Psychiatry Research 112 (2002) 101–110

Depression and anxiety symptoms in relation to anthropometry and

metabolism in men
Ann-Charlotte Ahlberga,*, Thomas Ljungb, Roland Rosmondb, Bruce McEwenc,
¨
Goran ˚
Holmb, Hans Olof Akesson a
¨
, Per Bjorntorpb

a
¨
Department of Psychiatry, Sahlgrenska University Hospital, University of Goteborg, ¨
Goteborg, S-413 45 Sweden
b
¨
Cardiovascular Institute, Sahlgrenska University Hospital, University of Goteborg, ¨
Goteborg, S-413 45 Sweden
c
Department of Neurobiology, Rockefeller University, New York, NY, USA

Received 11 May 2000; received in revised form 11 February 2002; accepted 19 February 2002

Abstract

Depression is associated with an increased risk of developing cardiovascular disease and type 2 diabetes mellitus.
Abdominal obesity is also a high risk factor for these diseases. Therefore, symptoms of depression and anxiety were
examined in relation to abdominal obesity. A total of 59 middle-aged men volunteered for measurements with the
Hamilton Depression Scale (HDS), the Montgomery–Asberg ˚ Depression Rating Scale (MADRS), the Beck Depression
Inventory (BDI) and the Hamilton Anxiety Scale (HAS). These results were examined in relation to body mass
index (BMI), waistyhip ratio (WHR) and sagittal abdominal diameter, a measurement of intra-abdominal fat mass,
and metabolic variables. Men with WHR)1.0 (ns26) in comparison with men with normal WHR (-1.0, ns33)
showed significantly higher sum scores in all the scales used. There were positive correlations between the sum
scores of all the depression scales and the WHR or the sagittal abdominal diameter. BMI correlated comparatively
weakly only with the HDS. The correlations with the WHR remained when the influence of BMI was eliminated,
suggesting that obesity is less involved than centralization of body fat. Insulin and glucose were significantly related
to the HDS. Morning cortisol levels were negatively related to the BDI and (borderline) to the MADRS, suggesting
perturbations of the regulation of the hypothalamic–pituitary–adrenal axis. We conclude that men with abdominal
obesity have symptoms of depression and anxiety.
䊚 2002 Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Depression; Anxiety; Men; Waistyhip ratio (WHR); Hypothalamic–pituitary–adrenal axis

*Corresponding author. Tel.: q46-31-3421661; fax: q46-31-826540.

E-mail address: ann-charlotte.ahlberg@brevet.nu (A.-C. Ahlberg).

