Cross-sectional relationship between chronic stress and mineral concentrations in hair
of elementary school girls
Barbara Vanaelst 1,2, Nathalie Michels 1, Inge Huybrechts 1,3, Els Clays 1, Maria R Flórez 4,5,
Lieve Balcaen 2,4, Martin Resano 5, Maite Aramendia 2,6, Frank Vanhaecke 4, Noellie Rivet 7,
Jean-Sebastien Raul 7, Anne Lanfer 8, Stefaan De Henauw 1,9
1
Department of Public Health, Ghent University, De Pintelaan 185, 2 block A, 9000 Ghent,
Belgium
2
Research Foundation – Flanders (FWO), Egmontstraat 5, 1000 Brussels, Belgium
3
Dietary Exposure Assessment Group, International Agency for Research on Cancer
(IARC/WHO), 150 Cours Albert Thomas, 69372 Lyon CEDEX 08, France
4
Department of Analytical Chemistry, Ghent University, Krijgslaan 281, S12, 9000 Ghent
Belgium
5
Department of Analytical Chemistry, University of Zaragoza, c/ Pedro Cerbuna 12, 50009,
Zaragoza, Spain
6
Centro Universitario de la Defensa - Academia General Militar de Zaragoza. Carretera de
Huesca s/n, 50090 Zaragoza, Spain
7
Department of Toxicology, Institute of Legal Medicine, University of Strasbourg, 11 Rue
Humann, 67085 Strasbourg, France
8
BIPS Institute for Epidemiology and Prevention Research, Achterstrasse 30, 28359 Bremen,
Germany
9
Department of Health Sciences, Vesalius, University College Ghent, Keramiekstraat 80,
9000 Ghent, Belgium
Corresponding author:
Barbara Vanaelst, Department of Public Health, Ghent University, University Hospital, Block
A, 2nd floor, De Pintelaan 185, B-9000 Ghent, Belgium, tel: +32 9 332 36 85, fax: +32 9 332
49 94, e-mail address: barbara.vanaelst@ugent.be
Short title: Chronic stress and hair minerals in elementary school girls
Laboratory information:
Elemental hair analysis: Atomic & Mass Spectrometry (A&MS) research unit, Department of
Analytical Chemistry, Ghent University, Krijgslaan 281 - S12, 9000 Ghent, Belgium
http://www.analchem.ugent.be/A&MS/
1
Hair cortisone determination: Toxicology Laboratory, Institute of Legal Medicine, Strasbourg
University, 11 Rue Humann, 67085 Strasbourg Cedex, France
http://www-ulpmed.u-strasbg.fr/iml/laboratoire.htm
2
Abstract
Chronic stress exposure is associated with diverse negative health outcomes. It has been
hypothesized that stress may also negatively affect the body’s mineral status. This study
investigates the association between chronic stress and long-term mineral concentrations of
calcium (Ca), copper (Cu), iron (Fe), magnesium (Mg), phosphorus (P) and zinc (Zn) in scalp
hair among elementary-school girls. Complete information on child-reported stress estimates
(Coddington Life Events Scale (CLES)), hair cortisone and hair mineral concentrations, and
predefined confounders in the stress-mineral relationship (i.e. age, body mass index (BMI),
physical activity, diet, hair colour and parental education) was provided cross-sectionally for
140 girls (5-10yrs old). The relationship between childhood stress measures (predictor) and
hair minerals (outcome) was studied using linear regression analysis, adjusted for the
abovementioned confounders. Hair cortisone concentrations were inversely associated with
hair mineral concentrations of Ca, Mg, Zn and the Ca/P ratio. Children at risk by life events
(CLES) presented an elevated Ca/Mg ratio. These findings were persistent after adjustment
for confounders. This study demonstrated an independent association between chronic stress
measures and hair mineral levels in young girls, indicating the importance of physiological
stress-mineral pathways independently from individual or behavioural factors. Findings need
to be confirmed in a more heterogeneous population and on longitudinal basis. The precise
mechanisms by which stress alters hair mineral levels should be further elucidated.
Keywords: hair, cortisone, life events, stress, mineral, child
3
1
1. Introduction
2
Stress is defined as the process in which environmental demands or events are interpreted and
3
appraised by the individual as taxing or exceeding his/her resources, resulting in
4
psychological and biological changes with risk for disease [1]. Acute, adaptational stress
5
responses exert temporal and beneficial effects to cope with the stressful situation, while
6
chronic activation of the stress system may adversely affect health with depression,
7
cardiovascular and auto-immune diseases, and psychosomatic complaints as potential
8
manifestations [2;3]. Furthermore, it has been hypothesized that prolonged stress may
9
negatively affect the body’s mineral status. Decreased zinc, selenium, iron and magnesium
10
concentrations have been observed in long-term psychological stress, although literature in
11
this regard is scarce [4-10].
