Content uploaded by Walid El Ansari
Author content
All content in this area was uploaded by Walid El Ansari on Aug 01, 2016
Content may be subject to copyright.
90
Cent Eur J Public Health 2014; 22 (2): 90–97
SUMMARY
Aims: The current study assessed, by university and sex, the association between nutritional behaviour (twelve independent variables), and
stress and depressive symptoms (dependent variables) in a sample from three UK countries.
Methods: A cross-sectional survey was undertaken among undergraduates enrolled across seven universities in England, Wales and Northern
Ireland (N
=
3,706). Self-administered questionnaires included a 12-item food frequency questionnaire, Cohen’s Perceived Stress Scale and
modified Beck Depression Inventory. Sex and university comparisons were undertaken. Univariable and multivariable regression analyses were
computed for each of the two outcomes – perceived stress and depressive symptoms.
Results: The frequencies of consuming of the various food groups differed by university and sex, as did depressive symptoms and perceived
stress. Multivariable regression analyses indicated that consuming ‘unhealthy’ foods (e.g. sweets, cookies, snacks, fast food) was significantly
positively associated with perceived stress (females only) and depressive symptoms (both males and females). Conversely, consuming ‘healthy’
foods (e.g. fresh fruits, salads, cooked vegetables) was significantly negatively associated with perceived stress and depressive symptoms scores
for both sexes. There was significant negative association between consuming fish/sea food and depressive symptoms among males only. For
males and for females, consuming lemonade/soft drinks, meat/sausage products, dairy/dairy products, and cereal/cereal products were not as-
sociated with either perceived stress or depressive symptoms.
Conclusions: The associations between consuming ‘unhealthy’ foods and higher depressive symptoms and perceived stress among male and
female students as well as the associations between consuming ‘healthy’ foods and lower depressive symptoms and perceived stress among
male and female students in three UK countries suggest that interventions to reduce depressive symptoms and stress among students could also
result in the consumption of healthier foods and/or vice versa.
Key words: self reported, food consumption, dietary habits, depression, stress, university students, college health
Address for correspondence: W. El Ansari, Faculty of Applied Sciences, University of Gloucestershire, Oxstalls Campus, Oxstalls Lane, Gloucester
GL2 9HW, United Kingdom. E-mail: walidansari@glos.ac.uk.
FOOD AND MENTAL HEALTH: RELATIONSHIP
BETWEEN FOOD AND PERCEIVED STRESS AND
DEPRESSIVE SYMPTOMS AMONG UNIVERSITY
STUDENTS IN THE UNITED KINGDOM
Walid El Ansari1, Hamed Adetunji2, Reza Oskrochi3
1Faculty of Applied Sciences, University of Gloucestershire, Gloucester, United Kingdom
2Faculty of Public Health and Health Informatics, Umm Al Qura University, Makkah Al Mukarrama, Kingdom of Saudi Arabia
3Faculty of Technology, Design and Environment, Oxford Brookes University, Oxford, United Kingdom
INTRODUCTION
There are concerns about the rates of depression, stress and
other psychiatric symptoms among university students, regard-
less of their academic disciplines (1–3). Entry to college, whilst
exciting, can be stressful and taxing for many young adults, and
depressive symptoms and stress are health problems among col-
lege students across the globe (4–7).
The sources of stress for university/college student are numer-
ous. These include: achieving academic success despite financial
constraints (8); adapting to changes in academic workloads, sup-
port networks and new environments; psychosocial changes in
one’s social and support systems (9); and being away from family
home, with increased responsibility, in a period that could begin
experimentation with drugs and other risk-taking behaviours (9).
In addition, some disciplines e.g. medical courses could be more
‘stressful’ (10). Likewise, features of university accommodation
such as residence hall status (e.g. conflict with a roommate) or
aspects related to faculty/staff (e.g. differences with tutors/lectur-
ers) can be important predictors of stress (11).
Many young adults experience psychiatric episodes and mental
illnesses during university time (12). The burdens facing univer-
sity students were positively associated with higher depression,
not only by mediation via perceived stress but also directly (2).
In turn, stress has direct and indirect adverse effects on health,
and one way stress may affect health is by influencing the foods
people select to eat (13). At the general population level, there
seems to be a collective effect of diet on mood (14), and regard-
less of the underlying reason of the mood disorder, the way we
eat affects the way we feel (15) and possibly vice versa. Many
examples illustrate such food-mood relationships.
Carbohydrates and Sugars: carbohydrate consumption seems
to alleviate depressive moods (16), and this has been viewed as
part of the link for developing obesity (17). For sugars, when
91
stressed, students with increased appetite selected significantly
more types of sweet foods (e.g. desserts, chocolate/candy bars,
candy, ice cream) (18). A diet high in refined carbohydrates and
sugars is common in depressive illness (19).
Healthy Eating Choices and Fat Content of Foods: a study of
stress and appetite and eating habits of students found that under
normal conditions, 80% of students made healthy eating choices,
however, only 33% ate healthy when stressed (18). Equally, stress
not only increased food consumption in certain individuals but
also shifted their food choices from lower fat to higher fat foods
(13). Likewise, research over six years of the diet and lifestyle
of volunteers free of depression found that participants with high
consumption of trans-fats (pastries and fast food) had up to 48%
increase in the risk of depression when compared to participants
who did not consume these fats (20).
Fish and Meat: omega-3 fatty acids (dietary consumption of
fish) might be linked to depression, with research reporting a
beneficial association between fish consumption and depression
(21). Conversely, others did not support such relations (22), and
the vegetarian diet seemed not to adversely affect mood (despite
low intake of omega-3 fatty acids) (23). In connection to meat, a
study showed lower odds of depression with high meat consump-
tion (22), while others found that depressive symptoms were
associated with less frequent meat consumption (24).