0165-1781/02/$ - see front matter 䊚 2002 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0 1 6 5 - 1 7 8 1 Ž 0 2 . 0 0 1 9 2 - 0
102 A.-C. Ahlberg et al. / Psychiatry Research 112 (2002) 101–110
1. Introduction
Observations by Kretschmer as early as 1921
have shown that subjects with a pychnic body
build are frequently depressed. In these studies,
several anthropometric measurements were per-
formed, including the waist and hip circumfer-
ences, allowing retrospective calculation of the
waistyhip circumference ratio (WHR), indicating
abdominal obesity. Such persons also showed a
delay in removal of a glucose load from the
circulation (Hirsch, 1932), indicating impaired
glucose tolerance, and disturbances of the regula-
tion of the autonomic nervous system (Hertz,
1931), and were susceptible to developing gout,
stroke and atherosclerosis (Kretschmer, 1921).
Recent research has shown that abdominal obe-
sity, as estimated by the WHR, is a powerful,
independent predictor for cardiovascular disease,
type 2 diabetes mellitus and stroke in both men
and women. Abdominal obesity is also associated
with insulin resistance, impaired glucose tolerance,
dyslipidemia and hypertension, which might be
the immediate triggers for disease development
(Bjorntorp,
¨ 1987).
Similar to abdominal obesity, depression has
been shown to be a powerful risk factor for
cardiovascular disease and type 2 diabetes mellitus
(Eaton et al., 1996; Pratt et al., 1996). The
observations by Kretschmer open up the possibility
that the WHR and its associated risk factors might
carry the risk for development of these diseases
that are prevalent in depression.
Previous studies have suggested that subjects
with abdominal obesity display traits of depression
and anxiety, expressed as mood changes, sleep
disturbances and frequent use of antidepressants
and anxiolytics (Lapidus et al., 1989; Wing et al.,
1991; Rosmond et al., 1996; Rosmond and Bjorn- ¨
torp, 1998). However, these studies have appar-
ently not utilized standardized, internationally
evaluated instruments for psychiatric examination,
and the results are therefore not conclusive.
Interestingly, abdominal obesity displays similar
neuroendocrine abnormalities (Rosmond and
¨
Bjorntorp, 1998; Rosmond et al., 2000) to those
observed in depression and anxiety (Carroll, 1982;
Roy-Byrne et al., 1986; Gold and Chrousos, 1998),
although less pronounced. Such abnormalities
might explain the mechanisms whereby elevated
WHR and its associated risk factors appear, which
seems to be due to peripheral hormonal perturba-
tions induced by the neuroendocrine abnormalities
(Bjorntorp,
¨ 1996). We therefore wanted to reex-
amine with established instruments whether
depression and anxiety are associated with an
accumulation of abdominal fat mass, and whether
this may be coupled to the similar neuroendocri-
ne–endocrine abnormalities found in both condi-
tions. This is of interest because depression and
anxiety may be associated with risks for serious
disease via induction of elevated WHR with asso-
ciated metabolic risk factors.
The aim of the present study was therefore to
examine if depressive and anxiety symptoms are
associated with abdominal obesity. This was
expected to provide information on whether the
risk for development of somatic disease in depres-
sion is mediated via abdominal obesity. Another
question was to examine whether the similar neu-
roendocrine abnormalities in depression–anxiety
and abdominal obesity are a common denominator
for the appearance of excess central body fat with
associated risk factors.
The results from the psychiatric examinations
were evaluated not only in relation to WHR, but
also to obesity (BMI), and several other variables
previously known to be strongly associated with
WHR and BMI. The somatic data are reported
separately (Ljung et al., 2000). Here we report the
psychiatric associations to these data.
2. Methods
2.1. Material
A total of 59 men volunteered to take part in
the study and were recruited by an advertisement
in a local newspaper. They were selected from a
larger sample of approximately 200 men to obtain
two groups with different WHR, but comparable
BMI. This was done to examine the influence of
fat distribution (WHR) without the influence of
obesity (BMI). The results of this selection pro-
cedure were that the WHR values did not overlap,
while BMI values were not different. In the adver-
 A.-C. Ahlberg et al. / Psychiatry Research 112 (2002) 101–110
tisement, men with perceived weight problems
were invited to participate. There was no mention
of psychiatric examinations in the advertisement.
Selection bias of psychosocial or employment
status was not involved. No subject was excluded
due to the presence of alcohol problems, which
was carefully examined. Mean age was 52.5"3.5
years (mean"S.D.). The men were all fully
employed and reported no major psychosocial
problems and moderate alcohol consumption. Sev-
en men were smokers, and in the subsequent
subgrouping of the WHR, there were three smokers
in the high-WHR group and four in the low.
The study was performed after informed consent
and was approved by the Ethical Committee of
¨
the University of Goteborg.
2.2. Procedures
The psychiatric interview was performed over
the course of approximately 1 h without knowledge
of other measurements. Four scales were used: the
Hamilton Depression Scale (HDS) (Hamilton,
1960); the Montgomery–Asberg˚ Depression Rat-
ing Scale (MADRS) (Montgomery and Asberg, ˚ cortisol concentration after dexamethasone
1979); the Beck Depression Inventory (BDI)
(Beck et al., 1961); and the Hamilton Anxiety
Rating Scale (HAS) (Hamilton 1959). The items
included in these scales are found in Tables 1–4.
The results of the BDI and HDS scales correlated
with a correlation coefficient of 0.80 (95% confi-
dence interval, 0.69–0.88). Corresponding values
for comparisons of the BDI and the MADRS were
0.79 (0.66–0.87), and those of the HDS and the
MADRS, 0.82 (0.71–0.89).
WHR was measured in the overnight fasting
state with the subjects standing in a normal respi-
ratory position. The waist circumference was
measured horizontally halfway between the lower
costal arc and the iliac crest, and the hip circum-
ference over the widest part of the gluteo-femoral
region. Measurements in cm were divided to obtain
the WHR. The men were divided into two groups,
those with WHR -1.0 (ns33), (mean"S.D.)
0.92"0.044 considered to be a normal value
(World Health Organization, 1998), and those with
WHR )1.0 (ns26), 1.06"0.035.
103
Body weight was recorded in underwear to the
nearest 0.1 kg, and height to the nearest cm, and
the BMI was calculated. BMI (mean"S.D.) for
the group with WHR -1 was 27.7"2.8, and for
the group with WHR )1, 30.2"2.3 kgym2.
Sagittal abdominal diameter, a close approxi-
mation of visceral (intra-abdominal) fat mass
(Kvist et al., 1988), was measured as the distance
between the examination table and the highest
point of the abdomen in the supine position.
An oral glucose tolerance test was performed
with 100 g of glucose in the fasting state in the
morning, with measurement of glucose and insulin
before, and 30, 60, 90 and 120 min after glucose
ingestion.
Cortisol in serum was measured on four morn-
ings before breakfast. Urine was collected for 24
h for cortisol measurements and expressed as total
excretion, or corrected for creatinine. On the final
day at approximately 22:00 h, 0.5 mg of dexame-
thasone (Decadron䉸; Merck, Darmstadt, Germany)
was taken and cortisol was measured again the
next morning. Dexamethasone suppression was
measured as the difference between the average of
the four non-inhibited cortisol levels, and the
administration.
Cortisol, testosterone, insulin and insulin-like
growth factor I (IGF-I) were measured by com-
mercially available radioimmunoassays (Orion
Diagnostica, Turku, Finland; ICN Medicals, Costa
Mesa, CA, USA; Pharmacia, Uppsala, Sweden;
Nichols, San Juan Capistrano, CA, USA,
respectively).
Blood pressure, determined using a mercury
manometer, and heart rate were measured twice in
the supine position after 5 min of rest and with 1
min between measurements, and then averaged.
Glucose, triglycerides, total, low-density lipo-
protein (LDL) and high-density lipoprotein (HDL)
cholesterol were measured by automated methods
as previously described (Ljung et al., 2000).
2.3. Statistical methods
Fisher’s permutation test was applied (Bradley,
1968; Oden´ and Wedel, 1975). Two-tailed tests of
significance were used in analyses and P-0.05
104 A.-C. Ahlberg et al. / Psychiatry Research 112 (2002) 101–110

