21.9
1.7
S-24
32
16.4
0.9
R-12
32
18.4
1.2
R-33
21
23.6
2.3
R-58
6
21.8
1.2
C: Wk of control phase; S: wk of starvation phase; R: wk of rehabilitation; N: number
of Subjects; BMI: Body Mass Index (kg/m2); SD: standard deviation
Patients with anorexia nervosa, a disorder
characterized by self-imposed restriction of
food intake leading to severe weight loss,
have a similar BMI range when they enter
treatment (9). After 12 weeks of controlled
re-feeding, the mean BMI of the Keys
participants was 18.4 (SD 1.2; range: 15.6-
20.7.), well below their control weights. At
that time most men left the formal part of the
experiment, except for twelve, who stayed for
another 2 months to be monitored on an
unrestricted diet. During the first two weeks
of unrestricted re-feeding, most participants
ate large amounts of food (estimated up to
11,000 calories per day), leading the Keys
researchers to re-institute dietary controls.
After a few weeks caloric intake seemed to
level off at more reasonable levels of between
3200-4500 calories per day, although some
continued to eat much more. Follow-up data
were obtained for 21 participants at 8 months
of rehabilitation, and for 6 participants at 58
weeks. Table 2 indicates that weights were
highest at week 33, with a mean BMI of 23.6
and a range of 19.6 to 29.2. By BMI
parameters, five participants were overweight
at this time (BMI 25-30), but none were
obese (BMI>30).
In the Keys study the participants lost 25% of
their control weights during semi-starvation
and the mean weight gain exceeding control
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 7 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
weights was 110% at R33. By one year (R
58) weight was obtained for only 8
participants, who had nearly returned to their
control weights. The fluctuations in body fat
were more extreme. During semi-starvation,
the men dropped to 30% of their control body
fat and during rehabilitation body fat rose to
approximately 140% (Fig.1).
Fig. 1. Body weight and body fat expressed as percentages of control values.
C, control; S, weeks of semistarvation, N=32; R, weeks of rehabilitation,
N=21 at R33, N=8 at R58.
From Keys et al. (1950). Copyright 1950 by the University of MN Press
By 58 weeks body fat was still somewhat
above control values in the 8 participants for
whom data were obtained. Based on data
gathered in the follow-up, changes in weight
were more drastic and lasted longer than
reported in the Keys study. The mean weight
gain exceeding control weights recalled by 16
follow-up participants was 22 pounds, i.e.
114% instead of 110% of control body
weight. Most took longer than 58 weeks to
return to their control weight. Seven men had
problems for 6 months to a year, 5 had
problems for 2 years, one had problems for 3
years, and one had problems that resolved in
4-5 years. Three men described minimal
problems with abnormal eating or being
overweight. Their weight and eating patterns
normalized within 6 months. Three men
never returned to their control weight.
The deceased participant who video-taped his
experience in 1990 (6) specifically addressed
the fact that he got “fat” after the experiment,
weighing from a low of 122 lbs during semi-
starvation to 225 lbs about six months later
(or 50 lbs above his control weight of 175
lbs). He described feeling hungry and eating
almost continuously for a long time. It took
him three years to return to his control weight
and a normal eating pattern. If his data are
included, the average weight gained above
control weight in 17 subjects is 27 lbs. or 116
% of control weight.
3.6. Cognitive Changes and Body
Perception:
On follow-up, all the men agreed with the
original study findings that they experienced
some form of intellectual inefficiency during
semi-starvation, with either decreased
concentration, rate of learning, or poor
judgment. Twelve participants on follow-up
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 8 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
recalled decreased motivation and discipline
in their intellectual pursuits. Interestingly, on
testing during the experiment no objective
loss in intellectual ability and no faults in
memory or logic were observed (1). Some
men carried a full academic load during the
experiment.
We inquired about changes in body
perception or body image. Two men
interviewed said that they perceived others,
including study staff, to be overweight during
semi-starvation. Two were unaware of their
own emaciated appearance although they
could see others getting thin. During
rehabilitation, seven men were concerned
about accumulation of fat in the abdomen and
buttocks and five of these men reported that
during or after the rehabilitation phase they
were bothered by how 'fat' they felt. Three
months after the start of rehabilitation one
man wrote in his personal notes: “During this
week I regained my top standardization
(control) weight: however it certainly is not
in the same places as the 138 pounds I had on
me when I came to Minneapolis. My arms,
thighs, buttocks and midsection, all feel fuller
than I can ever recall, my face also looks
fatter. However, these reactions may be
conditioned by what I got used to during
semi-starvation.” Six men denied having
concerns about changes related to their body.
3.7. Psychological Changes- With Analysis
and Review of MMPI Test Results:
The Keys researchers rated all men in good
mental health in the control period. During
semi-starvation all Keys participants were
described as experiencing psychological
changes: intense preoccupation with thoughts
of food, social introversion, loss of sexual
interest, decrease in self-initiated activity, and
emotional changes tending toward irritability
and depression. While all participants on
follow-up confirmed an intense preoccupation
with food, a reduction in sexual interest,
sociability, and self-initiated activity, only 15
of the 19 men recalled increased irritability
and only eight recalled feeling depressed
during semi-starvation or rehabilitation.
Remarkably, during rehabilitation the scores
indicating recovery from depression
correlated with calories received (Fig. 2).
Fig 2. Mean MMPI Depression ratings in Z,L,G,T groups with various calorie diets-
Z group with lowest calories, T group with highest calories.
C, control period; S, semi-starvation in wks; R, rehabilitation in wks;
N, number of participants
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 9 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
During early rehabilitation when diet was
tightly controlled (R6 and R12), those who
received the least calories (group Z) had high
depression scores for the longest time and
those who received the most calories (group
T) seemed to improve the fastest. During the
period of unrestricted food (R20 and R 32),
most men remaining in the study ate a high
number of calories and all Depression scale
scores rapidly returned towards normal.
The Minnesota Multiphasic Personality
Inventory (MMPI) was published by
Hathaway, a clinical psychologist, and
McKinley, a neuropsychiatrist, at the
University of Minnesota in 1943 (5); its use in
the Keys study was fortuitous. It permitted the
Keys researchers to summarize psychological
changes which occurred to varying degrees in
all participants under the term “semi-
starvation neurosis”, characterized by
significantly increased scores on the first
three MMPI scales of Hypochondriasis (Hs),
Depression (D), and Hysteria (Hy). High
scores on the first three scales had become
known as the “neurotic triad” after a
publication in 1943 on normative data of
Minnesota general medical outpatients by
McKinley and Hathaway (10).