12
The relationship between stress and minerals may operate through a behavioural and a
13
physiological pathway. On the behavioural side, stress has been shown to alter an individual’s
14
dietary pattern with an increase in the consumption of energy-dense “convenience foods”
15
(limiting purchasing/preparing time) or “comfort foods” (rich in sugar and fats), positively
16
stimulating sensations of reward and pleasure [11-13]. This unhealthy food consumption is
17
often at the expense of healthy mineral-rich foods and may lead to obesity and an unbalanced,
18
even deficient dietary mineral intake [14]. The physiological pathway operates through
19
activation of the stress system and the production of stress hormones such as cortisol, shifting
20
the body’s metabolism to a catabolic state, thereby increasing oxidative stress and increasing
21
the need for anti-oxidants (e.g. minerals for enzyme function) [15;16]. Moreover, stress has
22
been associated with common gastro-intestinal disorders and inhibition of nervus vagus
23
activation, perturbing gastric emptying, gastroduodenal/colonic motility and intestinal transit
24
[17-19]. As a consequence, nutrients may be less efficiently digested, absorbed and
25
metabolized. Stress hormones may also directly affect mineral absorption, distribution or
4
26
excretion (e.g. cortisol influences the parathyroid hormone and renal calcium handling,
27
together affecting calcium homeostasis) [20]. Summarized, stress may hinder an adequate
28
mineral intake, increase the body’s need for minerals through changes in metabolism or
29
redistribute the minerals to tissues with higher requirements, although more detailed
30
mechanisms need to be explored.
31
The last decade has been characterized by an increased interest in the study of stress and its
32
adverse health effects on young children [21-24]. In the field of stress and mineral status
33
however, no studies have been undertaken in children. From a methodological point of view,
34
scalp hair may offer opportunities compared to other biological matrices (such as serum or
35
urine) for the study of the stress-mineral relationship in children. Particularly for ‘longer-
36
term’ studies the hair matrix may be recommended as it is non-invasive and it presents a
37
retrospective window of stress hormones and mineral levels in the past [25;26]. Nevertheless,
38
this stress-mineral relationship has not been studied in scalp hair before. This study therefore
39
investigates the association between chronic stress and long-term mineral concentrations of
40
calcium (Ca), copper (Cu), iron (Fe), magnesium (Mg), phosphorus (P), zinc (Zn) and ratios
41
thereof in scalp hair of elementary-school girls by using child-reported stress estimates (life
42
events) and stress hormone measurements in scalp hair. We hypothesize that stress as
43
physiological event may be negatively associated with hair minerals levels in children,
44
independently from individual and behavioural factors such as the child’s age, BMI, physical
45
activity, dietary habits, socio-economic status and hair colour [27-33].
46
2. Methodology
47
Study Participants
48
140 girls aged 5-10 years old (Mean age=8.46 yr, SD= 1.11 yr) participated in this study as
49
part of the baseline survey of the ChiBS project (Children’s Body composition and Stress)
5
50
(February-June 2010, N=523). The ChiBS project, a study embedded within the European
51
IDEFICS study (Identification and prevention of Dietary- and lifestyle-induced health EFfects
52
In Children and infants) [34], investigates chronic psychosocial stress and changing body
53
composition in children over a two-year follow-up period (2010-2012), but also examines the
54
utility of hair samples as biomarker or diagnostic tool for stress and mineral status in children
55
[35]. More detailed research goals, methodology and participation characteristics of the
56
ChiBS project are described elsewhere [35].
57
In this study, the population was limited to the female participants of the ChiBS project, as
58
one of the survey modules, specifically hair sampling, was only performed in girls in order to
59
obtain the required hair length of 6 cm (N=263/523 or 50.3%, Figure 1). Parents were asked
60
to sign a consent form in which the option was offered to participate in the full ChiBS
61
programme or in a selected set of measurement modules, resulting in distinct participation
62
numbers for the different measurement modules (Figure 1). Hair samples were obtained from
63
the vertex posterior region of the scalp from 218 girls and hair colour noted. Only the most
64
proximal 6 cm of the hair strands were analyzed which is, based on an average hair growth
65
rate of 1 cm per month, representative for a period of 6 months in the past [36]. Each hair
66
sample obtained was split into two fractions and sent to expert-laboratories for determination
67
of cortisone and mineral levels, respectively (more details below). In addition, children were
68
subjected to routine anthropometric measurements and questionnaire administration (e.g.
69
stressful life events, dietary habits etc) at the same time point as hair sampling, at the
70
children’s schools. Data for all measurement modules was completed for 140 girls (Figure 1).
71
(insert Figure 1)
72
The ChiBS project was conducted according to the guidelines laid down in the Declaration of
73
Helsinki and was approved by the Ethics Committee of the Ghent University Hospital.
6
74
Hair mineral analysis
75
The hair contents of Ca, Cu, Fe, Mg, P and Zn were quantitatively determined via inductively
76
coupled plasma - mass spectrometry (ICP-MS), after microwave-assisted acid digestion of the
77
samples in the Department of Analytical Chemistry of Ghent University. Detailed procedures
78
and validation data of the ICP-MS method applied are described elsewhere [27]. The Ca/Mg,
79
Ca/P, Fe/Cu and Zn/Cu ratio were calculated, as they were suggested in literature to be
80
relevant mineral ratios [37-39].
81
Childhood stress measures
82
Coddington Life Events Scale (CLES)
83
To investigate the relationship between childhood psychosocial stress and hair mineral
84
concentrations, estimates of child-reported stress were obtained through a questionnaire on
85
stressors or life events. Completion of the questionnaires was assisted by trained interviewers.