Table 1
Scores for items in the Hamilton Depression Scale for the total group of men (ns59) and those with a WHR above (ns26) or
below (ns33) 1.0

Item All WHR)1.0 WHR-1.0 P-value

1 Depressed mood 0.1"0.4 0.2"0.5 0.1"0.3 NS
2 Guilt 0.1"0.3 0.1"0.3 0.1"0.4 NS
3 Suicide 0.1"0.2 0.1"0.3 0.0"0.2 NS
4 Insomnia, initial 0.1"0.4 0.2"0.5 0.1"0.3 NS
5 Insomnia, middle 0.2"0.5 0.2"0.6 0.1"0.3 NS
6 Insomnia, delayed 0.3"0.6 0.3"0.7 0.2"0.6 NS
7 Work and interests 0.1"0.4 0.2"0.5 0 0.030
8 Retardation 0.1"0.3 0.2"0.5 0 NS
9 Agitation 0.0"0.1 0.0"0.2 0 NS
10 Anxiety, psychic 0.3"0.5 0.5"0.6 0.1"0.3 0.006
11 Anxiety, somatic 0.2"0.5 0.4"0.6 0.1"0.2 0.005
12 Gastrointestinal symptoms 0.2"0.4 0.3"0.5 0.1"0.3 0.023
13 Somatic, general 0.2"0.4 0.4"0.5 0.1"0.2 0.006
14 Genital 0.1"0.3 0.2"0.4 0.0"0.2 0.021
15 Hypochondriasis 0.1"0.4 0.2"0.6 0 NS
16 Insight 0.0"0.2 0.1"0.3 0 NS
18 Diurnal variation 0.1"0.2 0.1"0.3 0 NS
Means"S.D. Items 4–6, 12–14 and 16 measured on a scale from 0 to 2, and items 1–3, 7–11, 15 and 17 from 0 to 4. WHR,
waistyhip circumference ratio.

was considered significant. With the aid of logistic
regression analysis, the items were examined to
explore the significance of differences between
groups in multivariate tests. Spearman’s rank test
was utilized to examine correlations.
Mantel’s test was applied in order to assess the
correlation between WHR and other variables
when the influence of BMI was eliminated. The
expected value of HDS sum was determined by
multiple regression analysis as ysaq
b=min(WHR,1)qc=max(WHR,y1.0) in order
to elucidate the change below and above WHRs
1.
3. Results
The scores for the HDS for all men, as well as
for those with WHR above and below 1.0, are
listed in Table 1. The following six items showed
a significantly higher score for those with WHR
)1 compared to WHR -1: work and interests
(Ps0.03); anxiety, psychic (Ps0.006); anxiety,
somatic (Ps0.005); gastrointestinal symptoms
(Ps0.023); somatic, general (Ps0.006); and
genital (Ps0.021).
In the MADRS (Table 2), three of the 11 items
registered showed higher values in men with WHR
)1 compared to those with WHR -1, in anxiety
(Ps0.007), concentration disturbance (Ps0.015)
and global assessment of degree of illness (Ps
0.005).
In the BDI (Table 3), those with WHR )1 had
a significantly higher score than those with lower
WHR in four items, namely: irritability (Ps
0.021); work inhibition (Ps0.024); fatigability
(Ps0.012); and loss of libido (Ps0.017).
Table 4 shows the results of registrations with
the HAS instrument. Men with WHR )1.0
showed, in comparison to men with WHR -1.0,
more respiratory symptoms. Other individual items
showed no differences. When all psychic items
(1–6) were considered together, there was a bor-
derline (Ps0.07) higher value for the men with
elevated WHR. With the somatic items (7–14),
this difference was fully significant (Ps0.027),
and this was also the case for the sum of all the
items (Ps0.028).
The sum score of the items included for the
three depression scales was calculated and is pre-
sented in Table 5. None of the men reached scores
 A.-C. Ahlberg et al. / Psychiatry Research 112 (2002) 101–110 105

Table 2
˚
Scores for items in the Montgomery–Asberg Depression Rating Scale for the total group of men (ns59) and those with a WHR
above (ns26) or below (ns33) 1.0