3.8. The MMPI of 32 Study Completers
Which Formed the Basis for the “Semi-
starvation Neurosis”
Fig. 3. Mean Scores on the MMPI for 32 semi-starvation participants during the
control period (C), after 24 wks of starvation (S24), and after 12 and 33 wks
of rehabilitation (R12 and R33). Note: N was only 20 at R33.
Hypochondriasis (Hs); Depression (D); Hysteria (Hy); Psychopathic deviate
(Pd); Masculinity-femininity (Mf); Paranoia (Pa);
Psychasthenia (Pt); Schizophrenia (Sc); Hypomania (Ma).
From Keys et al. (1950)(1). Copyright 1950 by the University of MN Press
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 10 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
Fig. 3 displays MMPI scores for the 32 semi-
starvation participants who completed the
experiment. All scales were uncorrected for
K, a later test development that corrected
empirically for psychological defensiveness,
and had a mean score of 50 with a standard
deviation of 10 in the normative population.
The highest mean score during semi-
starvation was Depression (more than 2 SD
above normal), followed by Hysteria at 70.
This profile type suggests a depressive
reaction with anxiety in a hysteroid
personality (11). The Depression score and
MMPI profiles of 20 participants obtained
after 33 weeks of rehabilitation returned to
the “normal” pre-semi-starvation level. Of
note is that the Psychopathic deviate (Pd)
scale and the MMPI scales indicating
“psychotic” tendencies (Pa (paranoia), Pt
(psychasthenia), Sc (schizophrenia), and Ma
(hypomania) remained within normal limits
during all phases of the experiment in the
participants who completed the experiment.
The normal Pd score indicates a lack of
impulsivity, aggressiveness, rebelliousness
and a general lack of anti-social features, not
unexpected in conscientious objectors with
religious affiliations. In essence, these men
were normal as a group before the
experiment, the psychological changes during
semi-starvation were measurable on the
MMPI and formed the basis for inferring a
“semi-starvation neurosis”, and they returned
to normal values after the imposed stress.
Validity scales L, F, K in the MMPI
descriptions or profiles were not included,
and no K correction was made for the 5 scales
(Hs, Pd, Pt, Sc, and Ma), which could affect
interpretation of the MMPI’s for non-
volunteer samples.
According to Schiele and Brozek (8), 5 of the
32 study completers developed symptoms
that went beyond the range of “semi-
starvation neurosis”. They developed serious
clinical worsening during the starvation
phase, which was reflected in changes in their
MMPI. We were able to interview 2 of these
participants: One participant who had a mild
cyclothymic history prior to the experiment,
experienced several episodes during the
initial part of the starvation phase lasting no
more than a few days of exaggerated mood
changes consisting of elation followed by
“deep, dark depression”. During the last 3-4
weeks of the starvation phase he suffered a
severe depression and he thought he was
going “crazy”, although he was able to
continue with the experiment. The MMPI
indicated a high depression score and an
increase in the Pt and Sc scales consistent
with his psychological decompensation (such
a profile is consistent with that of patients
diagnosed as schizophreniform psychosis).
The severe depression gradually resolved
during the rehabilitation phase (8). At follow-
up, he denied having any further psychiatric
problems and went on to have a varied and
successful national and international
academic career.
Another participant had a history of being a
hard worker, driving himself hard to finish a
task, but then being “compelled to go to bed
at home for an extended period of
recuperation (8). He experienced a severe
depression and psychological decompensa-
tion at the end of the starvation period that
culminated in two attempts at unconscious”
versus accidental self-mutilation by chopping
off three fingers of his left hand while
chopping wood. He recovered quickly after a
brief hospitalization and was able to complete
the experiment. At follow-up, this man
described a successful life, but reported a
low-grade chronic depression for 4-5 years
prior to the interview related to stressful life
experiences.
3.9. The MMPI of Four Non-Completers.
The MMPI scores of the 4 men who failed to
complete the experiment due to their
significant dietary violations and failure to
lose sufficient amounts of weight were also
normal during the control period. They
showed the typical “semi-starvation neurosis”
increases in the Hypochondriasis, Depression
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 11 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
and Hysteria scales during the starvation
phase, yet 3/4 showed a significant increase
on the Pd scale during starvation, indicating
increased anger, impulsivity, aggressiveness
or passive-aggressiveness, rebelliousness, or
other anti-social features. There were also
marked elevations of the “psychotic” side of
the MMPI profile, although not reaching
abnormal values. Longitudinal MMPI profiles
for one of these participants under the
experimental conditions are shown in Fig. 4.
K-correction would have led to a psychotic-
like profile (Fig.4).
Fig. 4. Scores on the MMPI participant No. 232 during the control period (C), and
after 12 and 24 weeks of starvation (S12 and S24). Hypochondriasis (Hs);
Depression (D); Hysteria (Hy); Psychopathic deviate (Pd); Masculinity-
femininity (Mf); Paranoia (Pa); Psychasthenia (Pt); Schizophrenia (Sc);
Hypomania (Ma). From Keys et al. (1950)(1). Copyright 1950 by the
University of MN Press.
Two of the four men who failed to complete
the experiment due to dietary violations had
to be hospitalized psychiatrically because of
severe decompensation and “pre-psychotic”
symptoms. During the first few weeks of
semi-starvation one of the men began to have
strange dreams of eating senile and insane
people”. His MMPI profile suggests similar-
ity to patients diagnosed with a
schizoaffective psychosis (Fig 5).
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 12 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
Fig. 5. Scores on the MMPI for participant No. 234 during the control period (C),
and after 10 weeks of semi-starvation (S10). Hypochondriasis (Hs);
Depression (D);Hysteria (Hy); Psychopathic deviate (Pd); Masculinity-
femininity (Mf); Paranoia (Pa); Psychasthenia (Pt); Schizophrenia (Sc);
Hypomania (Ma). From Keys et al. (1950)(1). Copyright 1950 by the
University of MN Press.