‘Whole Foods’, Fruits and Vegetables: the role of nutrition
in the management of depression reported that the production of
neurotransmitters requires nutrients (amino acids, minerals and
B vitamins) found in whole grains, eggs, yogurt, beans, green
leaf vegetables, and corn (19). Individuals with the highest intake
of ‘whole food’ were least likely to be depressed compared to
those with the lowest adherence to such diet (25), and diets rich
in processed food may increase the risk of depression (19). For
fruits and vegetables, depressive symptoms were associated with
less frequent consumption of fruits/vegetables (24). Dietary fi-
bers are associated with higher alertness and less perceived stress
(26), and nutrition for depressed patients should include fruit,
vegetables and wholegrains (27).
The literature also suggested that there are sex differences.
Research found that only in females, perceived stress was as-
sociated with more frequent consumption of sweets/fast foods
and less frequent consumption of fruits/vegetables (24). Indeed,
the effect of stress level on food choice seems different for men
and women (28).
We explored the links between nutritional habits and mental
health indicators at seven universities in the United Kingdom.
Apart from a few exceptions, while most studies focused on
data from one university across several countries or within one
country (29, 30), less studies examined the same question (as-
sociations between mental health indicators and nutrition) across
large samples of diverse students at different universities of one
nation. This is despite calls that research on nutrition and on the
correlation between depression and food consumption should
be conducted across diverse student populations (31). In addi-
tion, others (24) reported that single-country studies are rarely
directly comparable to one another due to the differences in food
consumption measures or mental health indicators. The current
study bridged these knowledge gaps and surveyed students at
seven universities in the United Kingdom, employing the same
food consumption measures and the same mental health indicators
across the universities to assess the associations between mental
health indicators and nutrition. If evidence suggests that particular
nutritional habits are associated with stress and/or depressive
symptoms, then programmes and interventions addressing mental
health may also be associated with the consumption of healthier
foods and/or vice versa.
The current survey assessed the associations between mental
health indicators and nutritional habits of a representative sample
of undergraduate students (N = 3,706) across 7 universities in the
UK (2007–2008). We explored the links between self-reported
perceived stress and depressive symptoms on the one hand, and
the food habits in terms of the usual consumption of 12 selected
food groups on the other. The three objectives were to:
• describe the food consumption behaviour and two mental
health indicators (perceived stress and depressive symptoms)
of students by university and by sex;
• assesstheassociationsbetweenfoodconsumptionbehaviour
(individual food groups separately) and two mental health
indicators by sex (univariable analysis);
• assesstheassociationsbetweenfoodconsumptionbehaviour
(all food groups together) and two mental health indicators by
sex (multivariable analysis).
MATERIALS AND METHODS
Sample and Data Collection
In agreement with other student health and well-being studies
across various countries (8, 29–35), no monetary or course credit
incentives were provided for participation. Ethical approval was
obtained at each participating university, and data were collected
simultaneously at 7 universities in Northern Ireland, Wales and
England. Each questionnaire included participant information
outlining the research objectives. Data were confidential and
protected, participation was voluntary and anonymous, and stu-
dents were informed that by completing the questionnaire, they
consent to participate in the study. For quality assurance, data
were computer entered at one site to minimise potential data
entry errors. Based on the number of returned questionnaires, the
responseratewas≈80%.
Health and Well-being Questionnaire
Assessment of Dietary Intake: Students self-reported their
nutritional habits in a food frequency questionnaire (12 indica-
tor variables) that measured their consumption of sweets, cakes/
cookies, snacks and fast/canned food, fresh fruits, raw and cooked
vegetables and salads, meat and fish, milk products, and cereals
(Table 1). In line with others, this instrument was developed to
include food groups that are critical for studies of dietary habits,
and the face and content validity of the tool were established by
grounding the questionnaire on wide literature review. The intro-
ductory question, “How often do you eat the following foods?”
queried students about the frequency of their usual consumption
of each food group separately (5-point scale: ‘several times a day’,
‘daily’, ‘several times a week’, ‘1–4 times a month’, and ‘never’).
No formal test of validity was undertaken, but the questionnaire
was very similar to other food frequency questionnaires that had
92
been validated (36, 37). In addition, the authors agreed on the
high face validity of the food frequency questionnaire that the
study employed.
Stress and Depressive Symptoms measures: Perceived stress
was assessed with Cohen’s Perceived Stress Scale (PSS-4 items)
that measured the extent to which participants felt life situations
to be stressful (38). The four items assessed how unpredict-
able, uncontrollable and overloaded respondents find their lives
(5-point Likert scale, 1 = ‘Never’, 5 = ‘Very Often’). We then
recoded the responses across the four items, in such a way that
for all answers, the lowest code 1 ‘Never’ = positive mental
state; whereas the highest code 5 = negative mental state. Each
student’s total Perceived Stress Score was obtained by summing
their responses to all 4 items. Depressive symptoms were assessed
by a Modification of the Beck Depression Inventory (MBDI) (39,
40), which is a single statement per symptom (six-point Likert
scale, 0 = ‘Never’, 5 = ‘Almost Always’) measuring its frequency
in the last four weeks. For each participant, MBDI score was the
sum of the participant’s responses to all 20 items. The scale had
excellent internal consistency, as in our sample of students in
Northern Ireland, Wales and England, Cronbach’s alpha of MBDI
was 0.88, 0.87. and 0.89, respectively (0.88 for the whole sample).