Item All WHR)1.0 WHR-1.0 P-value

1 Depressed mood 0.1"0.4 0.3"0.6 0.1"0.2 NS
2 Anxiety 0.3"0.5 0.4"0.6 0.1"0.3 0.007
3 Reduced sleep time 0.4"0.7 0.5"0.7 0.3"0.6 NS
4 Loss of appetite 0.0"0.1 0.0"0.2 0 NS
5 Concentration disturbance 0.2"0.5 0.4"0.7 0.1"0.2 0.015
6 Loss of initiative 0.1"0.4 0.2"0.5 0.0"0.1 NS
7 Reduced emotional engagement 0.1"0.4 0.2"0.5 0.0"0.1 NS
8 Depressive thoughts 0.1"0.3 0.2"0.4 0.1"0.2 NS
9 Suicidal thoughts 0.0"0.2 0.1"0.3 0.0"0.1 NS
10 Mood 0.1"0.3 0.2"0.4 0.1"0.2 NS
11 Global assessment of degree of illness 0.2"0.5 0.4"0.8 0.0"0.1 0.005
Means"S.D. Items 1–11 measured on a scale from 0 to 3. WHR, waistyhip circumference ratio.

above the levels suggested for a diagnosis of
depression. As can be observed, men with WHR
)1 had a significantly higher sum score than men
with WHR -1 in all three scales used.
Since the various items in each of the three
scales are expected to measure overlapping phe-
nomena, multivariate analyses were performed to
examine which items were independently, signifi-
cantly (P-0.05) different between the two groups
of men, which were as follows: HDS, work and
interests, anxiety, psychic, and genital; MADRS,
global assessment of illness; and BDI, irritability
and fatigability.
Correlations between the mean scores of the
depression scales and anthropometric, endocrine,
metabolic and hemodynamic measurements are

Table 3
Scores for items in the Beck Depression Inventory for the total group of men (ns59) and those with a WHR above (ns26) or
below (ns33) 1.0

Item All WHR)1.0 WHR-1.0 P-value

1 Mood 0.0"0.2 0.0"0.2 0.0"0.2 NS
2 Pessimism 0.1"0.6 0.2"0.6 0.1"0.3 NS
3 Sense of failure 0.1"0.3 0.1"0.4 0.1"0.3 NS
4 Lack of satisfaction 0.1"0.3 0.1"0.3 0.1"0.3 NS
5 Guilty feelings 0.1"0.2 0.0"0.2 0.1"0.3 NS
6 Sense of punishment 0.1"0.3 0.1"0.4 0.0"0.2 NS
7 Self-hate 0.1"0.4 0.2"0.5 0.1"0.3 NS
8 Self-accusation 0.3"0.7 0.3"0.8 0.3"0.7 NS
10 Crying spells 0.1"0.3 0.1"0.3 0.1"0.3 NS
11 Irritability 0.3"0.7 0.6"0.9 0.1"0.3 0.021
12 Social withdrawal 0.1"0.4 0.3"0.5 0.1"0.3 NS
13 Indecisiveness 0.0"0.3 0.1"0.4 0 NS
14 Body image 0.2"0.6 0.2"0.6 0.2"0.1 NS
15 Work inhibition 0.2"0.4 0.3"0.6 0.1"0.3 0.024
16 Sleep disturbance 0.5"1.0 0.7"1.0 0.4"0.9 NS
17 Fatigability 0.5"0.5 0.7"0.5 0.3"0.5 0.012
18 Loss of appetite 0.1"0.3 0.1"0.3 0.2"0.4 NS
19 Weight loss 0.3"0.1 0.3"0.5 0.4"0.9 NS
20 Somatic preoccupation 0.2"0.4 0.3"0.5 0.1"0.3 NS
21 Loss of libido 0.2"0.4 0.4"0.5 0.1"0.3 0.017
Means"S.E.M. Items 1–21 measured on a scale from 0 to 3. WHR, waistyhip circumference ratio.
106 A.-C. Ahlberg et al. / Psychiatry Research 112 (2002) 101–110