86
The Coddington Life Events Scale for children (CLES-C) is a validated and well-established
87
36-item questionnaire, which measures the frequency and timing of both positive and negative
88
life events relevant for this age group during the last year (four trimesters). By measuring
89
significant life events in terms of Life Change Units (LCUs), the CLES-C can provide insight
90
into recent events that may be affecting the child’s health and results in a ‘life change units’
91
score [40;41]. Children with a score above the age-specific cut-off are considered to be at
92
higher risk to suffer from psychological problems. For this study, a ‘negative life events
93
score’ and the proportion of children at risk were calculated only for the last 6 months to
94
correspond with the 6 cm hair samples.
95
Hair Cortisone Analysis
7
96
Cortisone is a hormone that is only minimally produced by the adrenals but mainly origins
97
from cortisol metabolism, more specifically from the conversion of cortisol into cortisone by
98
11β-HSD2 (hydroxysteroid dehydrogenase). Cortisone may therefore serve as an additional
99
biomarker for stress research, as exemplified by the involvement of 11β-HSD activity in
100
stress [42-46]. Our research group has previously correlated elevated hair cortisone
101
concentrations with negative life events in the same study sample of elementary-school girls
102
[47]. Therefore, in this study, hair cortisone concentrations were measured as a biological
103
measure of stress. Cortisone was analysed in the most proximal 6 cm of the same hair samples
104
in which minerals were analysed at the Department of Toxicology, Institute of Legal
105
Medicine, Strasbourg University with Ultraperformance Liquid Chromatography-tandem
106
Mass Spectrometry (UPLC-MS/MS). Detailed procedures and validation data of the applied
107
UPLC-MS/MS method are described elsewhere [48].
108
Other variables
109
Within the ChiBS and the IDEFICS project, information was collected on parental education
110
(categorized using the International Standard Classification of Education (ISCED) [49]) and
111
children’s physical activity using a self-administered parentally reported questionnaire.
112
Physical activity was studied as the hours of playing outdoors and doing sports in a sports
113
club, which has been shown to correlate to moderate and vigorous physical activity as
114
measured by accelerometers [50;51]. Children’s dietary habits were assessed using the self-
115
administered parental questionnaire ‘Children’s Eating Habits Questionnaire - Food
116
Frequency Questions’ (CEHQ-FFQ). The CEHQ-FFQ is a 43 food item-containing
117
questionnaire developed and validated within the IDEFICS project [52;53] and is used as a
118
screening instrument to investigate dietary habits and food consumption frequency in
119
children. Based on these food consumption frequencies, a Youth Healthy Eating Index
120
(YHEI) [54] was calculated based on Feskanich et al. [55] with higher scores signalling
8
121
healthier diets. Also standardized routine anthropometric measurements (electronic scale
122
Tanita BC 420 SMA, Tokyo, Japan; stadiometer Seca 225, Birmingham, UK) were performed
123
and used to calculate Body Mass Index (BMI) z-scores (BMI=weight(kg)/height(m²)) [56].
124
Statistical analysis
125
Statistical analysis was performed using the Statistical Program PASW version 19.0 (SPSS
126
Inc, IBM, IL, USA). The two-sided level of significance was set at p<0.05. To study the
127
association between childhood stress and hair mineral concentrations, multiple linear
128
regression analysis was performed. The logarithmically transformed hair mineral
129
concentrations and mineral ratios were used as dependent (outcome) variables in 2 models
130
investigating another childhood stress measure: hair cortisone concentrations and the CLES
131
score were consecutively used as independent variables. After performing these analyses
132
without adjustment for confounder variables,
133
adjustment for age, BMI z-score, physical activity, YHEI, hair colour and parental education,
134
as these factors have been shown to be associated with hair mineral levels [27-33]. No
135
multicollinearity between these variables was shown. As the sample size for regression
136
analysis was limited to the children from whom complete information on all variables studied
137
was available (N=140 and N=107, figure 1), post-hoc power analysis was performed
138
according to Faul et al. using G*Power 3 [57;58]. A sample size of 140 and 107 children in a
139
regression model with 7 predictors to detect a medium effect size and with a probability level
140
of 0.05 resulted in a power of 0.92 and 0.80, respectively. Box-plots were created to
141
graphically present significant differences in hair mineral concentrations between children at
142
risk and not at risk for psychological problems by stressful life events in the last 6 months
143
(non-parametrical Mann-Whitney U-tests with non-transformed data mineral concentrations).
144
Also, this stress-variable was used as a dummy variable (at risk versus not at risk) in
145
regression analysis.
9
the regression analysis was repeated with
146
3. Results
147
Descriptive results of the participants’ socio-demographic characteristics, hair mineral
148
concentrations and stress measures are presented in Table 1.
149
(insert Table 1)
150
Results of regression analysis are described in Table 2. Hair cortisone concentrations were
151
inversely associated with hair mineral levels of Ca, Mg, Zn and Ca/P, both with and without
152
adjustment for the confounder variables. These findings indicate reduced levels of certain
153
minerals with increased childhood stress, independently from the child’s age, BMI, physical
154
activity, diet, hair colour and parental education.