Data Analysis
For descriptions, we computed the means of the consumption
frequency of the various food groups as well as mental health
indicators by country and sex. Perceived Stress Score variable
(PSS) was generated by dividing by 4 the sum of the responses
to all 4 items of Cohen’s Perceived Stress Scale (5-point Likert
scale response format, 1 = ‘Never’, 5 = ‘Very Often’). In addition,
we computed a Modification of the Beck Depression Inventory
(MBDI) score by summing up the responses to all 20 items that
measure this mental health indicator (Table 1). For the univariable
analysis, we assessed associations between consumption of each
food group individually and both mental health indicators using
linear regression stratified by sex and adjusted for university. For
the multivariable analysis, we used multiple linear regressions
to assess associations between consumption of all food groups
together and both mental health indicators, stratified by sex and
adjusted for university. Hence, we were able to control for the
effects of the other food groups while assessing the associations
of any given food group.
RESULTS
Characteristics of the Sample
The current analysis employed data comprising 3,706 students
(765 males, 2,699 females, 242 missing sex; M age 24.9 years ±
SD 8.6 years) from 7 universities in three countries of the UK:
England – University of Gloucestershire (N = 970, 26.2% of the
sample, M age = 23.36 years), Bath Spa University (N = 485,
13.1%, M age = 22.23), Oxford Brookes University (N = 208,
5.6%, M age = 31.63), University of Chester (N = 993, 26.8%,
M age = 26.02), Plymouth University (N = 169, 4.6%, M age =
24.63);Wales−SwanseaUniversity(N=406,11.0%,Mage=
25); and the Republic of Northern Ireland – University of Ulster
(N = 475, 12.8%, M age = 25.19). The sample comprised first year
undergraduates (42.6%, N = 1,491), second year undergraduates
(31.3%, N = 1,095), third year undergraduates (18.7%, N = 655),
andasmallerproportionof≥4thyearstudents(7.5%,N=262).
Within each of the universities, the sex distribution of the
participating students varied by university, e.g. Chester (86.9%
females), Gloucester (56.4%), Ulster (91.8%), Swansea (92.2%),
Plymouth (63.9%), Oxford Brookes (89.2%), and Bath Spa
(77.4%), reflecting the nature of the enrolled student populations
at the Faculties/Schools where data was collected, being Schools
of Health and Social Care/Nursing/Sports and Exercise etc.).
Food Consumption Behaviour and Mental Health
Indicators by University and by Sex
Across the universities, consuming sweets was more common
in females, and conversely, consuming fast/canned foods was
more common in males (Table 1). Females generally consumed
cake/cookies more commonly than males (except at Ulster,
Swansea, Plymouth, and Bath Spa). Females also consumed fresh
fruits more commonly (except at Ulster), and also consumed
salad/raw vegetables and cereal/cereal products more commonly
than males. Conversely, consuming lemonade/soft drinks, meat/
sausage products, fish/seafood was more common in males as
well as eating dairy/dairy products which was more common in
males (except at Plymouth and Bath Spa). For the mental health
indicators, females generally exhibited higher perceived stress,
and had higher depressive symptoms scores (except at Glouces-
tershire and Ulster). There were differences in perceived stress
scores and depressive symptoms levels across the universities.
Associations between Food Consumption Behaviour
(Each Food Group Individually) and Mental Health
Indicators by Sex (Univariable Analysis)
The univariable analysis suggested two features (Table 2). First,
generally, significant associations (between more of the differ-
ent food groups and perceived stress and depressive symptoms)
were more evident for females (perceived stress – 8 significant
food groups for females vs. 2 food groups for males; depressive
symptoms – 7 significant food groups for females vs. 5 food groups
for males). Secondly, for females, the significant associations
between food groups and each of perceived stress and depressive
symptoms were nearly equal (8 associations for perceived stress,
7 associations for depressive symptoms). However, for males, the
significant associations between food groups and each of perceived
stress and depressive symptoms were not equal (2 associations
for perceived stress, 5 associations for depressive symptoms).
These findings suggested that: a) the consumption of the different
food groups were more broadly associated with both stress and
depressive symptoms in females than in males; and, b) in males,
associations of the different food groups seem relatively more
prevalent with depressive symptoms than with perceived stress.
In addition, for females, some food groups were negatively
associated with stress (fresh fruits, salad/raw vegetables, cooked
vegetables, cereal/cereal products), while other food groups
were positively associated with stress (sweets, snacks, fast food/
canned food, lemonade/soft drinks) (Table 2). This same pattern
of relationships with the specific food groups was also generally
93
Chester Gloucestershire Ulster Swansea Plymouth Oxford Brookes Bath Spa
Female Male Female Male Female Male Female Male Female Male Female Male Female Male
N=755 N=114 N=511 N=393 N=425 N=38 N=367 N=31 N=108 N=61 N=174 N=21 N=356 N=103
Dietary Intake – Consumption of Food Groups*
Sweets 3.00 2.87 2.91 2.86 3.26 3.24 3.02 2.97 2.87 2.77 2.94 2.86 3.00 2.92
Cake/cookies 2.41 2.37 2.47 2.42 2.62 2.70 2.44 2.48 2.35 2.49 2.55 2.40 2.42 2.53
Snacks 2.73 2.78 2.60 2.82 2.96 2.76 2.70 2.65 2.33 2.64 2.62 2.43 2.55 2.74
Fast food/
canned food 2.40 2.60 2.25 2.64 2.55 2.62 2.39 2.68 2.06 2.36 2.13 2.15 2.28 2.47
Fresh fruits 3.65 3.23 3.77 3.21 3.63 3.66 3.90 3.52 4.06 3.39 4.01 3.75 3.79 3.31
Salad/raw
vegetables 3.31 3.08 3.50 3.06 3.28 3.08 3.48 3.19 3.82 3.39 3.82 3.43 3.50 3.04
Cooked
vegetables 3.51 3.20 3.59 3.19 3.60 3.66 3.57 3.48 3.86 3.48 3.84 3.85 3.54 3.12
Lemonade/
soft drinks 2.75 2.94 2.64 2.82 2.92 3.14 2.70 2.87 2.21 2.38 2.47 2.50 2.55 2.62
Meat/sausage
products 3.09 3.24 3.10 3.72 3.21 3.55 3.14 3.39 3.01 3.61 3.03 3.29 3.00 3.22
Fish/sea food 2.41 2.47 2.49 2.62 2.35 2.40 2.46 2.71 2.47 2.49 2.62 2.75 2.37 2.47
Dairy/dairy
products 3.77 3.79 3.67 3.74 3.85 4.13 3.83 4.03 3.99 3.82 3.91 4.00 3.83 3.73
Cereal/cereal
products 3.55 3.34 3.55 3.44 3.69 3.64 3.70 3.58 4.04 3.43 3.94 3.80 3.66 3.41
Mental Health – Perceived Stress and Depressive Symptoms
PSS** 11.00 10.04 10.80 10.40 11.56 10.56 10.76 9.96 11.44 10.36 11.08 10.00 11.04 10.6
MBDI*** 24.87 23.00 16.47 20.70 23.45 26.08 23.45 18.93 29.21 21.08 25.17 19.28 27.41 25.34
Table 1. Food consumption and mental health indicators by university and sex
observed for depressive symptoms in females (except for cereal/
cereal products).