He then began impulsively to break the
controlled diet, concealed his dietary
violations and began minor shoplifting sprees
stealing candy and trinkets that had no value
for him. He wrote voluminously and showed
a distinct flight of ideas. He then developed a
violent emotional outburst threatening suicide
and violence; he was admitted to a psychiatric
ward. The clinical picture suggested hypoma-
nia: he was overly talkative, elated and
emotionally labile. He quickly returned to
normal after being allowed to eat ad libitum.
We learned that this man died sometime after
the original experiment. We were unable to
get any follow-up information for him.
The other man began to use enormous
amounts of chewing gum (40-60 packs per
day) and later stole some food items. He
began to compulsively root in garbage cans
and actually ate garbage. Since he failed to
lose weight despite drastic cuts in his diet, he
was dropped from the experiment at the end
of the starvation period. Subsequently he ate
huge amounts of food, became sick and
vomited. He sought psychological help and
voluntarily sought admission to the
psychiatric ward twice in a short period of
time. "His symptoms subsided over a period
of weeks and he eventually made a
satisfactory adjustment” (8). We interviewed
this participant at follow-up. He told us he
gained about 30 pounds above his control
weight after the starvation period by
overeating and was bothered by looking fat. It
took him about a year to return to normal
eating and normal weight. He denied any
further psychiatric problems and eventually
earned a PhD.
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 13 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
4. Comparison to Eating Disorders
4.1. Importance of Temperament and
PersonalityRelationship to Nutritional
Status, Psychological State and Abnormal
Eating Pattern.
Personality and temperamental differences
have been described between the restricting
and bulimic subtypes of anorexia nervosa (12,
13). Those who lose weight by persistently
curtailing their food intake tend to be more
introverted and conscientious, more parsimo-
nious, better organized; in short they display
higher emotional self-control and self-
discipline, whereas those who develop binge-
eating and then vomit to lose weight tend to be
more outgoing, sometimes more impulsive,
and tend to be less controlled behaviorally and
emotionally
The rise in the scores on the “neurotic” scales
of the MMPI’s with prolonged semi-starvation
in the 32 men who completed the experiment is
similar to the elevation in the “neurotic triad"
observed in eating disorders (14). In general,
restricting anorexia nervosa patients show less
psychopathology on the MMPI than patients
who binge eat (14-16). In restricting anorexia
nervosa the depression scale is generally the
highest; no published study has shown a mean
profile exceeding 70 on a clinical scale (16).
Notably, Vitousek and Manke (16) have
pointed out that the mean semi-starvation
profile for the 32 participants who completed
the experiment in the Keys study looks
remarkably similar to the average profile for
restricting anorexia nervosa. Like the MMPIs
in the Keys study participants, once they
regained pre-starvation weights, the MMPI of
low weight anorexia nervosa patients (with and
without binge eating) becomes more normal
with weight gain in a hospital setting (17, 18),
indicating that the MMPI cannot be interpreted
without taking the clinical context into
account.
Patients who binge eat (both bulimic anorexia
nervosa patients and normal weight bulimia
nervosa patients) consistently show higher Pd
scores and sometimes higher scores in the
“psychotic” end of the MMPI (particularly the
Paranoid (Pa), Psychasthenia (Pt), and
Schizophrenia (Sc) scores), a profile frequently
associated with impulse and characterological
problems (14-16, 19). Interestingly, the MMPI
profile of female bulimia nervosa patients (14)
is similar to that of the male semi-starvation
subjects who failed to complete the experiment
due to dietary violations who showed marked
psychological worsening and developed binge
eating and kleptomania (Fig.4 and 5). The
main difference is a peak in the Hy scale in the
semi-starvation subjects not seen in the bulimia
nervosa patients.
4.2. Hunger, Preoccupation with Food and
Eating Patterns:
Keys and Brozek (1) erroneously assumed "a
decreased appetite" in anorexia nervosa and
believed that “continuous hunger and mental
absorption with food” were absent in anorexia
nervosa. In fact, the behaviors described in the
experimental participants, such as collecting
recipes and cookbooks, and hoarding food and
“strange” items, and even kleptomania, are
characteristic of anorexia nervosa. Anorexia
nervosa patients are preoccupied with and
dream about food, reflecting hunger sensations,
even if they may deny being hungry (2, 20).
When asked about appetite in one study
evaluating 44 hospitalized pre-treatment
anorexia nervosa patients, 54.5% claimed to
have no appetite, and 45.5% admitted to a
strong appetite, but were afraid of losing
control of their appetite (21).
Keys and Brozek (1) were also unaware that
abnormal eating patterns such as binge eating
can occur in anorexia nervosa, as observed by
Halmi et al (21). In a systematic study by
Casper et al. (12), the clinical importance of
bulimic behavior in anorexia nervosa was
analyzed leading to its recognition as a distinct
diagnostic subcategory (DSM IV). Food
restriction, due to dieting, often with
significant weight loss is a recognized
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 14 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
antecedent to most cases of binge eating and
bulimia nervosa (22, 23).
4.3. Energy Levels and Physical Activity
Patterns in the Minnesota Semi-Starvation
Study and Anorexia Nervosa.
All signs of physiological starvation described
in the Keys study, with the exception of the
markedly diminished energy levels and
physical activity patterns, have also been
described in anorexia nervosa (2-4).
As weight loss progressed, the Keys
researchers noted and the follow-up
participants vividly recalled the marked loss of
energy, increasing lethargy and weakness.
None of the follow-up participants recalled
periods of physical activation, even though the
Keys researchers noted in some short-lived
“quickening”, usually associated with transient
euphoric feelings lasting from a few hours to
several days followed by a “low” period.
This lethargy is in sharp contrast to the mostly
normal, if restless, energy levels and normal or
higher than normal activity levels in anorexia
nervosa (24, 25). Bruch (26) noted that “the
subjective feeling of not being tired and …the
drive for activity continues until the emaciation
is far advanced". In the Keys study the men's
alertness decreased, unlike the “paradoxical
liveliness” in anorexia nervosa already
described by Lasègue (27). It is interesting that
the Keys researchers recognized that this 'drive
for physical activity' seems to be a fundamental
clinical and psychopathological feature for
differentiating anorexia nervosa from the
lethargy and slow movements observed in
other conditions of human starvation (1).
Several studies have experimentally confirmed
normal or higher activity levels in severely
underweight and fasting anorexia nervosa
patients (28, 29). Physiologically these activity
levels appear to be starvation-dependent (30).
As a result, unlike the men in the Keys study
who dreaded exercise, anorexia nervosa
patients like to exercise, behavior with the
added benefit of reducing stress while
accelerating the desired weight loss (25, 31-
33).