155
(insert Table 2)
156
Children at risk for psychological problems by the occurrence of stressful events in the last 6
157
months demonstrated an elevated Ca/Mg ratio, as presented in Table 2 and illustrated in
158
Figure 2. The CLES negative events score (as continuous predictor variable) was only
159
associated with the Ca/Mg ratio without adjustment for the individual and behavioural factors
160
(Table 2).
161
(insert Figure 2)
162
Analyses were adjusted for a number of possible confounders (see methodology section). In
163
this study, the following variables were significantly associated with one or more of the
164
dependent variables: (1) age (hair Fe, Fe/Cu and Ca/Mg), (2) BMI Z-score (hair Ca/Mg), (3)
165
physical activity (hair Zn, Fe/Cu and Zn/Cu), (4) hair colour (hair P, Ca and Mg), and (5) the
166
YHEI (Ca/Mg) (Table 2).
167
4. Discussion
10
168
This paper was the first to investigate the stress-mineral relationship in children over a long-
169
term in the past, and using scalp hair as biological matrix. We demonstrated an independent
170
association of chronic stress measures (i.e. hair cortisone and CLES) on hair mineral levels of
171
elementary-school girls. As stress-mineral research in scalp hair remained unexplored until
172
now, evaluation of our findings is limited to previous observations in other biological
173
matrices such as serum or urine.
174
Higher levels of hair cortisone were associated with reduced hair levels of Ca, Mg, Zn and
175
Ca/P, which is in agreement with previous findings in serum and urine [4-6;9]. Since analyses
176
were adjusted for individual and behavioural factors (age, BMI, diet, physical activity, hair
177
colour and parental education), the changes in mineral levels may be ascribed to the unique
178
physiological contribution of increased stress.
179
Surprisingly however, a relationship between hair mineral concentrations and hair
180
cortisone was observed, while no association was found for the hair minerals with the CLES
181
negative event score (except for the unadjusted relationship with the Ca/Mg ratio). This is
182
unexpected as the CLES score and hair cortisone may be assumed to behave similarly in
183
relation to hair mineral levels based on the following information: (1) environmental stressor
184
exposure (i.e. assessed by CLES questionnaire) is generally linked to a physiological stress
185
response (i.e. assessed by hair cortisone measurements) [1;59], (2) our research group
186
previously showed a strong correlation between the CLES questionnaire and hair cortisone
187
concentrations [47] and (3) both stress assessment methods represent the same time period in
188
the past (last 6 months). This study may thus indicate that not stressor exposure in general but
189
more specifically the body’s physiological stress response is related to hair minerals or
190
mineral metabolism, such as a stress-induced increased excretion of metabolites (such as
191
cortisone and minerals) into hair. Another interpretation could be the more sensitive
192
representation of stress by hair cortisone measurements compared to the child-reported CLES
11
193
questionnaire, as a result of which the regression analysis with the CLES score may be
194
attenuated compared to the cortisone results. Nevertheless, if CLES life events were studied
195
categorically (‘at risk’ versus ‘not at risk’), a robust relationship was observed with the Ca/Mg
196
ratio, indicating that life events may be associated with minerals only in more extreme cases
197
of repeated stressor exposure, i.e. if the cut-off score for becoming at risk for psychological
198
problems is reached.
199
Despite the association between stress (particularly cortisone and CLES at risk) and hair
200
mineral levels observed in this study, levels of Cu, Fe and P in hair were not associated with
201
any of the studied stress estimates, nor with the ‘at risk’ status for psychological problems by
202
stressful events (Table 2, Figure 2). For Cu, this is in line with previous findings [4;5], while
203
for Fe our results are in disagreement with findings from Singh et al., Moore et al. and Chen
204
et al. (although the latter refers to animal research) [5;10;60]. We may hypothesize that the
205
metabolism or homeostasis of Cu, Fe and P is not, or only to a lesser extent, influenced by
206
stress compared to the other studied minerals.
207
Although it has been hypothesized that stress may affect the body’s mineral intake or mineral
208
requirements, specific metabolic pathways have largely remained undefined. Nonetheless,
209
some explanations were provided in literature.
210
Singh et al. partly contributed a decrease in plasma Zn concentrations (in response to
211
stress) to a decrease in Zn-binding proteins (such as albumin) and to a removal of Zn from the
212
circulation by other tissues: glucocorticoids may stimulate hepatic metallothionein synthesis
213
and thereby orchestrate the sequestration of Zn by the liver. No influence of altered urinary
214
Zn excretion was shown by these authors [5]. However, as indicated by Roy et al. [61], Zn
215
deficiency may activate the hypothalamus-pituitary-adrenal axis, causing glucocorticoid
216
production, suggesting that our observed cortisone-Zn relationship may also operate in the
217
reverse direction.
12
218
Reduced Fe levels after stress have been explained by an increase in ferritin
219
concentrations (an intracellular Fe storage protein), indicating a shift from circulating to
220
storage iron [5]. In this context, hair Fe could be considered an excretion or storage pathway.
221
On the other hand, animal studies indicated decreased iron absorption in relation to
222
psychological stress, possibly through changed expression of iron transporters [60].