found in Table 6. The descriptive results of meas- Table 5

urements of the somatic variables have been pre- Sum score of the various items used in the three scales for
depression
sented separately (Ljung et al., 2000). The HDS
mean score correlated significantly with the WHR, Scale WHR-1 WHR)1 P-value
BMI, sagittal abdominal diameter, insulin and glu-
HDS 1.2"1.8 4.1"4.1 -0.001
cose in the fasting state and during oral glucose MADRS 0.9"1.1 2.7"3.4. 0.007
tolerance test (OGTT), and negatively with morn- BDI 2.7"2.7 4.9"4.7 0.044
ing cortisol and (borderline) HDL. The mean score
Values are means"S.D. Abbreviations: HDS, Hamilton
of the MADRS correlated significantly with WHR, ˚
Depression Scale; MADRS, Montgomery–Asberg Depression
sagittal diameter, sum of glucose values during Rating Scale; BDI, Beck Depression Inventory; WHR, waisty
OGTT, and correlated to a borderline level with hip circumference ratio.
fasting insulin and glucose, sum of insulin values
during OGTT, and negatively with morning serum BMI was eliminated. The increase in HDS sum
cortisol. The BDI mean score showed a significant with increasing WHR was not significantly differ-
negative correlation with morning serum cortisol, ent for WHR -1 vs. WHR G1.
and a borderline correlation with WHR, sagittal
abdominal diameter, and sum of glucose values 4. Discussion
during OGTT. The HAS mean score correlated
with the sum of glucose values, negatively with The men studied here, from the point of view
morning cortisol and borderline with fasting of the relationship between psychological–psychi-
glucose. atric variables and WHR, have previously been
Using Mantel’s test, it was assessed that there examined as far as somatic variables are concerned
was a significant correlation between WHR and (Ljung et al., 2000). In the study presented here,
anxiety, psychic (Ps0.021), and between WHR we found that men with WHR )1.0, in compari-
and HDS sum (Ps0.035) when the influence of son with men with WHR -1.0, showed signifi-

Table 4
Scores for items in the Hamilton Anxiety Rating Scale for the total group of men (ns59) and those with a WHR above (ns26)
or below (ns33) 1.0

Item All WHR)1.0 WHR-1.0 P-value

1 Anxiety 0.4"0.8 0.5"1.0 0.2"0.6 NS
2 Tension 0.5"0.8 0.6"0.4 0.3"0.7 NS
3 Fear 0.0"0.3 0.1"0.4 0 NS
4 Sleep disturbances 0.4"0.9 0.5"1.0 0.3"0.8 NS
5 Concentration difficulties, memory 0.5"0.8 0.7"1.0 0.3"0.7 NS
6 Depressive mood 0.3"0.7 0.3"0.8 0.3"0.6 NS
7 Somatic muscular 0.6"1.0 0.7"1.1 0.4"0.9 NS
8 Somatic sensory 0.6"1.1 0.8"1.4 0.4"1.0 NS
9 Circulatory 0.3"1.3 0.3"0.7 0.3"1.6 NS
10 Respiratory 0.2"0.6 0.5"0.8 0 0.007
11 Gastrointestinal 0.3"0.7 0.5"0.9 0.2"0.4 NS
12 Urogenital 0.3"0.6 0.3"0.7 0.3"0.6 NS
13 Autonomic 0.3"0.6 0.3"0.6 0.2"0.7 NS
14 Behavior during interview 0.0"0.3 0.1"0.4 0 NS
Sum of 1–6 2.0"2.5 2.7"3.1 1.4"1.9 0.070
Sum of 7–14 2.6"3.2 3.5"4.1 1.9"2.2 0.027
Total sum score 4.5"5.3 6.2"6.8 3.0"3.0 0.028
Means"S.D. Items 1–6 are considered as psychic and 7–14 as somatic symptoms. WHR, waistyhip circumference ratio. 0–5,
no anxiety; 6–14, minor anxiety; )15, major anxiety.
 A.-C. Ahlberg et al. / Psychiatry Research 112 (2002) 101–110 107

Table 6
Correlations between the mean scores of the depression and anxiety scales and anthropometric, endocrine, metabolic and hemody-
namic determinations (ns59)

HDS MADRS BDI HAS

Rho P Rho P Rho P Rho P
WHR 0.413 -0.01 0.318 -0.05 0.243 -0.10 0.211 NS
BMI 0.277 -0.05 0.201 NS 0.153 NS y0.043 NS
Sagittal abdominal diameter 0.427 -0.01 0.339 -0.05 0.261 -0.10 0.159 NS
Insulin 0.298 -0.05 0.235 -0.10 0.066 NS 0.040 NS
Sum insulin OGTT 0.198 NS 0.229 -0.10 0.043 NS y0.075 NS
Glucose 0.383 -0.01 0.236 -0.10 0.213 NS 0.252 -0.10
Sum glucose OGTT 0.441 -0.01 0.362 -0.01 0.260 -0.10 0.258 -0.05
Testosterone y0.085 NS y0.126 NS y0.077 NS y0.014 NS
IGF-1 y0.018 NS 0.035 NS 0.102 NS 0.100 NS
HDL y0.255 -0.10 y0.192 NS y0.076 NS y0.031 NS
SBP 0.148 NS 0.139 NS 0.040 NS 0.147 NS
DBP y0.007 NS 0.028 NS y0.111 NS 0.046 NS
HR y0.053 NS y0.042 NS y0.024 NS 0.083 NS
TG 0.135 NS 0.144 NS y0.210 NS y0.004 NS
Cholesterol 0.063 NS 0.116 NS y0.140 NS 0.086 NS
Morning serum cortisol y0.335 -0.05 y0.221 -0.10 y0.365 -0.01 y0.262 -0.05
Urinary cortisol (mmolymol creatinine) y0.041 NS 0.030 NS y0.119 NS 0.056 NS
Urinary cortisol (nmolyperiod) 0.132 NS 0.101 NS 0.082 NS 0.094 NS
Dexamethasone suppression 0.074 NS 0.159 NS 0.114 NS y0.017 NS
Spearman rank correlation coefficient. Abbreviations: HDS, Hamilton Depression Scale; MADRS, Montgomery–Asberg ˚ Depres-
sion Rating Scale; BDI, Beck Depression Inventory; HAS, Hamilton Anxiety Scale; WHR, waistyhip circumference ratio; BMI,
body mass index; OGTT, oral glucose tolerance test; IGF-1, insulin-like growth factor 1; HDL, high-density lipoprotein cholesterol;
SBP, DBP: systolic, diastolic blood pressure; HR, heart rate; and TG, triglycerides. Morning serum cortisol is the mean value of
four measurements between 08:45 and 09:00 h on different days. Dexamethasone suppression was calculated as the mean morning
salivary cortisol value minus the inhibited salivary morning cortisol value after dexamethasone.