In contrast, for males, the relationships between food con-
sumption and mental health indicators were ‘narrower’. Only
salad/raw vegetables were negatively associated with stress, and
only fast food/canned food was positively associated with stress.
For depressive symptoms, males exhibited a wider portfolio of
associations, one that shared some similar features to that of the
females (fresh fruits, salad/raw vegetables, cooked vegetables
were negatively associated with depressive symptoms). However,
for males, cake/cookies were positively associated with depressive
symptoms; such relationship was not evident in females.
Associations between Food Consumption Behaviour
(All Food Groups Together) and Mental Health In-
dicators by Sex (Multivariable Analysis)
Table 3 shows the multivariable analysis between food
consumption frequency and perceived stress and depressive
symptoms, stratified by sex and controlled for all other variables
under investigation. As for sweets/cookies/snacks/fast food, more
frequent consumption of these foods was significantly associated
with higher perceived stress (females only) and higher depressive
symptoms for both males and females. Less frequent fruits/veg-
etables consumption (e.g. fresh fruits, salads, cooked vegetables)
*Mean of the consumption frequency scale (1 = never, 5 = several times/ day); **Perceived Stress Scale by Cohen, higher scores indicate higher perceived stress; ***Modi-
ed Beck Depression Inventory, higher scores indicate stronger depressive symptoms; all cell values represent mean scores except MBDI where cell values are the sum
of responses.
was significantly associated with higher perceived stress and
higher depressive symptoms for both sexes. In addition, there
was a significant negative association between consuming fish/
sea food and depressive symptoms among males only. For males
and for females, consuming lemonade/soft drinks, meat/sausage
products, dairy/dairy products, and cereal/cereal products was not
associated with either perceived stress or depressive symptoms.
DISCUSSION
Nutrition is a modifiable determinant of non-communicable
diseases, and from the modifiable risk factors perspective, the role
of nutrition in psychiatry is now more significant than initially
considered (19). For instance, alternative non-pharmacological
treatments for depressive symptoms (e.g. nutritional supplements)
are significant treatment options (41). However, conflicting find-
ings have increased people’s interest among the different domains
of dietary composition and mood (31). Hence, there have been
calls to understand the relationships between depression/depressive
symptoms and other health behaviours such as eating and nutrition.
Our current findings contribute to the evidence base of the rela-
tionships between food consumption behaviour and mental health
indicators, expanding our awareness of these relationships and
their associations with sex across these young adult populations.
94
Perceived Stress Score (PSS) Depressive Symptoms Score (M-BDI)
Female Male Female Male
Food group p-value Estimate* p-value Estimate* p-value Estimate* p-value Estimate*
Sweets (chocolate, candy, etc.) 0.001 0.053 0.288 -0.029 < 0.001 0.081 0.071 0.066
Cake/cookies 0.155 0.024 0.845 -0.006 0.168 0.027 0.007 0.099
Snacks (chips, peanuts, etc.) 0.024 0.038 0.819 -0.007 0.001 0.068 0.058 0.069
Fast food/canned food** < 0.001 0.095 0.008 0.084 < 0.001 0.106 < 0.001 0.183
Fresh fruits < 0.001 -0.085 0.168 -0.036 < 0.001 -0.111 0.047 -0.074
Salad/raw vegetables < 0.001 -0.048 0.013 -0.065 < 0.001 -0.071 0.014 -0.091
Cooked vegetables < 0.001 -0.061 0.127 -0.042 < 0.001 -0.072 0.017 -0.089
Lemonade/soft drinks 0.003 0.038 0.840 -0.005 0.002 0.060 0.573 0.021
Meat/sausage products 0.665 0.006 0.748 -0.010 0.680 0.008 0.942 -0.003
Fish/sea food 0.138 -0.023 0.500 0.019 0.824 0.004 0.133 0.055
Dairy/dairy products 0.249 -0.017 0.840 0.006 0.050 -0.038 0.517 0.024
Cereal/cereal products*** 0.001 -0.049 0.518 -0.017 0.073 -0.035 0.740 -0.012
Table 2. Associations between food consumption behaviour and perceived stress and depressive symptoms (univariable analysis)
Each food group adjusted only for university; separate models for males and females and for both mental health indicators; estimates are the Standardized Coefcients;
*change in the corresponding score (PSS or M-BDI) per one unit of the food group frequency scale; **e.g. pizza, hamburger, french fries, canned ravioli, etc.; ***e.g. whole-
wheat bread, cereals, oatmeal, etc.; bolded cells indicate signicant relationships.
