Acta Pdiatrica ISSN 0803-5253

REGULAR ARTICLE

Coeliac disease is more common in children with high socio-economic status

LA Whyte1, S Kotecha2, WJ Watkins2, HR Jenkins (huw.jenkins2@wales.nhs.uk)1
1.Department of Paediatric Gastroenterology, Childrens Hospital of Wales, Cardiff, UK
2.Department of Child Health, Cardiff University, School of Medicine, Cardiff, UK

Keywords
Deprivation, Epidemiology, Gastroenterology,
General paediatrics, Socio-economic status
Correspondence
Dr. Huw R Jenkins, Consultant Paediatric
Gastroenterologist, Department of Child Health,
Childrens Hospital of Wales, Heath Park, Cardiff
CF14 4XW, UK.
Tel: +44 29 2074 4558 |
Fax: +44 29 2074 4559 |
Email: huw.jenkins2@wales.nhs.uk
Received
3 August 2013; revised 21 September 2013;
accepted 30 October 2013.
DOI:10.1111/apa.12494

ABSTRACT
Aim: There are a number of genetic and environmental factors that are associated with an
increased risk of developing coeliac disease. Our aim was to determine whether socioeconomic deprivation increases or reduces the development of the disease.
Methods: A cross-sectional study identified all children <16 years old diagnosed with
coeliac disease in the same tertiary paediatric centre between January 1995 and
December 2011. Data, including age at diagnosis and postcode, were collected and linked
with the quintile rank of the Welsh Index of Multiple Deprivation score 2008, a measure of
socio-economic status.
Results: We included 232 patients and identified a graded association between the
prevalence of coeliac disease and socio-economic deprivation, which showed a higher rate
in children living in more affluent areas. The largest difference was between the lowest
deprivation level (rate/1000 = 1.16) and the highest deprivation level (rate/
1000 = 0.49).
Conclusion: In our population, coeliac disease was more common in children in the higher
socio-economic groupings. The reasons for this are not clear, but perhaps both the hygiene
hypothesis and the health seeking behaviours of parents with high socio-economic status
are possible factors in the more frequent diagnosis of coeliac disease in this group.

AIM
Coeliac disease is a multisystem disease, with an estimated
prevalence of approximately 1% (1,2). Its manifestations
include a small intestinal enteropathy secondary to exposure to gluten, which is completely reversible on a gluten
exclusion diet (3). Historically, patients with CD presented
with severe gastrointestinal symptoms or malnutrition.
More recently, with better serological testing available and
the new guidelines from the European Society of Paediatric
Gastroenterology, Hepatology and Nutrition (ESPGHAN)
published in 2012, symptomatic children will be diagnosed
without biopsy (4). As new environmental factors are
identified and validated as predictors of coeliac disease,
the diagnostic criteria may change to also include asymptomatic children in the group who do not need endoscopy.
This clearly has implications for health economics. Studies
concentrating on potential environmental factors are therefore important as they can help to shape future practice and
management of children with suspected coeliac disease.
Socio-economic deprivation in childhood is an important
consideration, and although this has been studied less

Abbreviations

rigorously, the results are conflicting. One prospective

epidemiological study in Sweden showed that maternal
low socio-economic status was associated with a higher risk
of coeliac disease diagnosis in boys of <2 years of age, but
not girls (5). This is in keeping with the Oxford record
linkage study (6) that showed that low maternal socioeconomic status was a risk factor for coeliac disease.
Another Swedish study showed that children from lower
socio-economic status families had an increased risk of
coeliac disease, even after confounding factors such as
infant feeding practises and infectious episodes were
considered (7). Conversely, one population-based study,

Key notes


ESPGHAN, European Society of Paediatric Gastroenterology,

Hepatology and Nutrition; WIMD, Welsh Index of Multiple
Deprivation.

2013 Foundation Acta Pdiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 289294

Our study identified 232 children <16 years old diagnosed with coeliac disease by a tertiary paediatric
centre.
It found that coeliac disease was more common in
children from affluent areas, with a clear graded
association between the prevalence of the disease
and socio-economic status.
Children diagnosed under the age of two followed the
same trend as older children when it came to the links
between socio-economic status and prevalence of the
disease.