cantly higher scores in several of the items
measured, as well as in mean scores when the
HDS, MADRS and BDI were utilized for meas-
urements of depressive symptoms, and the HAS
for anxiety. These scales have been thoroughly
evaluated and are widely used internationally. It
therefore seems clear that a particularly elevated
WHR is associated with depressive and anxiety
symptoms.
The summed values of the HAS were only fully
significant for the somatic anxiety values in the
HAS scores, particularly respiratory symptoms. An
increased volume of intra-abdominal fat may well
cause some respiratory difficulties, and this rela-
tionship might then be explained by elevated
masses of intra-abdominal fat.
The HDS and MADRS contain items estimating
psychic anxiety (numbers 10 and 2, respectively)
and these showed significantly higher values for
the men with WHR)1.0, which was, however,
not found with the HAS. The HDS and MADRS
probably measure overlapping phenomena with the
HAS. It therefore seems likely that psychic anxiety
shows an association with WHR. This was also
the case for somatic anxiety in the HDS scale.
Major depression and anxiety disorder are highly
prevalent and frequently comorbid diagnoses
(Breier et al., 1986; Kessler et al., 1996; Fava et
al., 2000; Kaufman and Charney, 2000). Depres-
sion and anxiety share many overlapping symp-
toms, including fatigue, impaired concentration,
irritability, sleep disturbance and somatization, in
addition to subjective experiences of nervousness,
worry and restlessness. (Ninan, 1999). They also
seem to share a common pathophysiology (Weiss
et al., 1994). The main error discussed in the
108 A.-C. Ahlberg et al. / Psychiatry Research 112 (2002) 101–110
literature as being responsible for anxiety and
depression is the dysregulation of the noradrenerg-
ic and 5-hydroxytryptamine systems, along with
the stress hormones, with hyperactivity of neurons
that utilize corticotropin releasing factor (CRF).
CRF has repeatedly been reported to be elevated
in the cerebrospinal fluid of patients with depres-
sion (Nemeroff et al., 1984; Plotsky et al., 1998).
As many as 75% of patients with major depression
have been reported to have a hyperactive hypotha-
lamic-pituitary-adrenal (HPA) axis, as character-
ized by hypercortisolemia (Laird and Benefield,
1995; Nemeroff, 1998; Ninan, 1999). HPA-axis
dysfunction has also been reported in some cases
of generalized anxiety disorders, as evidenced by
dexamethasone suppression test results (Tiller et
al., 1988; Avery et al., 1995). There are probably
both genetic predisposition and environmental
influences that act upon the neural circuits that
mediate stressyfear responsiveness and mood.
The men examined were recruited by an adver-
tisement in a local newspaper and might therefore
have been a selected group, not representative of
the population at large. The WHR, BMI and
sagittal diameter of the group of men with WHR
-1.0 are, however, not significantly different from
those of a larger cohort of randomly selected men
at a comparable age, recruited from the same city
(Rosmond et al., 1998), and may therefore be
considered to be representative of the general
population from these aspects.
WHR is an approximation of centralization of
body fat stores. Sagittal abdominal diameter is an
anthropometric estimate of intra-abdominal, vis-
ceral fat mass (Kvist et al., 1988), and also showed
significant associations to the mean score of the
HDS, MADRS and, borderline with BDI, but not
with HAS. BMI, an index of total fat mass, and
therefore of obesity, showed a significant relation-
ship to the HDS only. When the influence of BMI
was eliminated, the relationships to WHR
remained. It therefore seems likely that the depres-
sive symptoms according to these scales are more
closely related to centralization of body fat stores
than to obesity per se. Relationships were also
found between the HDS and fasting insulin and
glucose values. These are well-known associations
with abdominal obesity (Bjorntorp,
¨ 1993), and the
significant relationships disappeared when adjusted
for WHR.
The measurements of BMI, WHR and sagittal
abdominal diameter have previously shown asso-
ciations to abnormalities in HPA-axis regulation,
particularly when morning cortisol is low in diur-
nal curves (Rosmond et al., 1998). In the men
studied in this report, where cortisol measurements
were directly compared with WHR, men with
higher WHR had lower morning cortisol values