Perceived Stress Score (PSS) Depressive Symptoms Score (M-BDI)
Female Male Female Male
Food group or subscale p-value Estimate#p-value Estimate#p-value Estimate#p-value Estimate#
Sweets/cookies/snacks/fast food* 0.017 0.051 0.853 0.008 0.001 0.072 < 0.001 0.158
Fruits/vegetables** 0.002 -0.067 0.025 -0.092 < 0.001 -0.081 0.004 -0.115
Lemonade/soft drinks 0.207 0.027 0.949 -0.003 0.128 0.032 0.270 -0.047
Meat/sausage products 0.950 0.001 0.398 -0.035 0.788 -0.006 0.112 -0.065
Fish/sea food 0.894 -0.003 0.156 0.057 0.196 0.027 0.006 0.108
Dairy/dairy products 0.873 -0.003 0.582 0.023 0.095 -0.036 0.339 0.039
Cereal/cereal products*** 0.059 -0.042 0.869 -0.007 0.976 -0.001 0.872 0.007
Table 3. Associations between food consumption behaviour and perceived stress and depressive symptoms (multivariable
analysis)
Each food group adjusted for university and for all other variables in the table, separate models for males and females and for both mental health indicators; estimates
are the Standardized Coefcients; #change in the corresponding score (PSS or MBDI) per one unit of the food group frequency scale; *Sweets/cookies/snacks/fast food
subscale: mean of four items (sweets, cakes/cookies, snacks, fast food); **Fruits/vegetables subscale: mean of three items (fresh fruits, salads, cooked vegetables);
***e.g. whole-wheat bread, cereals, oatmeal, etc.; bolded cells indicate signicant relationships.
For the current study, we found that across all the participating
universities, the consumption of sweets was more common in
females. This is in agreement with research of university students
in Germany, Denmark, Poland, and Bulgaria (34) that used the
same (food frequency and mental health) research instruments
as employed in the present study, and reported that overall, more
women reported frequent consumption of sweets. In our sample
of students, females also consumed fresh fruits more commonly
(except at Ulster), and generally ate salad/raw vegetables products
more commonly than males. Earlier reports similarly found that
female students ate more fruits than males (42), and our findings
were again consistent with a European study where largely, more
women reported frequent consumption of fruits and salads (34).
We are also in agreement with the study conducted in Hong Kong
(43), where females had better nutritional habits than men (were
more likely to report eating fruit/vegetables), postulating support
that women behaved ‘healthier’ than men in terms of consuming
fruit (44). Conversely, we found that consumption of meat/sausage
products and fish/seafood were more common in males, in support
of research (34) where more men reported frequent consumption
of meat and fish. Indeed, El Ansari and colleagues (34) found
that irrespective of the four countries they examined, more men
regularly ate meat than women, which is also in agreement with
Turkish males adolescents who reported significantly more meat
serving per day compared to females (45).
Regarding the differences between the participating universi-
ties in terms of their students’ nutritional behaviours and mental
health levels, it is not straightforward to tease out the factors
that could explain such differences. Such factors might include
student-level variables and/or university-level features (32).
95
Similar difficulties have been noted in the relationships between
student health outcomes and well-being and student- and school-
level factors (46). Understanding the different dimensions of
the university characteristics warrants further consideration,
particularly as these dimensions relate to determinants of student
health/well-being. Few multi-level studies collected sufficient
student- and university-level information to be able to further our
understandings of such relationships. Others (46) also noted the
scarcity of multi-level studies that explored the manner in which
schools may affect student health/well-being. Likewise, El Ansari
et al. (32, 35) suggested that differences between participating
universities in terms of health practices could be related to many
features, e.g., university characteristics and its environment, poli-
cies, student selection, and resultant composition of the student
population; or related to the region where a university is located;
or the country and its political and health stances. Many con-
founding factors that could confound such complex associations
are usually not measured.
In connection with objectives two and three, we assessed the
associations between food consumption behaviour and two mental
health indicators by sex twice: first employing each food group
individually (univariable analysis); and then by employing all the
food groups together (multivariable analysis). Our initial univari-
able analysis shed light on the relationships between each food
group (individually) and two mental health indicators. However,
as it was equally critical to understand the relationships between
the various food groups (collectively) and the mental health indi-
cators, hence we subsequently undertook multivariable analysis
to control for all the other variables under investigation. This is in
agreement with others who have voiced that it is more important
to study the overall dietary pattern than isolated nutrients (47).
As for sweets/cookies/snacks/fast food, for our UK sample,
more frequent consumption of these food groups was significantly
positively associated with higher perceived stress in females only.
This is in agreement with research across Germany, Poland and
Bulgaria that employed the same instruments as the current study
and reported the same findings: in females only, perceived stress
was associated with more frequent consumption of sweets/cook-
ies/snacks/fast food (24). However, for depressive symptoms,
whilst our findings indicated that more frequent consumption
of sweets/cookies/snacks/fast food were positively significantly
associated with higher depressive symptoms for both males and
females, others (24) reported this relationship only for females.