289

Coeliac disease and socio-economic status

Whyte et al.

comparing the relatively poor area of Russian Karelia and

wealthy area of neighbouring Finland, found that coeliac
disease was less common in children in the lower socioeconomic group in Karelia than in Finland (8). The largest
and most recent study of the association between socioeconomic status and coeliac disease in adults found that
coeliac disease was less common in those with lower socioeconomic class (9). Similarly, an unpublished PhD from the
Sheffield/Nottingham area found that incidence rate of
coeliac disease in adults was twice as high in the least
deprived and most affluent quintiles (0.17/1000), compared
with those with coeliac disease in the most deprived and
poorest quintile (0.07/1000) (10). Other studies have found
no difference between socio-economic status (2).
With these conflicting and largely adult-based studies in
mind, our aim was to investigate whether children living in
areas of low socio-economic status in Wales have a higher
or lower incidence of coeliac disease than those living in
areas of higher socio-economic status.

METHODS
A cross-sectional study was performed by identifying all the
paediatric patients (age <16) diagnosed with coeliac disease
in the same tertiary medical centre between January 1995
and December 2012.
The children were diagnosed using the ESPGHAN
criteria 1990 at the tertiary regional paediatric gastroenterology centre. Approximately 80% were followed up at the
same tertiary centre and the other 20% at their local
paediatric unit. These children were from a well-defined
geographical area (covering all Cardiff, Newport and Powys
postcodes), where there was no overlap with any other
paediatric gastroenterology services, as all endoscopies in
children were undertaken only in that unit.
The patients were identified using the local area database
kept by the Department of Paediatric Gastroenterology. In
addition, to ensure that no patients had been missed, we
searched the adult gastroenterology departmental database,
the histopathology departmental database and the endoscopy departmental database for coeliac disease. We did not
identify any other patients with coeliac disease in this way,
and we could therefore confirm maximal ascertainment of
all cases within this geographical area. Data were collected
on each case and included current postcode, gender and age
at diagnosis.
The patients identified were given an unique identifier
number. A list of the unique identifier numbers were linked
with their current postcode. It was not possible to identify
the postcodes at diagnosis as the data were not stored on
our database. However, this is unlikely to affect the results
as evidence shows us that socio-economic status does not
change with age (1113).
The postcodes were then linked with the lower super
output areas, and a quintile of deprivation for each patient
was identified according to the Welsh Index of Multiple
Deprivation (WIMD) for 2008 (14). Lower super output
areas are standard divisions used across Wales and England

290

for collecting, aggregating and reporting statistics. Lower

super output areas are the lowest geographical level for
which clinical information is available this contains an
average of 1500 people, with a minimum of 1000.
The postcode of each patient was also linked with the
Townsend deprivation index, which is another score used
UK-wide to rate deprivation. This score was then linked
with the WIMD score to compare the results to see whether
both deprivation indices identified a similar trend between
coeliac disease prevalence and deprivation.
Statistical analyses
Prevalence rates of coeliac disease by quintile of rank of
WIMD 2008 score were calculated using the total paediatric population for Cardiff, the Vale of Glamorgan, Newport
and Powys as taken from the UK national census 2008 (15).
Multivariate logistic regression analyses were performed to
estimate the effects of possible confounders such as age and
gender. A separate analysis of the WIMD of those children
under two years at diagnosis was performed to consider
whether this group was different from those presenting at
an older age. Chi-square tests were employed to assess any
differences in rates across the WIMD quintiles.
The prevalence per WIMD quintile was then compared
with the prevalence per Townsend deprivation quintile to
see how robust the conclusions were and whether the
deprivation quintiles gave the same prevalence of coeliac
disease in each group.
PASW 18 (SPSS Inc., Chicago, IL, USA) was used for all
statistical analyses.
Ethical considerations
This project was assessed by the local research ethics
committee within the hospital and did not require formal
research ethics committee approval as the patient information remained anonymous, and the final conclusions drawn
could not be linked back to individual patients or groups.