(Ljung et al., 2000). This was also found in
relation to the depression scales in this report.
Lower morning cortisol indicates a malfunction of
the HPA axis (McEwen, 1998; Rosmond et al.,
1998). Under such conditions, total diurnal cortisol
secretion may not show any abnormalities (Ros-
mond et al., 1998), as found in this work with
urinary output of cortisol. It therefore seems pos-
sible that the correlations found here between
WHR and depressive symptoms might have been
associated with perturbations of the regulation of
the HPA axis.
Alcohol consumption could be involved in the
observations reported in several ways. Alcohol
directly activates the HPA axis (Cicero, 1980),
and may well be a precipitating or accompanying
problem in depression. Although notoriously
uncertain, reports from the participating men did
not indicate that alcohol was involved. This can,
however, not be definitely excluded. The few
smokers (ns7) were distributed essentially equal-
ly in both subgroups, and smoking should therefore
not have been of significance for the results.
Other confounding factors should also be con-
sidered. We have previously found that psychoso-
cial and socioeconomic handicaps are followed by
perturbations of the regulation of the HPA axis
and elevated WHR (Lapidus et al., 1989; Larsson
et al., 1989). This factor was not examined in
detail. However, it seems unlikely that such factors
were involved in the men studied, who were fully
employed and apparently without major psycho-
social problems.
In summary, this study shows that elevated
WHR, a measurement of body fat centralization
and metabolic perturbations, is associated with
depressive symptoms in several established scales.
Anxiety is also apparently involved, although
 A.-C. Ahlberg et al. / Psychiatry Research 112 (2002) 101–110
weaker. Abdominal obesity and depression–anxi-
ety show similar neuroendocrine abnormalities,
that are likely to induce accumulation of fat in
central depots and other risk factors for cardiovas-
cular disease and type 2 diabetes mellitus. It is
hypothesized that this is the pathogenic pathway
that may lead from frequent depressive periods to
somatic disease.
After this study was completed, a report was
published showing that visceral fat mass, measured
with computerized tomography scan, was elevated
in patients with melancholic depression (Thakore
et al., 1997). The results of this study seem to
strengthen the notion that regulation of the HPA
axis, which is clearly abnormal in melancholic
depression, is involved in visceral fat
accumulation.
Acknowledgments
This study was supported by grants from the
John D. and Catherine T. MacArthur Foundation
Research Network on Socioeconomic Status and
Health and the Swedish Medical Research Council
(Project No B96-19X-00251-34B).
References
Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., Erbaugh,
J., 1961. An inventory for measuring depression. Archives
of General Psychiatry 4, 561–571.
¨
Bjorntorp, P., 1987. The associations between obesity, adipose
tissue distribution and disease. Acta Medica Scandinavica
723, 121–134.
¨
Bjorntorp, P., 1993. Visceral obesity: a ‘civilization syndrome’.
Obesity Research 1, 206–222.
¨
Bjorntorp, P., 1996. The regulation of adipose tissue distribu-
tion in humans. International Journal of Obesity and Related
Metabolic Disorders 20, 291–302.
Bradley, J.W., 1968. Distribution-free Statistical Tests. Prentice
Hall, London, pp. 68–86.
Breier, A., Charney, D.S., Heninger, G.R., 1986. Agoraphobia
with panic attacks. Development, diagnostic stability, and
course of illness. Archives of General Psychiatry 43,
1029–1036.
Cicero, T.J., 1980. Sex differences in the effects of alcohol
and other psychoactive drugs on endocrine function. In:
Israel, Y., Kalant, O., Kalant, H. (Eds.), Research Advances
in Alcohol and Drug Problems. Plenum, New York, pp.
544–593.
Carroll, B.J., 1982. The dexamethasone suppression test for
melancholia. British Journal of Psychiatry 140, 292–304.
109
Eaton, W.W., Armenian, H., Gallo, J., Pratt, L., Ford, D.E.,
1996. Depression and risk for onset of type II diabetes. A
prospective population-based study. Diabetes Care 19,
1097–1102.
Fava, M., Rankin, M.A., Wright, E.C., Alpert, J.E., Nierenberg,
A.A., Pava, J., Rosenbaum, J.F., 2000. Anxiety disorders in
major depression. Comprehensive Psychiatry 41, 97–102.
Gold, P.W., Chrousos, G.P., 1998. The endocrinology of atyp-
ical depression: relation to neurocircuitry and somatic con-