This discrepancy in findings might be due to two points. First,
our much larger sample (3,706 students) than that of the European
study (1,839 students) possibly rendered enough power for the
current study to pick up findings that other research with rela-
tively smaller sample sizes did not (24). Second, the discrepancy
in findings might also be due to the extent of ‘diversity’ of the
sample, as previous work (31) suggested that research on nutrition
and on the correlation between depression and food consumption
should be conducted across diverse student populations. In the
European study (24), the diversity was between Germany, Poland
and Bulgaria; in our study, it was England, Wales and Northern
Ireland. Because our sample had more universities from England
that the other two nations, it is possible that the extent of diversity
is different than that of the European study. Nevertheless, we are
also in partial agreement with other research that examined the as-
sociation of food consumption behaviour with one or both mental
health indicators among college students across seven cities (31),
and found an increase of perceived stress with the increase of the
intake frequency level of ready-to-eat food and snack food for the
whole sample. However, Liu et al. (31) did not explore relation-
ships by sex, therefore, the overall associations they observed
might have resulted from associations limited to females only.
In connection to fresh fruits, salads and cooked vegetables, the
current study found that less frequent consumption of these food
groups was significantly associated with higher perceived stress
and higher depressive symptoms scores for both sexes. Again
this finding is in partial agreement with others (24) where higher
fruits/vegetables consumption was associated with lower levels of
depressive symptoms among females. We are also in agreement
with research (31) that found an increase of depression with the
decrease of the frequency level of eating fruits, although they
did not stratify their findings by sex, so it is not absolutely clear
whether the overall associations they found was true for both
males and females or could have again resulted from associa-
tions limited to females or males only. Others have also reported
that psychological stress has been associated with less fruit and
vegetable intake (48).
As for fish/sea food, the current study found a significant
positive association between the consumption of fish/sea food
and depressive symptoms among males only. This is congruent
with others (46) who have reported that fish consumption (rich
in omega-3 fatty acids) might be causally associated with mood
stabilization. We are also in partial agreement with Hakkarainen
et al. (22), where dietary intake of omega-3 fatty acids showed
no association with low mood level (we found no relationship
between stress and fish/sea food for both sexes; and no relation-
ship between fish/sea food and depressive symptoms in females).
For both sexes, we found that the consumption of meat/sausage
products, dairy/dairy products, and cereal/cereal products were not
associated with either perceived stress or depressive symptoms.
This is in partial agreement with others who researched university
students (24) and found that for both sexes, milk products and
cereal products were not associated with either perceived stress
or depressive symptoms.
In understanding the food consumption behaviour-mental
health associations, generally, the links between depressive symp-
toms and health behaviours consider a two-way feature: depres-
sive symptoms can be associated with increased ‘health-compro-
mising’ behaviours that place individuals at risk for other health
consequences beyond the psychological domain; and, depressive
symptoms can be associated with decreased ‘health-promoting’
behaviours and may prevent full-health potential across multiple
domains of health (49). For instance, in adolescents, depression/
negative mood was related to health-compromising attitudes/
behaviours, e.g., weight dissatisfaction, negative body image,
while a significant negative relationship between depressed mood
and health-promoting eating behaviours, e.g. eating breakfast and
lunch has also been reported (50–52).
This study has limitations and generalization needs to be cau-
tious. Across the 7 universities, data were collected employing
the same method and research tool, and data processing was the
same across universities. However, as there could be regional
variations and/or systematic differences between the selected
universities, students’ data from each given university may not
be representative of the whole university. Nevertheless, a high
96
response rate at each site suggested that selection bias was not
a probable danger to the study’s internal validity. The study is
a cross-sectional survey, hence the direction of the association
between food consumption and mental health cannot be ascer-
tained. We did not have enough information on social class and
entry tariff across the university courses that participants were
studying on, which could have been useful. Participants self-
reported the levels of mental health indicators and dietary food
consumption where such responses may be subject to sociability
and social desirability. We did not assess serving sizes, which
might be different across sites. The food frequency questionnaire
was not compared against objective methods of food consump-
tion measurement. Nonetheless, the tool was comparable to other
published food frequency questionnaires that have been validated
(36, 37). Future research would need to consider such limitations.
In addition, studies are needed to clarify the temporal relation-
ship between the emotional/mental and nutritional domains (53).
Significant associations between food and mood warrant further
research to determine causality.
CONCLUSION
In summary, the frequencies of consumption of the various
food groups as well as depressive symptoms and perceived stress
differed by university and sex. Consuming ‘unhealthy’ foods (e.g.
sweets/cookies/snacks/fast food) was significantly positively
associated with perceived stress (females only) and depressive
symptoms for both males and for females. Conversely, consuming
‘healthy’ foods (e.g. fresh fruits, salads, cooked vegetables) was
significantly negatively associated with perceived stress and de-
pressive symptoms scores for both sexes. There was a significant
positive association between the consumption of fish/sea food
and depressive symptoms among males only. For both males and
females, the consumption of lemonade/soft drinks, meat/sausage
products, dairy/dairy products, and cereal/cereal products were
not associated with either perceived stress or depressive symp-
toms. The study findings should support the implementation of
health promotion and prevention programmes at universities.
Positive effects on food consumption behaviours can be expected
from interventions aimed at decreasing or preventing depressive
symptoms and/or reducing perceived stress levels among students,
and vice versa. Efforts to reduce depressive symptoms and stress
among female students may lead to the consumption of healthier
foods and/or vice versa. Strategies might include stress manage-
ment programmes, university environment enabling relaxation
and wellbeing, health-oriented courses, and the organization of
studies and curricula with a focus on stress reduction.
Acknowledgements
The authors wish to acknowledge the universities, faculties and students
who participated in this study. In addition, we acknowledge the UK
Student Health Group (J. John, P. Deeny, C. Phillips, S. Snelgrove, X.
Hu, S. Parke, M. Stoate, A. Mabhala).