4.4. Body Perception Changes
Two men interviewed said that they perceived
others, including study staff, to be overweight
during semi-starvation while two were
unaware of their own emaciated appearance.
During the rehabilitation phase seven men
were concerned about accumulation of fat in
the abdomen and buttocks and five of these
men reported they were bothered by how 'fat'
they felt. Nonetheless, these realistic concerns
about fat accumulation during weight gain are
different from the body image changes in
anorexia nervosa patients, who sometimes
describe themselves as “fat” despite being
extremely thin (34). In anorexia nervosa
patients, the tendency for increased fat
accumulation in the abdominal area during
rapid therapeutic hospital nutritional
rehabilitation is common (35).
DISCUSSION
The Minnesota Experiment on the Biology of
Human Starvation (1) measured and
documented for the first time that prolonged
caloric under-nutrition resulting in significant
loss of body weight in normal men led to down
regulation of physiological functions and
changes in psychological function, all
reversible with qualitatively and quantitatively
adequate food intake. Our follow-up evaluation
provides new data suggesting that the severe
food restriction resulted in abnormal eating
patterns and weight changes that were more
prolonged than reported by Keys et al (1),
given that the original follow was no longer
than 8 months.
Furthermore, our long-term follow-up study
suggests that voluntary severe food restriction
for 6 months was not associated with life-long
physical, cognitive or emotional adverse
effects, despite significant individual suffering
during the experiment. All but one man stated
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 15 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
they would again participate in the experiment.
All participants ultimately led interesting and
productive lives, perhaps not unexpected in
view of their voluntary participation as
conscientious objectors and the original
selection criteria for physically and
psychologically healthy subjects.
Previous publications have noted the relevance
of the Minnesota experiment to anorexia
nervosa, emphasizing the similarity to signs of
starvation in anorexia nervosa (2-4, 36) and
therapeutic benefits from recognizing signs of
starvation in anorexia nervosa (3,4). The study
by Kalm and Semba (37) is particularly
relevant because it involved tape-recorded
follow-up interviews of 18 Keys study
participants in 2003-2004, thus following and
overlapping with our 19 participant follow-up
in 2002. Their detailed description of the
Minnesota experiments provides a lively
account interspersed with the men's
recollections of the semi-starvation and
rehabilitation experiences. While their general
findings converge with our findings, our study
focuses more specifically on changes in eating,
weight, psychological state, and the relevance
to eating disorders.
Keys and Brozek (1) recognized from the
men's reactions during rehabilitation that
prolonged semi-starvation can result in
overeating and binge eating and hence
introduced controlled meals. Similar eating
patterns have been observed under conditions
of enforced starvation. Polivy et al (38)
compared Canadian WWII combat veterans
who were interned in German prisoners of war
(POW) camps with non-POW veterans. The
POW group reported significantly more binge
eating than the non-POW group; the greater the
weight loss the more binge eating years later.
Favaro et al (39) observed that former political
prisoners with severe weight loss who survived
Nazi concentration camps were at greater risk
of developing recurrent binge eating than the
ex-partisans with less weight loss. Binge eating
often, but not always, started immediately after
liberation and then persisted for many years, in
some for 51 years. In Holocaust survivors,
Sindler et al. (40) described intense
preoccupation with and sensitivity to food-
related issues. By contrast, a retrospective
study of Holocaust survivors in Israel did not
find that prolonged and severe starvation
resulted in the experience of binge eating
compared to a control group of immigrants,
matched for schooling and origin (41). The
sufferings of the malnourished concentration
camp survivors and the dangers of eating in
excess are movingly described in a paper by
Brooks (42).
Observations from the Keys study suggest that
food restriction does not lead to binge eating in
everyone. Personality and temperament seems
to play a part in the type of response. One
follow-up participant, Max Kampelman, who
described himself as having “considerable
ability to focus on goals and being a bit
stubborn”, was able to “concentrate when
others couldn’t study”. He finished law school
during the starvation phase and had the least
psychological worsening on interview and on
the MMPI (1,8). During rehabilitation he never
felt a loss of control over eating, never binge
ate, and had returned to normal eating and his
former weight within six months. Several other
men denied binge eating and described
returning to normal eating and weight quickly.
In contrast, the 4 participants who failed to
complete the experiment due to their
significant dietary violations and developed
binge eating and kleptomania, showed marked
psychological worsening during semi-
starvation. The Keys researchers recognized
that the emergence of individual differences in
basic personality make-up was associated with
the response to semi-starvation. They remarked
that “Ex post facto” it appeared that men with a
more stable personality makeup showed
minimal deterioration while those with latent
personality weaknesses developed more severe
symptoms” (8).
The Keys study authors found underlying
personality characteristics to be a factor in the
severity of MMPI changes with long-term
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 16 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
caloric under-nutrition. These observations
apply to eating disorders. The MMPI profiles
of Keys study participants who had difficulty
adhering to the requirements of semi-starvation
resemble those of patients with bulimia
nervosa (14), while the mean semi-starvation
profile of the 32 subjects who completed the
Keys experiment looks remarkably similar to
the average profile for restricting anorexia
nervosa patients (16). Also, like the return to
more normal MMPI’s associated with weight
regain in the Keys study, the MMPIs of low
weight anorexia nervosa patients, with and
without binge eating, become more normal
with weight gain (17, 18). Thus, the Keys
study supports the fact that a return to a normal
nutritional state is required for eating disorder
patients to improve psychologically.
Psychologically, the lethargy and emerging
depressive symptoms in the Keys study
participants stand in contrast to the mental
alertness and continued vigor expressed by
anorexia nervosa patients. Physically, loss of
energy, fatigue and muscle weakness
experienced by participants in the Keys study
are not observed until the cachectic end stage
of anorexia nervosa. For most of the disorder,
anorexia nervosa patients preserve their
energy, remaining active and mobile and not
infrequently show greater than normal activity
levels (2, 24, 30).
Another difference from anorexia nervosa was
the absence of pathological body perception
disturbances and disparagement in the Keys
study participants. Some men expressed
reasonable concern and critical evaluation of
their changing body shape during nutritional
rehabilitation. In anorexia nervosa, the critical
evaluation of their changing body tends to
intensify the fear of weight gain.