sequences. Proceedings of the Association of American
Physicians 111, 22–34.
Hamilton, M., 1959. The assessment of anxiety states by
rating. British Journal of Medical Psychology 32, 50–55.
Hamilton, M., 1960. A rating scale for depression. Journal of
Neurology, Neurosurgery, and Psychiatry 23, 56–62.
Hertz, T., 1931. Pharmacodynamische Untersuchungen an
¨ Neurologie und Psychia-
Konstitutionstypen. Zeitschrift fur
trie 3, 134–139.
Hirsch, O., 1932. Blutzuckerbelastungsproblem zur blutchem-
¨
ischen Fundierung der Korperbautypen. ¨ Neu-
Zeitschrift fur
rologie und Psychiatrie 4, 140–147.
¨
Kretschmer, E., 1921. Korperbau und Character. Harcourt,
New York. English version: Physique and Character, 1936.
Kaufman, J., Charney, D., 2000. Comorbidity of mood and
anxiety disorders. Depression and Anxiety 12 (Suppl. 1),
69–76.
Kessler, R.C., Nelson, C.B., McGonagle, K.A., Liu, J., Swarts,
M., Blazer, D.G., 1996. Comorbidity of DSM-III-R major
depressive disorder in the general population: results from
the US National Comorbidity Survey. British Journal of
Psychiatry Suppl. 30, 17–30.
Kvist, H., Chowdhury, B., Grangard, ˚ ´ U., Sjostrom,
U., Tylen, ¨ ¨
L., 1988. Total and visceral adipose-tissue volumes derived
from measurements with computed tomography in adult
men and women: predictive equations. American Journal of
Clinical Nutrition 48, 1351–1361.
¨
Lapidus, L., Bengtsson, C., Hallstrom, ¨ T., Bjorntorp,
¨ P., 1989.
Obesity, adipose tissue distribution and health in women—
results from a population study in Gothenburg, Sweden.
Appetite 12, 25–35.
¨
Larsson, B., Seidell, J., Svardsudd, K., Welin, L., Tibblin, G.,
¨
Wilhelmsen, L., Bjorntorp, P., 1989. Obesity, adipose tissue
distribution and health in men—the study of men born in
1913. Appetite 13, 37–44.
Ljung, T., Holm, G., Friberg, P., Andersson, B., Marin, ˚ P.,
¨
Dallman, M., McEwen, B., Bjorntorp, P., 2000. The activity
of the hypothalamic–pituitary–adrenal axis and the sympa-
thetic nervous system in relation to waistyhip circumference
ratio in men. Obesity Research 8, 487–495.
McEwen, B.S., 1998. Protective and damaging effects of stress
mediators. New England Journal of Medicine 338, 171–179.
˚
Montgomery, S.A., Asberg, M., 1979. A new depression scale
designed to be sensitive to change. British Journal of
Psychiatry 134, 382–389.
´ A., Wedel, H., 1975. Arguments for Fischer’s permu-
Oden,
tation test. Annals of Statistics 3, 518–520.
110 A.-C. Ahlberg et al. / Psychiatry Research 112 (2002) 101–110
Pratt, L.A., Ford, D.E., Crum, R.M., Armenian, H.K., Gallo,
J.J., Eaton, W.W., 1996. Depression, psychotropic medica-
tion, and risk of myocardial infarction. Prospective data
from the Baltimore ECA follow-up. Circulation 96,
3123–3129.
¨
Rosmond, R., Bjorntorp, P., 1998. Psychiatric ill-health of
women and its relationship to obesity and body fat distri-
bution. Obesity Research 6, 338–345.
Rosmond, R., Lapidus, L., Marin, ¨
P., Bjorntorp, P., 1996.
˚
Mental distress, obesity and body fat distribution in middle-
aged men. Obesity Research 4, 245–252.
¨
Rosmond, R., Dallman, M.F., Bjorntorp, P., 1998. Stress-related
cortisol secretion in men: relationships with abdominal
obesity and endocrine, metabolic and hemodynamic abnor-
malities. Journal of Clinical Endocrinology and Metabolism
83, 1853–1859.
¨
Rosmond, R., Holm, G., Bjorntorp, P., 2000. Food-induced
cortisol secretion in men in relation to anthropometric,
metabolic and hemodynamic variables in men. Interntional
Journal of Obesity and Related Metabolic Disorders 24,
416–422.
Roy-Byrne, P.P., Uhde, T.W., Post, R.M., Gallucci, W., Chrou-
sos, G.P., Gold, P.W., 1986. The corticotropin-releasing
hormone stimulation test in patients with panic disorder.
American Journal of Psychiatry 143, 896–899.
Thakore, J.H., Richards, P.J., Reznek, R.H., Martin, A., Dinan,
T.G., 1997. Increased intra-abdominal fat deposition in
patients with major depressive illness as measured by
computed tomography. Biological Psychiatry 41,
1140–1142.
Wing, R.R., Matthews, K.A., Kuller, L.H., Meilahn, E.N.,
Plantinga, P., 1991. Waist to hip ratio in middle-aged
women. Associations with behavioral and psychosocial fac-
tors and with changes in cardiovascular risk factors. Arte-
riosclerosis and Thrombosis 11, 1250–1257.
World Health Organization, 1998. Obesity. Preventing and
Managing the Global Epidemic. Report of a WHO consul-
tation on obesity, Geneva, 3–5 June 1997, WHOyNUTy
NCDy98.1. WHO, Geneva.