Conflict of Interest
None declared
REFERENCES
1. Levine RE, Litwins SD, Frye AW. An evaluation of depressed mood in two
classes of medical students. Acad Psychiatry. 2006 May-Jun;30(3):235-7.
2. Mikolajczyk RT, Maxwell AE, Naydenova V, Meier S, El Ansari W.
Depressive symptoms and perceived burdens related to being a student:
Survey in three European countries. Clin Pract Epidemiol Ment Health.
2008 Jul 3;4:19.
3. Mikolajczyk RT, Maxwell AE, El Ansari W, Naydenova V, Stock C,
Ilieva S, et al. Prevalence of depressive symptoms in university students
from Germany, Denmark, Poland and Bulgaria. Soc Psychiatry Psychiatr
Epidemiol. 2008 Feb;43(2):105-12.
4. Eller T, Aluoja A, Vasar V, Veldi M. Symptoms of anxiety and depres-
sion in Estonian medical students with sleep problems. Depress Anxiety.
2006;23(4):250-6.
5. El Ansari W, Stock C, Phillips C, Mabhala A, Stoate M, Adetunji H, et al.
Does the association between depressive symptomatology and physical
activity depend on body image perception? A survey of students from
seven universities in the UK. Int J Environ Res Public Health. 2011
Feb;8(2):281-99.
6. Gambetta-Tessini K, Mariño R, Morgan M, Evans W, Anderson V. Stress
and health-promoting attributes in Australian, New Zealand, and Chilean
dental students. J Dent Educ. 2013 Jun;77(6):801-9.
7. El Ansari W, Labeeb S, Moseley L, Kotb S, El-Houfy A. Physical and
psychological well-being of university students: survey of eleven faculties
in Egypt. Int J Prev Med. 2013 Mar;4(3):293-310.
8. El Ansari W, Stock C. Is the health and wellbeing of university students
associated with their academic performance? Cross sectional findings
from the United Kingdom. Int J Environ Res Public Health. 2010
Feb;7(2):509-27.
9. Piko B. Perceived social support from parents and peers: which is the
stronger predictor of adolescent substance use? Subst Use Misuse. 2000
Mar;35(4):617-30.
10. Chan GC, Koh D. Understanding the psychosocial and physical work
environment in a Singapore medical school. Singapore Med J. 2007
Feb;48(2):166-71.
11. Dusselier L, Dunn B, Wang Y, Shelley MC 2nd, Whalen DF. Personal,
health, academic, and environmental predictors of stress for residence
hall students. J Am Coll Health. 2005 Jul-Aug;54(1):15-24.
12. Mowbray CT, Megivern D, Mandiberg JM, Strauss S, Stein CH, Collins
K, et al. Campus mental health services: recommendations for change.
Am J Orthopsychiatry. 2006 Apr;76(2):226-37.
13. Zellner DA, Loaiza S, Gonzalez Z, Pita J, Morales J, Pecora D, et al. Food
selection changes under stress. Physiol Behav. 2006 Apr 15;87(4):789-93.
14. Meyer BJ, Kolanu N, Griffiths DA, Grounds B, Howe PR, Kreis IA.
Food groups and fatty acids associated with self-reported depression:
an analysis from the Australian National Nutrition and Health Surveys.
Nutrition. 2013 Jul-Aug;29(7-8):1042-7.
15. Low Dog T. The role of nutrition in mental health. Altern Ther Health
Med. 2010 Mar-Apr;16(2):42-6.
16. Wurtman RJ, Wurtman JJ. Carbohydrates and depression. Sci Am. 1989
Jan;260(1):68-75.
17. Arnow B, Kenardy J, Agras WS. Binge eating among the obese: a descrip-
tive study. J Behav Med. 1992 Apr;15(2):155-70.
18. Kandiah J, Yake M, Jones J, Meyer M. Stress influences appetite and
comfort food preferences in college women. Nutr Res. 2006;26(3):118-23.
19. Popa TA, Ladea M. Nutrition and depression at the forefront of progress.
J Med Life. 2012 Dec 15;5(4):414-9.
20. Sánchez-Villegas A, Toledo E, de Irala J, Ruiz-Canela M, Pla-Vidal
J, Martínez-González MA. Fast-food and commercial baked goods
consumption and the risk of depression. Public Health Nutr. 2012
Mar;15(3):424-32.
21. Appleton KM, Woodside JV, Yarnell JW, Arveiler D, Haas B, Amouyel
P, et al. Depressed mood and dietary fish intake: direct relationship or
indirect relationship as a result of diet and lifestyle? J Affect Disord. 2007
Dec;104(1-3):217-23.
22. Hakkarainen R, Partonen T, Haukka J, Virtamo J, Albanes D, Lönnqvist J.
Is low dietary intake of omega-3 fatty acids associated with depression?
Am J Psychiatry. 2004 Mar;161(3):567-9.
23. Beezhold BL, Johnston CS, Daigle DR. Vegetarian diets are associated
with healthy mood states: a cross-sectional study in seventh day adventist
adults. Nutr J. 2010 Jun 1;9:26.
24. Mikolajczyk RT, El Ansari W, Maxwell AE. Food consumption frequency
and perceived stress and depressive symptoms among students in three
European countries. Nutr J. 2009 Jul 15;8:31.
97
25. Akbaraly TN, Brunner EJ, Ferrie JE, Marmot MG, Kivimaki M, Singh-
Manoux A. Dietary pattern and depressive symptoms in middle age. Br
J Psychiatry. 2009 Nov;195(5):408-13.
26. Bourre JM. Effects of nutrients (in food) on the structure and function
of the nervous system: update on dietary requirements for brain. Part 2:
macronutrients. J Nutr Health Aging. 2006 Sep-Oct;10(5):386-99.