Gender differences may play a part in the
response to semi-starvation. The Keys subjects
were all male, whereas eating disorders occur
predominantly in females. This sex difference
might contribute to the symptoms in anorexia
nervosa, such as the body perception
disturbance. It might be involved in the
continued normal or high activity levels, since
severely food- restricted female compared to
male rats show higher running wheel activity
(43).
This follow-up study could have been
improved utilizing a more highly structured
interview format. One might also question
whether a sample of 20 of 36 original
participants is truly representative. There is,
however, no reason to believe that 19 men who
were interviewed plus the deceased participant
who had made a videotaped interview were
different from the 10 other men who had died
or the 6 men who could not be located.
Furthermore, the accuracy of information
based on retrospective recall when tested
against the facts has limitations. For example,
only 8/19 men we interviewed recalled feeling
depressed, although the Keys study reported
that all men had some measurable depressive
symptoms. Remarkably, all men spontaneously
remembered severe lethargy with weight loss
and a lack of motivation and surprisingly,
nearly all recalled periods of abnormal eating
and their maximum weights as well as the
period of time it took to return to their pre-
experiment body weights.
Acknowledgements:
Our co-author, Irving Gottesmann, PH.D., the
eminent behavior geneticist, contributed with
everlasting good humor his expertise and
counsel to the paper. His expertise is reflected
in the interpretation of the MMPI, his counsel
in the findings that both personality, i.e.
genetics and environmental stressors interact in
producing the clinical outcome.
Irving Gottesmann died on June 29th, 2016.
We thank Scott Crow, MD, Professor of
Psychiatry, University of Minnesota, for his
general support of the study and his reading of
the final manuscript. Permission to reproduce
the figures was obtained from the University of
Minnesota Press.
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 17 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
References
1. Keys A, Brozek J, Henschel A,
Mickelson O, Taylor H. The Biology of
Human Starvation. Minneapolis, MN:
University of Minnesota Press; 1950.
2. Casper RC, Davis JM. On the course of
anorexia nervosa. The American
Journal of Psychiatry. 1977 Sep;134
(9):974-8. PubMed PMID: 900307.
3. Garfinkel PE, Garner DM. Anorexia
Nervosa: A Multidimensional
Perspective. New York, NY: Bruner
Meisel U; 1982
4. Casper RC. Restless Activation and the
Drive for Activity in Anorexia Nervosa
may reflect a Disorder of Energy
Homeostasis. 2016 Int J Eat Disord.
Aug;49(8):750-2. PMID:27315579;
Epub 2016 Jun17.
5. Hathaway SR, McKinley JC. The
Minnesota Multiphasic Personality
Inventory (Rev. ed.). Minneapolis, MN:
University of Minnesota Press; 1943.
6. Office of Instructional Services.
Minnesota Semi-starvation Experiment
[Video]. Fort Collins, CO: Colorado
State University; 1990.
7. Kampelman MM. Entering New
Worlds: the memoirs of a private man
in public life. New York: Harper
Collings Publishers; 1991.
8. Schiele BC, Brozek J. Experimental
neurosis resulting from semistarvation
in man. Psychosomatic medicine. 1948
Jan-Feb;10(1):31-50. PubMed PMID:
18905659.
9. Beumont PJ, Al-Alami M, Touyz SW.
Relevance of a Standard Measurement
of Undernutrition to the Diagnosis of
Anorexia Nervosa: Use of Quetelet's
Body Mass Index (BMI). International
Journal of Eating Disorders.
1988;7:399-405.
10. McKinley JC, Hathaway SR. The
identification and measurement of
psychoneuroses in medical practice.
The Journal of the American Medical
Association. 1943;122:161-7.
11. Gillberstadt H, Duker J. A handbook
for clinical and actuarial MMPI
interpretation. Oxford, England: W.B.
Saunders Co.; 1965.
12. Casper RC, Eckert ED, Halmi KA,
Goldberg SC, Davis JM. Bulimia. Its
incidence and clinical importance in
patients with anorexia nervosa.
Archives of General Psychiatry. 1980
Sep;37(9):1030-5. PubMed PMID:
6932187.
13. Strober M. The significance of bulimia
in juvenile anorexia nervosa.
International Journal of Eating
Disorders. 1981;1:28-43.
14. Casper RC, Hedeker D, McClough JF.
Personality dimensions in eating
disorders and their relevance for
subtyping. Journal of the American
Academy of Child and Adolescent
Psychiatry. 1992 Sep;31(5):830-40.
PubMed PMID: 1400113.
15. Cumella EJ, Wall AD, Kerr-Almeida N.
MMPI-2 in the inpatient assessment of
women with eating disorders. Journal of
personality assessment. 2000
Dec;75(3):387-403. PubMed PMID:
11117153.
16. Vitousek K, Manke F. Personality
variables and disorders in anorexia
nervosa and bulimia nervosa. Journal of
abnormal psychology. 1994
Feb;103(1):137-47. PubMed PMID:
8040475.
17. Leon GR, Lucas AR, Colligan RC,
Ferdinande RJ, Kamp J. Sexual, body-
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 18 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
image, and personality attitudes in
anorexia nervosa. Journal of abnormal
child psychology. 1985 Jun;13(2):245-
57. PubMed PMID: 4008754.
18. McCormick LM, Keel PK, Brumm MC,
Watson DB, Forman-Hoffman VL,
Bowers WA. A pilot study of
personality pathology in patients with
anorexia nervosa: modifiable factors
related to outcome after hospitalization.
Eating and weight disorders : EWD.
2009 Jun-Sep;14(2-3):e113-20.
PubMed PMID: 19934624.
19. Hatsukami D, Owen P, Pyle R, Mitchell
J. Similarities and differences on the
MMPI between women with bulimia
and women with alcohol or drug abuse
problems. Addictive behaviors.
1982;7(4):435-9. PubMed PMID:
6963852.
20. Garfinkel PF. Perception of hunger and
satiety in anorexia nervosa.
Psychological medicine. 1974;4:309-
15.
21. Halmi KA, Goldberg SC, Eckert ED,
Casper RC, Davis JM. Pretreatment
evaluation in anorexia nervosa.
Proceedings of the National Institute of
Child Health and Human Development
and National Institute of Mental Health
on Anorexia Nervosa. Anorexia
Nervosa: Raven Press; 1977.