223
Grases et al. mainly attributed decreases in Ca and Mg status in response to stress to
224
changes in renal excretion [9]: stress-related cortisol prevents tubular Ca reabsorption
225
mediated by aldosterone, thereby inducing increased urinary Ca. As cortisol inhibits
226
aldosterone activity in renal cells, increases in Mg excretion are also observed. Grases and
227
colleagues associated anxiety to catecholamine production which may also increase urinary
228
Mg excretion and thus lower Mg status in serum and maybe in hair.
229
Despite the explanations mentioned above, it is clear that further investigations into
230
the effects of psychosocial stress on mineral metabolism are needed. Irrespective of the
231
precise mechanism by which stress is associated with hair minerals (e.g. changes in
232
metabolism, changes in diet), this study has pointed to another health impact of stress
233
exposure, even in young children.
234
Strengths, limitations and future research
235
This study investigated the association between childhood stress and hair minerals, as this
236
association remained un-investigated. Next to the novelty of the study, other strong
237
methodological features are the use of the validated and state of the art ICP-MS and UPLC-
238
MS/MS technique to measure hair mineral concentrations and hair cortisone levels,
239
respectively, and the adjustment of all analyses for child’s age, BMI, physical activity, dietary
240
habits, socio-economic status and hair colour, whereby the unique physiological contribution
241
of stress on hair mineral levels could be studied. A next asset of this study is the assessment of
242
the environmental (i.e. CLES questionnaires) and biological (i.e. hair hormone measurements)
13
243
stress dimension, which permitted studying the individual associations of these stress-
244
estimates with hair minerals. However, the use of child-reported stress questionnaires may be
245
subject to recall- or reporting-bias. In addition, no corrections for multiple testing were
246
performed (e.g. Bonferroni) as we considered this too stringent for our analysis, although this
247
may have led to an increased likelihood of significant findings because of the high number of
248
regressions analyzed. Post-hoc power analysis indicated that our sample size to detect a
249
medium effect size in our regression analysis was sufficient to reach a power of 0.80,
250
although a larger sample size would have allowed studying smaller effect sizes. Another
251
limitation that should be considered is the exclusively female population under study within a
252
small age range, limiting the generalisability of our results. Findings need thus to be
253
confirmed in a more heterogeneous population sample (i.e. boys and girls, childhood to
254
adolescence) and in a longitudinal study design, since the cross-sectional design of this study
255
cannot determine causality. As not much is known in the field of stress and minerals, more
256
specifically hair minerals, evaluation and discussion of our observations to previously
257
reported findings remain limited. Therefore, we again point to the importance of further
258
research. Nevertheless, this study may initiate further hair mineral research in relation to
259
stress and stress-related health effects or behaviour in children. Particularly for large-scale
260
epidemiological research in children, hair mineral analysis may offer considerable advantages
261
as hair sampling is easy, non-invasive, inexpensive and the samples are easily stored. For
262
boys on the other hand, a sufficient length of hair should be available to obtain retrospective
263
measures of several months in the past. Last, we recommend further investigation into the
264
detailed
265
levels/accumulation (e.g. are some minerals more susceptible to stress than others?; is the
266
stress-mineral relationship stressor dependent? etc.) in order to further endorse our findings.
267
Conclusions
mechanisms
and
processes
by
14
which
stress
influences
hair
mineral
268
This study strengthened previous indications of a relationship between stress and mineral
269
levels. More specifically, we demonstrated an independent association between chronic stress
270
measures (i.e. hair cortisone and life events) and hair mineral levels in young girls, a
271
previously unexplored research area. However, findings need to be confirmed in a more
272
heterogeneous population and on a longitudinal basis. Furthermore, the precise mechanisms
273
by which stress alters hair mineral levels should be further elucidated in order to fully
274
understand the importance of stress on this aspect of health.
275
Acknowledgements
276
The project was financed by the European Community within the Sixth RTD Framework
277
Program Contract No. 016181 (FOOD) and the research council of Ghent University
278
(Bijzonder Onderzoeksfonds). Barbara Vanaelst, Lieve Balcaen and Maite Aramendia are
279
financially supported by the Research Foundation - Flanders (Grant n°: 1.1.894.11.N.00,
280
1.2.031.09.N.01, 1.2.031.09.N.01, respectively). Nathalie Michels is financially supported by
281
the research council of Ghent University (Bijzonder onderzoeksfonds). María R. Flórez is
282
financially supported by Gent University (project BOF 01SB0309) and the Spanish Ministry
283
of Economy and Competitiveness (project CTQ2009-08606). The authors wish to thank the
284
ChiBS children and their parents who generously volunteered and participated in this project.