27. Forsyth AK, Williams PG, Deane FP. Nutrition status of primary care
patients with depression and anxiety. Aust J Prim Health. 2012;18(2):172-6.
28. Zellner DA, Saito S, Gonzalez J. The effect of stress on men's food selec-
tion. Appetite. 2007 Nov;49(3):696-9.
29. El Ansari W, Maxwell AE, Mikolajczyk RT, Stock C, Naydenova V,
Krämer A. Promoting public health: benefits and challenges of a Europe-
anwide research consortium on student health. Cent Eur J Public Health.
2007 Jun;15(2):58-65.
30. El Ansari W, Labeeb S, Kotb S, Yousafzai MT, El-Houfey A, Stock C.
Correlates of smoking, quit attempts and attitudes towards total smoking
bans at university: findings from eleven faculties in Egypt. Asian Pac J
Cancer Prev. 2012;13(6):2547-56.
31. Liu C, Xie B, Chou CP, Koprowski C, Zhou D, Palmer P, et al. Perceived
stress, depression and food consumption frequency in the college students
of China Seven Cities. Physiol Behav. 2007 Nov 23;92(4):748-54.
32. El Ansari W, Stock C; UK Student Health Group, Snelgrove S, Hu X,
Parke S, Davies S, et al. Feeling healthy? A survey of physical and psy-
chological wellbeing of students from seven universities in the UK. Int
J Environ Res Public Health. 2011 May;8(5):1308-23.
33. El Ansari W, Stock C. Factors associated with smoking, quit attempts
and attitudes towards total smoking bans at university: a survey of seven
universities in England, Wales and Northern Ireland. Asian Pac J Cancer
Prev. 2012;13(2):705-14.
34. El Ansari W, Stock C, Mikolajczyk RT. Relationships between food
consumption and living arrangements among university students in four
European countries - a cross-sectional study. Nutr J. 2012 Apr 24;11:28.
35. El Ansari W, Clausen SV, Mabhala A, Stock C. How do I look? Body
image perceptions among university students from England and Denmark.
Int J Environ Res Public Health. 2010 Feb;7(2):583-95.
36. Osler M, Heitmann BL. The validity of a short food frequency question-
naire and its ability to measure changes in food intake: a longitudinal
study. Int J Epidemiol. 1996 Oct;25(5):1023-9.
37. Roddam AW, Spencer E, Banks E, Beral V, Reeves G, Appleby P, et al.
Reproducibility of a short semi-quantitative food group questionnaire and
its performance in estimating nutrient intake compared with a 7-day diet
diary in the Million Women Study. Public Health Nutr. 2005 Apr;8(2):201-
13.
38. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived
stress. J Health Soc Behav. 1983 Dec;24(4):385-96.
39. Schmitt M, Beckmann M, Dusi D, Maes J, Schiller A, Schonauer K. Reli-
ability of the simplified Beck Depression Inventory (BDI-V). Diagnostica.
2003;49(4):147-56. (In German.)
40. Schmitt M, Maes J. Suggested simplification of the Beck Depression
Inventory (BDI). Diagnostica. 2000;46(1):38-46. (In German.)
41. Rechenberg K, Humphries D. Nutritional interventions in depression and
perinatal depression. Yale J Biol Med. 2013 Jun 13;86(2):127-37.
42. Schw eyer FX, Le Corre N. Daily nutrition in students. Pre’venir.
1994;26:87-92. (In French.)
43. Lee RL, Loke AJ. Health-promoting behaviors and psychosocial well-
being of university students in Hong Kong. Public Health Nurs. 2005
May-Jun;22(3):209-20.
44. Monneuse MO, Bellisle F, Koppert G. Eating habits, food and health
related attitudes and beliefs reported by French students. Eur J Clin Nutr.
1997 Jan;51(1):46-53.
45. BaşM,AltanT,DinçerD,AranE,KayaHG,YüksekO.Determination
of dietary habits as a risk factor of cardiovascular heart disease in Turkish
adolescents. Eur J Nutr. 2005 Mar;44(3):174-82.
46. Saab H, Klinger D. School differences in adolescent health and wellbeing:
findings from the Canadian Health Behaviour in School-aged Children
Study. Soc Sci Med. 2010 Mar;70(6):850-8.
47. Sanchez-Villegas A, Martínez-González MA. Diet, a new target to prevent
depression? BMC Med. 2013 Jan 3;11:3.
48. Cartwright M, Wardle J, Steggles N, Simon AE, Croker H, Jarvis
MJ. Stress and dietary practices in adolescents. Health Psychol. 2003
Jul;22(4):362-9.
49. Perry CL, Jessor R. The concept of health promotion and the prevention
of adolescent drug abuse. Health Educ Q. 1985 Summer;12(2):169-84.
50. Tomori M, Rus-Makovec M. Eating behavior, depression, and self-esteem
in high school students. J Adolesc Health. 2000 May;26(5):361-7.
51. Stice E, Presnell K, Spangler D. Risk factors for binge eating onset in
adolescent girls: a 2-year prospective investigation. Health Psychol. 2002
Mar;21(2):131-8.
52. Kaplan SL, Landa B, Weinhold C, Shenker IR. Adverse health behaviors
and depressive symptomatology in adolescents. J Am Acad Child Psy-
chiatry. 1984 Sep;23(5):595-601.
53. German L, Kahana C, Rosenfeld V, Zabrowsky I, Wiezer Z, Fraser D,
et al. Depressive symptoms are associated with food insufficiency and
nutritional deficiencies in poor community-dwelling elderly people. J
Nutr Health Aging. 2011 Jan;15(1):3-8.
Received July 18, 2013
Accepted in revised form April 24, 2014