22. Eckert ED, Halmi KA, Marchi P, Grove
W, Crosby R. Ten-year follow-up of
anorexia nervosa: clinical course and
outcome. Psychological medicine. 1995
Jan;25(1):143-56. PubMed PMID:
7792349.
23. Fairburn CG, Welch SL, Doll HA,
Davies BA, O'Connor ME. Risk factors
for bulimia nervosa. A community-
based case-control study. Archives of
general psychiatry. 1997 Jun;54(6):509-
17. PubMed PMID: 9193191.
24. Casper RC. Behavioral activation and
lack of concern, core symptoms of
anorexia nervosa? The International
Journal of Eating Disorders. 1998
Dec;24(4):381-93. PubMed PMID:
9813763.
25. Casper RC. The 'drive for activity' and
"restlessness" in anorexia nervosa:
potential pathways. Journal of Affective
Disorders. 2006 May;92(1):99-107.
PubMed PMID: 16448703.
26. Bruch H. Eating disorders: obesity,
anorexia nervosa, and the person
within. New York: Basic Books; 1973.
27. Lasegue E. De l'anorexie hysterique.
Archives Generales de Medecine.
1873;1:385-403.
28. Casper RC, Schoeller DA, Kushner R,
Hnilicka J, Gold ST. Total daily energy
expenditure and activity level in
anorexia nervosa. The American
Journal of Clinical Nutrition. 1991
May;53(5):1143-50. PubMed PMID:
1850575.
29. Pirke KM, Trimborn P, Platte P, Fichter
M. Average total energy expenditure in
anorexia nervosa, bulimia nervosa, and
healthy young women. Biological
psychiatry. 1991 Oct 1;30(7):711-8.
PubMed PMID: 1958768.
30. Hebebrand J, Exner C, Hebebrand K,
Holtkamp C, Casper RC, Remschmidt
H, et al. Hyperactivity in patients with
anorexia nervosa and in semistarved
rats: evidence for a pivotal role of
hypoleptinemia. Physiology &
behavior. 2003 Jun;79(1):25-37.
PubMed PMID: 12818707.
31. Casper RC, Jabine LN. An eight-year
follow-up: outcome from adolescent
compared to adult onset anorexia
nervosa. Journal of Youth &
Adolescence. 1996;25:499-515.
Eckert E.D. et al. Archives of Psychology, vol. 2, issue 3, March 2018 Page 19 of 19
Copyright © 2018, Archives of Psychology. All rights reserved. http://www.archivesofpsychology.org
32. Davis C, Katzman DK, Kaptein S,
Kirsh C, Brewer H, Kalmbach K, et al.
The prevalence of high-level exercise in
the eating disorders: etiological
implications. Comprehensive
psychiatry. 1997 Nov-Dec;38(6):321-6.
PubMed PMID: 9406737.
33. Kron L, Katz JL, Gorzynski G, Weiner
H. Hyperactivity in anorexia nervosa: a
fundamental clinical feature.
Comprehensive psychiatry. 1978 Sep-
Oct;19(5):433-40. PubMed PMID:
679677.
34. Casper RC, Halmi KA, Goldberg SC,
Eckert ED, Davis JM. Disturbances in
body image estimation as related to
other characteristics and outcome in
anorexia nervosa. The British Journal of
Psychiatry; 1979 Jan;134:60-6. PubMed
PMID: 570073.
35. Mayer L, Walsh BT, Pierson RN, Jr.,
Heymsfield SB, Gallagher D, Wang J,
et al. Body fat redistribution after
weight gain in women with anorexia
nervosa. The American journal of
clinical nutrition. 2005 Jun;81(6):1286-
91. PubMed PMID: 15941877.
36. Russell SA. Hunger: An Unnatural
History. New York, NY: Basic Books;
2005.
37. Kalm LM, Semba RD. They starved so
that others be better fed: Remembering
Ancel Keys and the Minnesota
Experiment. Journal of Nutrition. 2005;
135:1347-1352.
38. Polivy J, Zeitlin SB, Herman CP, Beal
AL. Food restriction and binge eating: a
study of former prisoners of war.
Journal of Abnormal Psychology. 1994
May;103(2):409-11. PubMed PMID:
8040513.
39. Favaro A, Rodella FC, Santonastaso P.
Binge eating and eating attitudes among
Nazi concentration camp survivors.
Psychological Medicine. 2000
Mar;30(2):463-6. PubMed PMID:
10824666.
40. Sindler AJ, Wellman NS, Stier OB.
Holocaust survivors report long-term
effects on attitudes toward food. Journal
of nutrition education and behavior.
2004 Jul-Aug;36(4):189-96. PubMed
PMID: 15544727.
41. Bachar E, Canetti L, Berry EM. Lack
of long-lasting consequences of
starvation on eating pathology in
Jewish Holocaust survivors of Nazi
concentration camps. Journal of
Abnormal Pychology. 2005
Feb;114(1):165-9. PubMed PMID:
15709823.
42. Brooks J. 'Uninterested in anything
except food': the work of nurses feeding
the liberated inmates of Bergen-Belsen.
Journal of clinical nursing. 2012
Oct;21(19-20):2958-65. PubMed
PMID: 22834907.
43. Doerries LE, Stanley EZ, Aravich PF.
Activity-based anorexia: relationship to
gender and activity-stress ulcers.
Physiology and Behavior
1991:50(5):945-9.
44. Yilmaz Z, Hardaway JA, Bulik CM.
Genetics and epigenetics of eating
disorders. Advances in Genomics and
Genetics. 2015;5:131-50.
... Additionally, humans need food to survive and eating plays a large role in our daily routines. Abstaining from eating results in strong physical urges and increased thoughts about food and attention for food-related cues (Keys 1950;Eckert et al., 2018) that are very difficult to resist and impossible to fully overcome. Thus, individuals are prone to engage with their project of dietary restraint at all times: " [My] life went concretely so that I would just think all days long how I could skip eating as much as possible and how I could exercise enough" (Savuskoski et al., 2012, p. 79). ...