RESULTS
The results have been analysed against the WIMD scores
and are presented by deprivation rank as a measure of
socio-economic status. To try to reduce confusion in the
analysis, we wish to clarify that the lowest quintile of
deprivation equates to the highest socio-economic status
groups and the highest quintile of deprivation refers to the
lowest socio-economic status group.
Demographics of the cohort
A total of 253 children were confirmed with the diagnosis of
coeliac disease during the study period, but 21 cases could
not be linked to a lower super output area either because
the postcode recorded was incorrect or they were part of
new housing which is not included in the current files. Two
postcodes were outwith the geographical area, and these
were excluded. This left a total of 232 children (126 female
and 106 male) whose data were analysed. The age range at
diagnosis for the paediatric patients was 39 weeks to

2013 Foundation Acta Pdiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 289294

Whyte et al.

Coeliac disease and socio-economic status

15 years. The median age at diagnosis for this group was

eight years. Deprivation scores were then attached to the
linked lower super output areas using the WIMD score.
CD and socio-economic status
For children, the prevalence rate of coeliac disease in the
population was 0.78/1000. The prevalence was highest in
those children who had the highest socio-economic status
(i.e. the least deprived group) and lowest in those with the
lowest socio-economic status (i.e. the most deprived group).
There was a clear graded prevalence between the deprivation quintiles, with a modest increased prevalence in the
lowest socio-economic group, deprivation level 5, when
compared with deprivation level 4. The prevalence rate of
coeliac disease in the high socio-economic status populations was 1.16/1000 and 0.49/1000 in the lowest socioeconomic status population (Table 1 and Fig. 1).
A chi-square test showed a significant difference
(p < 0.001) over the deprivation quintiles with the greatest
difference being between the highest and lowest deprivation
levels. The odds ratio for the rate per 1000 of coeliac disease
was not significant between the lowest and low quintiles
(0.794, 95% CI 0.5501.146). The odds ratio of coeliac
disease was significant between the lowest deprivation and
the mid, high and highest deprivation (Table 2). Differences
between the low, middle, high, highest quintiles were not
significant (p > 0.05).
Rate of coeliac disease was modelled using logistic
regression on WIMD with age and gender as additional
confounders. Gender had no effect, and age had a significant but small effect. The relative differences between the
deprivation levels were not affected by the confounding.

Prevalence rate of CD in children with quintile rank of

townsend deprivation score
WIMD is a Wales only measure of deprivation. To ensure
that the results presented here are not some artefact of
WIMD, we also used a UK-wide measure based on
Townsend deprivation quintiles. The results were largely
very similar between the two measures. The most appreciable difference comes with the highest rate of deprivation
although the difference was not significant (Fig. 4).

CONCLUSIONS
In our well-defined South Wales childhood population,
there was a clear graded association between the prevalence
of coeliac disease and socio-economic status, with the rate
higher in children living in more affluent areas compared
with those living in areas with higher deprivation. There
was an increased prevalence in females in all areas of
deprivation. Children who were diagnosed under the age of
two years followed the same trend as older children in the
linkage between socio-economic status and coeliac disease
prevalence.
This is the first population-based cohort study to have
been undertaken looking specifically at children with
coeliac disease and socio-economic status. We are confident that every paediatric patient with coeliac disease

Gender and deprivation

Female children had an overall higher rate of coeliac
disease than male children (0.87/1000 and 0.69/1000,
respectively). Also female children had a higher rate of
coeliac disease in each of the individual deprivation groups.
None of the differences in rates were statistically significant
(Fig. 2).
Age and deprivation
Looking specifically at those who were two or under at the
time of diagnosis (n = 55), the prevalence was again higher
in those with lower deprivation. The numbers were not
great enough to justify splitting the genders (Fig. 3).

Figure 1 To show the rate of coeliac disease with deprivation quintile of WIMD
deprivation level.