285
15
Female participants
in the baseline ChiBS
survey
N=263
Hair sampling
N=218
Routine
anthropometry
Socio-demographic
information
Physical activity
information
Youth Healthy Eating
Index
N=218
N=208
N=201
N=146
CLES questionnaires
N=213
Hair cortisone
analysis
N=164
Hair mineral analysis
N=218
Total participants in
this study
N=140 *
N=107 **
Figure 1: Flowchart of study participants
CLES: Coddington Life Events Scale; Socio-demographic information: International Standard Classification of Education; *Total number of
participants if questionnaires (CLES questionnaire or emotion questionnaire) are used as stress measurement; **Total number of participants if
hair cortisone concentrations are used as stress measurement
16
Figure 2: Hair Ca/Mg concentrations in children at risk (N=24) and not at risk (N=116)
for psychological problems by events of last 6 months
The hair Ca/Mg ratio is higher in children at risk by events of last 6 months (Mann-Whitney
U-Test (non-logarithmically transformed data) p=0.016). The higher and lower ends of the
boxes represent the Q3 and Q1 of the hair mineral concentrations. The Q2 is indicated within
the boxes. The whiskers represent the concentration range, excluding outliers and extremes.
(Q=quartile)
17
Table 1: Personal characteristics of the participating elementary school girls
(N=140)
Median P25 P75
Age
8
8
9
BMI Z-score
-0.19 -0.87 0.66
Physical activity (hours/week)
14
10
19
Youth healthy eating index (score 0-80)
49
42
55
Hair mineral concentrations (µg/g)
Ca (N=139)
586
293 852
Cu
18
14
30
Fe
7
6
9
Mg (N=139)
28
18
46
P
141
128 155
Zn
222
200 249
Hair cortisone concentration (pg/mg) (N=107)
9
7
11
CLES negative event score (last 6 months)
29
0
53
N
%
CLES at risk by events last 6 months
24
17.1
Maximal parental education
ISCED level 2 and 3
44
31.5
ISCED level 4
24
17.1
ISCED level 5
72
51.4
Hair colour
blond
7
5
red
2
1.4
brown
102
72.9
dark brown
27
19.3
black
2
1.4
Values were rounded off to the measurement unit except for BMI Z-scores.
ISCED= International Standard Classification of Education, 2 ‘lower secondary education’, 3 ‘upper secondary
education’, 4 ‘post-secondary non-tertiary education’, 5 ‘first stage of tertiary education’
18
Table 2: see attachment
19
Reference List
[1] Cohen S, Kessler RC, Gordon LU (1997) Measuring stress: a guide for health and
social scientists. New York: Oxford University Press, Inc.
[2] Schneiderman N, Ironson G, Siegel SD (2005) Stress and health: Psychological,
behavioral, and biological determinants. Annu Rev Clin Psychol 1: 607-28.
[3] Mcewen BS (1998) Protective and damaging effects of stress mediators. N Engl J Med
338: 171-9.
[4] Pizent A, Jurasovic J, Pavlovic M, Telisman S (1999) Serum copper, zinc and
selenium levels with regard to psychological stress in men. J Trace Elem Med Biol 13:
34-9.
[5] Singh A, Smoak BL, Patterson KY, Lemay LG, Veillon C, Deuster PA (1991)
Biochemical Indexes of Selected Trace Minerals in Men - Effect of Stress. Am J Clin
Nutr 53: 126-31.
[6] Takase B, Akima T, Uehata A, Ohsuzu F, Kurita A (2004) Effect of chronic stress and
sleep deprivation on both flow-mediated dilation in the brachial artery and the
intracellular magnesium level in humans. Clin Cardiol 27: 223-7.
[7] Cernak I, Savic V, Kotur J, Prokic V, Kuljic B, Grbovic D, et al. (2000) Alterations in
magnesium and oxidative status during chronic emotional stress. Magnes Res 13: 2936.
[8] Seelig MS (1994) Consequences of Magnesium-Deficiency on the Enhancement of
Stress Reactions - Preventive and Therapeutic Implications - (A Review). J Am Coll
Nutr 13: 429-46.
20
[9] Grases G, Perez-Castello JA, Sanchis P, Casero A, Perello J, Isern B, et al. (2006)
Anxiety and stress among science students. Study of calcium and magnesium
alterations. Magnes Res 19: 102-6.
[10] Moore RJ, Friedl KE, Tulley RT, Askew EW (1993) Maintenance of Iron Status in
Healthy-Men During An Extended Period of Stress and Physical-Activity. Am J Clin
Nutr 58: 923-7.
[11] Dallman MF, Pecoraro NC, la Fleur SE (2005) Chronic stress and comfort foods: Selfmedication and abdominal obesity. Brain Behav Immun 19: 275-80.
[12] Torres SJ, Nowson CA (2007) Relationship between stress, eating behavior, and
obesity. Nutrition 23: 887-94.
[13] Rabinovitz S (2006) Stress and Food Craving. In: Yehuda S, Mostofsky DI, editors.
Nutrients, Stress and Medical Disorders.New Jersey: Human Press; pp 155-64.
[14] Kaidar-Person O, Person B, Szomstein S, Rosenthal RJ (2008) Nutritional deficiencies
in morbidly obese patients: A new form of malnutrition? Obes Surg 18: 1028-34.
[15] Costantini D, Marasco V, Moller AP (2011) A meta-analysis of glucocorticoids as
modulators of oxidative stress in vertebrates. J Comp Physiol B 181: 447-56.
[16] Vertuani S, Angusti A, Manfredini S (2004) The antioxidants and pro-antioxidants
network: An overview. Curr Pharm Des 10: 1677-94.