Article
Full-text available
Recovery from eating disorders is known to be difficult. Individuals with eating disorders are generally poorly responsive to change: their eating behaviour is rigid, and they are often inflexible as regards their eating-related goals. This opens up questions about their agency in eating. Why do individuals with eating disorders not “just stop” performing problematic eating behaviour, despite the enormous burden this behaviour may inflict in their lives? In this article, we seek to answer this question by providing a clear view of individuals’ own role in the development of and recovery from their eating disorders. Specifically, we adopt a situated perspective on agency to explain the poor responsiveness to change characteristic of these conditions. We suggest that in the pursuit of eating-related goals, individuals with eating disorders develop a project-like relation to eating, which not only pervasively changes the arrangement and content of their lives, but also their lives’ meaning. On this basis, we argue that the goal-inflexibility and rigidity of eating behaviour often ascribed to them might not be grounded in disorders or impairments of agency, but can in fact be understood in terms of their agency in eating. This is what we call the “life project account” of eating disorders.
... In addition, periods of binge eating reinforce further dieting due to fear of binge-related weight gain and increasing focus on weight control [21]. Restraint theory suggests that over time, the use of cognitive restraint to control eating may subvert a person's ability to identify hunger and satiety derived from their physiological cues [3,22,28]. That is, the more a person ignores their physiological hunger and satiety signals, the less likely they will notice and respond to these signals to guide eating. ...
Article
Full-text available
Purpose Bariatric surgery works, in part, by surgically changing signals of hunger and satiety to achieve weight loss. Not all patients experience optimal outcomes. One potential explanation is that post-surgery dieting may subvert the ability to identify physiological cues of hunger and fullness. Dieting behaviours (e.g. restriction/cognitive restraint) are correlated with disordered eating, and disordered eating implicated in poor outcomes. This study examines the experience of dieting after bariatric surgery. Method Seventeen adult participants who had undertaken bariatric surgery and residing in Australia participated in semi-structured interviews. Surgeries occurred in 2021 (n = 8), 2020 (n = 4), 2019 (n = 2), and one participant each had surgery in 2014, 2009, and 2004. Thematic analysis elicited themes related to post-operative dieting. Results All participants reported chronic pre-surgery dieting. Lifestyle change was the overarching post-surgical theme comprising (i) flexibility (e.g. allowing food, intuitive eating), and (ii) control, comprising surgery control (e.g. set portions, surgery instilled control) and dieting control (e.g. discipline, restriction/restraint). Descriptions of lifestyle change often mirrored pre-surgery descriptions of dieting. Conclusion Post-surgery lifestyle change appears to encompass a tension between flexible/adaptive approaches to eating and the need to maintain control. Control may emerge as practices that mirror pre-surgery dieting with the potential to interfere with adaptive eating behaviours or promote disordered eating. Dieting behaviours may be a precursor to the development of disordered eating. Health care practitioners should regularly assess dieting behaviour post-surgery to enable early intervention where warranted. Future research should consider how post-surgery re-emerging dieting may be identified and measured to aid in intervention. Graphical Abstract
... Bulimia nervosa is an eating disorder that develops in 20-30% of patients with AN [30]. Single participants of the Minnesota Starvation Experiment had an overshooting weight gain and/or developed abnormal eating behaviour lasting up to several years after the end of the experiment [31,32]. ...
Article
Full-text available
Off-label treatment of a 15-year-old female patient with anorexia nervosa with human recombinant leptin (metreleptin) for nine days was associated with self-reported increments of appetite and hunger resulting in rapid weight gain, substantial improvement of eating disorder cognitions and of depression. The results further substantiate the effects of metreleptin on both AN and depression. We contrast these results with the widespread view that leptin is an anorexigenic hormone. Randomized controlled trials are warranted to confirm the described effects.
... The pathogenesis of AN involves several genetic, neurobiological, psychological and socio-cultural factors, and increasing evidence points to an important role in metabolic dysfunctions, including disturbances at the level of glucose, insulin and IGF-1 regulation [40][41][42]. The lower insulin and HOMA-IR levels observed in AN1 than in AN2 and CG probably allow for the involvement of counter-regulatory mechanisms of long-term starvation. ...
Article
Full-text available
Anorexia nervosa (AN) is a widespread, metabo-psychiatric disorder with high relapse rates, comorbidity, and mortality. Many regulatory proteins and neurohormones studied to date play essential roles in the etiopathogenesis of eating disorders and the maintenance of psychopathological symptoms. Nevertheless, the regulatory and pathophysiological mechanisms of AN are still poorly understood. In the presented study, the plasma levels of apelin-13 (APE-13) and asprosin (ASP), as well as carbohydrate metabolism parameters and psychometric parameters, were evaluated in low-weight adolescent female patients with AN (AN1), after partial weight normalization (AN2) and in an age-matched healthy control group (CG) were evaluated. APE-13 levels were higher in the AN1 group than in the post-realimentation and the CG group. APE-13 levels were independent of insulin and glucose levels. Plasma ASP levels increased with increasing body weight in patients with AN, correlating with the severity of eating disorder symptoms in emaciation. The presented data suggest that APE-13 and ASP may be AN’s biomarkers-regulation of eating behavior by APE-13 and ASP, the close relationship between them and emotional behavior, and changes in neurohormone levels in patients with eating and affective disorders seem to support these hypotheses. Moreover, their plasma levels seem to be related to the severity of psychopathological symptoms of eating disorders.
Article
Postoperative eating behaviors and unhealthy weight control measures are leading causes of bariatric surgery complications. Candidates for bariatric surgery and individuals with eating disorders may share common risk factors, such as a history of dieting, and/or being bullied or teased for their weight. Binge-eating disorder, night eating syndrome, and bulimia nervosa are the most common eating disorders among candidates for bariatric surgery before the operation. Malnutrition, stress, and intense fear of weight gain can lead to the development of an eating disorder after surgery as well. Plugging, grazing, loss of control eating, dumping, and food avoidance are specific disordered behaviors that may present after bariatric surgery. To improve physical and psychological outcomes for individuals under their care, clinicians can screen for these disordered behaviors and distinguish them from a healthy diet and the expected postsurgical course.
Article
Analysis of actual transactions in a grocery store show that households enrolled in U.S. government’s Supplementary Nutritional Assistance Program (SNAP) purchase more unhealthy hedonic food, compared to food secure households. SNAP households purchased more unhealthy hedonic food regardless of whether they used government funds or cash to pay for their purchases. Three follow-up simulated grocery shopping studies were designed to understand the mechanisms underlying this food consumption pattern by SNAP households. SNAP households, compared to food secure households, reported stronger food craving. Stronger food craving was associated with lower unhealthiness perception of hedonic food. Overall, the results suggest that food cravings can reduce perceptions of unhealthiness and thus increase unhealthy food purchases. Interventions that reduce the effect of food cravings on shopping decisions might reduce unhealthy consumption.