[17] Kiecolt-Glaser JK (2010) Stress, Food, and Inflammation: Psychoneuroimmunology
and Nutrition at the Cutting Edge. Psychosom Med 72: 365-9.
[18] Yin J, Levanon D, Chen JDZ (2004) Inhibitory effects of stress on postprandial gastric
myoelectrical activity and vagal tone in healthy subjects. Neurogastroenterol Motil 16:
737-44.
[19] Mayer EA (2000) The neurobiology of stress and gastrointestinal disease. Gut 47:
861-9.
21
[20] Heshmati HM, Riggs BL, Burritt MF, McAlister CA, Wollan P, Khosla S (1998)
Effects of the circadian variation in serum cortisol on markers of bone turnover and
calcium homeostasis in normal postmenopausal women. J Clin Endocrin Metab 83:
751-6.
[21] Teicher MH, Andersen SL, Polcari A, Anderson CM, Navalta CP, Kim DM (2003)
The neurobiological consequences of early stress and childhood maltreatment.
Neurosci Biobehav Rev 27: 33-44.
[22] Michels N, Sioen I, Huybrechts I, Bammann K, Vanaelst B, De Vriendt T, et al.
(2012) Negative life events, emotions and psychological difficulties as determinants of
salivary cortisol in Belgian primary school children. Psychoneuroendocrinology 37:
1506-15.
[23] Vanaelst B, De Vriendt T, Ahrens W, Bammann K, Hadjigeorgiou C, Konstabel K, et
al. (2012) Prevalence of psychosomatic and emotional symptoms in European schoolaged children and its relationship with childhood adversities: results from the
IDEFICS study. Eur Child Adolesc Psychiatry 21: 253-65.
[24] Washington TD (2009) Psychological stress and anxiety in middle to late childhood
and early adolescence: manifestations and management. J Pediatr Nurs 24: 302-13.
[25] Kempson IM, Lombi E (2011) Hair analysis as a biomonitor for toxicology, disease
and health status. Chem Soc Rev 40: 3915-40.
[26] Meyer JS, Novak MA (2012) Minireview: Hair cortisol: A novel biomarker of
hypothalamic-pituitary-adrenocortical activity. Endocrinology 153: 4120-7.
[27] Vanaelst B, Huybrechts I, Michels N, Vyncke K, Sioen I, De Vriendt T, et al. (2012)
Mineral concentrations in hair of Belgian elementary school girls: reference values
and relationship with food consumption frequencies. Biol Trace Elem Res 150: 56-67.
22
[28] Chojnacka K, Zielinska A, Michalak I, Gorecki H (2010) The effect of dietary habits
on mineral composition of human scalp hair. Environ Toxicol Pharmacol 30: 188-94.
[29] Chojnacka K, Gorecka H, Gorecki H (2006) The effect of age, sex, smoking habit and
hair color on the composition of hair. Environ Toxicol Pharmacol 22: 52-7.
[30] Wang CT, Chang WT, Jeng LH, Liu PE, Liu LY (2005) Concentrations of calcium,
copper, iron, magnesium, and zinc in young female hair with different body mass
indexes in Taiwan. Journal of Health Science 51: 70-4.
[31] Nielsen FH, Lukaski HC (2006) Update on the relationship between magnesium and
exercise. Magnes Res 19: 180-9.
[32] Huybrechts I, Lin Y, De Keyzer W, Sioen I, Mouratidou T, Moreno LA, et al. (2011)
Dietary sources and sociodemographic and economic factors affecting vitamin D and
calcium intakes in Flemish preschoolers. Eur J Clin Nutr 65: 1039-47.
[33] Montain SJ, Cheuvront SN, Lukaski HC (2007) Sweat mineral-element responses
during 7 h of exercise-heat stress. Int J of Sport Nutr Exerc Metab 17: 574-82.
[34] Ahrens W, Bammann K, Siani A, Buchecker K, De Henauw S, Iacoviello L, et al.
(2011) The IDEFICS cohort: design, characteristics and participation in the baseline
survey. Int J Obes 35: S3-S15.
[35] Michels N, Vanaelst B, Vyncke K, Sioen I, Huybrechts I, De Vriendt T, et al. (2012)
Children's Body Composition and Stress - the ChiBS study: aims, design, methods and
population characteristics. Arch Publ Health 70: 17.
[36] Harkey MR (1993) Anatomy and physiology of hair. Forensic Sci Int 63: 9-18.
[37] Wang CT, Chang WT, Zeng WF, Lin CH (2005) Concentrations of calcium, copper,
iron, magnesium, potassium, sodium and zinc in adult female hair with different body
mass indexes in Taiwan. Clin Chem Lab Med 43: 389-93.
23
[38] Bialkowska M, Hoser A, Szostak WB, Dybczynski R, Sterlinski S, Nowicka G, et al.
(1987) Hair zinc and copper concentration in survivors of myocardial infarction. Ann
Nutr Metab 31: 327-32.
[39] Park SB, Choi SW, Nam AY (2009) Hair Tissue Mineral Analysis and Metabolic
Syndrome. Biol Trace Elem Res 130: 218-28.