Chapter
Everybody eats, and what we eat – or do not – affects the brain and mind. There is significant general, applied, academic, and industry interest about nutrition and the brain, yet there is much misinformation and no single reliable guide. Diet Impacts on Brain and Mind provides a comprehensive account of this emerging multi-disciplinary science, exploring the acute and chronic impacts of human diet on the brain and mind. It has a primarily human focus and is broad in scope, covering wide-ranging topics like brain development, whole diets, specific nutrients, research methodology, and food as a drug. It is written in an accessible format and is of interest to undergraduate and graduate students studying nutritional neuroscience and related disciplines, healthcare professionals with an applied interest, industry researchers seeking topic overviews, and interested general readers.
Article
Introduction Anorexia nervosa poses an important bioethical dilemma, since patients often refuse treatment despite the danger that this poses to their health, and it is not clear that their decision is autonomous. The aim of this study was to investigate the perceptions/performance of psychiatrists and clinical psychologists regarding the capacity and involuntary hospitalisation of patients with anorexia nervosa. Methods Seven psychiatrists, four clinical psychologists, and one third-year resident psychologist were interviewed. A qualitative research approach based on grounded theory was used. Results The data analysis showed that these professionals articulate patient care around one main category - hospitalisation as a last resort and the search for voluntariness, which implies a change in the usual healthcare dynamics. Around this central category, some important concepts emerge: role stress, informal coercion, weight, family and chronicity. Conclusions It is concluded that the difficulty of reconciling professional demands can undermine the quality of care and job satisfaction itself, which highlights the need for reflection and research into the foundations of the responsibilities assumed.
Article
Full-text available
2016 The Authors. International Journal of Eating Disorders Published by Wiley Periodicals, Inc. (Int J Eat Disord 2016).
Article
Full-text available
Eating disorders (EDs) are serious psychiatric conditions influenced by biological, psychological, and sociocultural factors. A better understanding of the genetics of these complex traits and the development of more sophisticated molecular biology tools have advanced our understanding of the etiology of EDs. The aim of this review is to critically evaluate the literature on the genetic research conducted on three major EDs: anorexia nervosa (AN), bulimia nervosa (BN), and binge eating disorder (BED). We will first review the diagnostic criteria, clinical features, prevalence, and prognosis of AN, BN, and BED, followed by a review of family, twin, and adoption studies. We then review the history of genetic studies of EDs covering linkage analysis, candidate gene association studies, genome-wide association studies, and the study of rare variants in EDs. Our review also incorporates a translational perspective by covering animal models of ED-related phenotypes. Finally, we review the nascent field of epigenetics of EDs and a look forward to future directions for ED genetic research.
Article
This paper reports on a study comparing clinical symptomatology, premorbid functioning, family relations, life stress, and parent medical and psychiatric history in two matched groups of juvenile anorexia nervosa patients, one with and one without the syndrome of bulimia. In comparison to nonbulimic anorectics, bulimics evidenced higher levels of core anorectic sympomatology and were more likely to show affective disturbance and alchohol use. Bulimia was more strongly associated with affective instability and other signs of behavioral deviance during childhood. The family environment of bulimics was characterized by significantly greater conflict and negativity, less cohesion, and less structure than the nonbulimic anorectic family, and bulimics experienced more stressful life changes prior to illness onset than did nonbulimic patients. Parents of bulimic anorectics also reported significantly higher levels of marital discord, were rated as more emotionally distant from their daughters, and exhibited greater psychiatric morbidity and physical health problems. Results were consistent with the author's hypothesis that the disposition toward bulimia is rooted in early disturbances in ego functioning, personality adaptation, and other maladaptive factors. Study of the pathological effects of psychological, genetic, and physiological factors on appetite controlling mechanisms in anorectic subgroups deserves further investigation.
Article
Many a medical man has wished for an easily applicable measuring device which would identify and characterize the psychoneurotic patient with a minimum use of the time consuming interview technic that is conventional in the psychiatric approach. One may not desire to deal with the psychoneuroses in one's practice, but the physician or surgeon is indeed insensitive to the problem or very young in the profession who has not been plagued by his inability appropriately to assess the role of the neurotic element in some of his patients.Without wishing to belabor the topic of the neuroses unnecessarily, because much has been published on it, we should like to remind the reader of a few general points that are pertinent to this report. Competent internists have estimated variously that from 30 to 70 per cent of the ambulatory patients who appear for medical attention come primarily because of one or
Article
Background. Prisoners in Nazi concentration camps lived through extreme situations that included starvation. We test our hypothesis that there is a greater lifetime presence of binge eating among survivors from concentration camps than in a control group.
Article
Aims and objectives. The aim of this article is to explore the work of nurses feeding and helping liberate the inmates of Bergen-Belsen concentration camp in the spring of 1945. Background. A considerable amount has been published on the relief of Belsen, but the majority of the research has focused on the medical staff and the army in general. The exception to this is an article published by Ellen Ben-Sefer, but its analysis of the actual work of the nurses is limited. Design. The data are explored through the medium of nurses' work, especially feeding work and its place in the historiography. Methods. This article will offer an analysis of archival material, including official reports and personal testimony and published primary sources. Results. Nurses were very much hidden from the liberation picture, partly through government policy, partly because of gendered ideologies and partly because of the desire of many to hide their memories. However, the data identify the pivotal role of nurses in the saving of lives and rehabilitation of inmates. This article demonstrates this role through the work of feeding. Conclusion. The nurses had to continually negotiate and renegotiate their place in the liberation. As women, their place at Belsen was problematic and often thwarted by the ideologies of the day. Nevertheless, the services of registered nurses during the war had proved indispensible. The Allied governments and Royal Army Medical Corps were ultimately to rely on their professional expertise. The nurses’ work in the feeding of starving inmates demonstrates their value. Relevance to clinical practice. This article demonstrates the importance of nurses’ feeding work and their role in the caring of people’s humanity.
Article
Quetelet's Index provides a measure of nutritional status which is both objective and easy to calculate. A review of the normative data published in the literature indicates that a BMI value of 16 provides a convenient criterion for the recognition of significant undernutrition. This value was applied successfully to a group of 150 anorexia nervosa patients. On the other hand, a BMI of over 18 appeared appropriate in the case of a small group of 22 bulimia nervosa subjects.