[40] Coddington RD (1972) Significance of Life Events As Etiologic Factors in Diseases
of Children .2. Study of Normal Population. J Psychosom Res 16: 205-13.
[41] Villalonga-Olives E, Valderas JM, Palacio-Vieira JA, Herdman M, Rajmil L, Alonso J
(2008) The adaptation into Spanish of the Coddington Life Events Scales (CLES).
Qual Life Res 17: 447-52.
[42] Welberg LAM, Thrivikraman KV, Plotsky PM (2005) Chronic maternal stress inhibits
the capacity to up-regulate placental 11 beta-hydroxysteroid dehydrogenase type 2
activity. J Endocrinol 186: R7-R12.
[43] Altuna ME, Lelli SM, de Viale LCSM, Damasco MC (2006) Effect of stress on
hepatic 11 beta-hydroxysteroid dehydrogenase activity and its influence on
carbohydrate metabolism. Can J Physiol Pharmacol 84: 977-84.
[44] Romer B, Lewicka S, Kopf D, Lederbogen F, Hamann B, Gilles M, et al. (2009)
Cortisol Metabolism in Depressed Patients and Healthy Controls. Neuroendocrinology
90: 301-6.
[45] Plenis A, Konieczna L, Oledzka I, Kowalski P, Baczek T (2011) Simultaneous
determination of urinary cortisol, cortisone and corticosterone in parachutists,
depressed patients and healthy controls in view of biomedical and pharmacokinetic
studies. Mol Biosyst 7: 1487-500.
24
[46] Yehuda R, Bierer LM, Andrew R, Schmeidler J, Seckl JR (2009) Enduring effects of
severe developmental adversity, including nutritional deprivation, on cortisol
metabolism in aging Holocaust survivors. J Psychiatr Res 43: 877-83.
[47] Vanaelst B, De Vriendt T, Huybrechts I, Michels N, Vyncke K, Sioen I, et al. (2012)
Cortisone in hair of elementary-school girls and its relationship with childhood stress.
Revisions under review.
[48] Vanaelst B, Rivet N, Ludes B, De Henauw S, Raul JS (2012) Measurement of cortisol
and cortisone in children's hair using ultra performance liquid chromatography and
tandem mass spectrometry. Revisions under review.
[49] UNESCO (1997) International Standard Classification of Education ISCED.
http://www.unesco.org/education/information/nfsunesco/doc/isced_1997.htm.
[50] Burdette HL, Whitaker RC, Daniels SR (2004) Parental report of outdoor playtime as
a measure of physical activity in preschool-aged children. Archives of Pediatrics &
Adolescent Medicine 158: 353-7.
[51] Verbestel V, De Henauw S, Maes L, Pitsiladis YP, Konstabel K, Bammann K, et al.
(2012) Validity of a parental-report questionnaire to assess 2-8 years old children's
physical activity and sedentary behaviour. Submitted.
[52] Lanfer A, Hebestreit A, Ahrens W, Krogh V, Sieri S, Lissner L, et al. (2011)
Reproducibility of food consumption frequencies derived from the Children's Eating
Habits Questionnaire used in the IDEFICS study. Int J Obes 35: S61-S68.
[53] Huybrechts I, Bornhorst C, Pala V, Moreno LA, Barba G, Lissner L, et al. (2011)
Evaluation of the Children's Eating Habits Questionnaire used in the IDEFICS study
by relating urinary calcium and potassium to milk consumption frequencies among
European children. Int J Obes 35: S69-S78.
25
[54] Gwozdz W, Reisch L, Idefics partners (2012). Enhancing healthy lifestyles: A crossgeographic analysis of factors influencing diets of European children. Submitted.
[55] Feskanich D, Rockett HRH, Colditz GA (2004) Modifying the Healthy Eating Index
to assess diet quality in children and adolescents. J Am Diet Assoc 104: 1375-83.
[56] Bammann K, Sioen I, Huybrechts I, Casajus JA, Vicente-Rodriguez G, Cuthill R, et
al. (2011) The IDEFICS validation study on field methods for assessing physical
activity and body composition in children: design and data collection. Int J Obes 35:
S79-S87.
[57] Faul F, Erdfelder E, Lang AG, Buchner A (2007) G*Power 3: A flexible statistical
power analysis program for the social, behavioral, and biomedical sciences. Behavior
Research Methods 39: 175-91.
[58] Faul F, Erdfelder E, Buchner A, Lang AG (2009) Statistical power analyses using
G*Power 3.1: Tests for correlation and regression analyses. Behavior Research
Methods 41: 1149-60.
[59] Vanaelst B, De Vriendt T, Huybrechts I, Rinaldi S, De Henauw S (2012)
Epidemiological approaches to measure childhood stress. Paediatr Perinat Epidemiol
26: 280-97.
[60] Chen JB, Shen H, Chen CJ, Wang WY, Yu SY, Zhao M, et al. (2009) The effect of
psychological stress on iron absorption in rats. Bmc Gastroenterology 9.
[61] Roy A, Evers SE, Avison WR, Campbell MK (2010) Higher zinc intake buffers the
impact of stress on depressive symptoms in pregnancy. Nutrition Research 30: 695704.
26