Table 1 Prevalence of coeliac disease by quintile rank of Welsh Index of Multiple Deprivation score

No Coeliac
Coeliac
Total

Lowest deprivation
Num (Rate/1000)

Low deprivation
Num (Rate/1000)

Mid deprivation
Num (Rate/1000)

High deprivation
Num (Rate/1000)

Highest deprivation
Num (Rate/1000)

Total
Num (Rate/1000)

66920
78 (1.16)
66998

48651
45 (0.92)
48696

42223
25 (0.59)
42248

63110
46 (0.73)
63156

77394
38 (0.49)
77432

298298
232 (0.78)
298530

2013 Foundation Acta Pdiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 289294

291

Coeliac disease and socio-economic status

Whyte et al.

Table 2 Regression modelling to show deprivation quintile of Welsh Index of Multiple Deprivation score and odds ratio of coeliac disease
Odds ratios with the least deprived as the reference category
Unadjusted

Lowest
Low
Mid
High
Highest

Adjusted for gender and age

OR

CI low

CI high

OR

CI low

CI high

1.000
0.794
0.508
0.625
0.421

0.550
0.324
0.434
0.286

1.146
0.797
0.900
0.621

1.000
0.794
0.506
0.607
0.417

0.550
0.323
0.421
0.283

1.146
0.794
0.877
0.615

1.6

Coeliac rate in children split by gender with

1.8

Coeliac rate in children by WIMD and

deprivation quintile

1.4

1.6

Townsend deprivation fifths

Male

1.2

Female

WIMD
Townsend

1.2

Rate per 1000

Rate per 1000

1.4

1
0.8
0.6

1
0.8
0.6

0.4

0.4

0.2
0
Lowest dep Low dep

Mid dep

High dep Highest dep

0.2

Total

WIMD deprivation level

0

Figure 2 Coeliac disease rate in children split by gender with deprivation

quintile.

Lowest dep Low dep

Mid dep

High dep Highest dep

Total

Deprivation level

Figure 4 Prevalence rate per quintile using WIMD and Townsend deprivation
quintiles.
3

Coeliac rate in children 2 years or less

Rate per 1000

2.5

with deprivation fifth

2
1.5
1
0.5
0
Lowest dep Low dep

Mid dep

High dep Highest dep

Total

WIMD deprivation level

Figure 3 Prevalence of coeliac disease in children under 2 years with

deprivation quintile.

within that defined area was identified as all endoscopic

diagnoses were made in the regional unit and there was
rigorous cross-checking of other relevant databases.

292

The WIMD scores are a recognised measure of how

deprived an area is within Wales, when considering eight
different categories of deprivation (14). Deprivation ranks
are calculated for each area according to how deprived
they are relative to each other. These indicators of deprivation are shown to be linked with increased morbidity and
mortality (15,16). However, the WIMD scores for areas give
the overall deprivation for the people living in that area, but
it may be the case specific households have very different
levels of deprivation within the same lower super output
areas. Although the Townsend deprivation scores have
similar limitations, by analysing the data using both of these
scores, we have provided more robust evidence that the
association we have identified is real.
This study was limited by the relatively small number of
patients in the geographical area studied. However, in
limiting the numbers to this geographical area, we could be
sure that all paediatric patients were identified, and we did
not have to consider patients from the same geographical
area that are cared for by another centre. A future study,

2013 Foundation Acta Pdiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 289294

Whyte et al.

ideally UK-wide, including more patients would be helpful

to confirm the findings of this study. Indeed, in using the
WIMD score, it is possible to compare the data with
another geographical area in England or Wales so to see
whether the findings are the same.
Previous data from other studies looking at socio-economic status versus prevalence have shown varying trends.
Our data clearly show that childhood coeliac disease is
associated with the highest socio-economic status quintiles.
We can only speculate on the reasons for this, but one
suggestion is that the parents of children in the higher socioeconomic status group are more likely to seek medical
attention than other those from a lower socio-economic
status group. There is significant evidence from groups who
have studied a variety of chronic diseases that those parents
with higher socio-economic status have more health seeking
behaviours than those in lower socio-economic status groups
(11,12,16,17). Indeed, we know that there is an increasing
incidence of children presenting with mild abdominal symptoms (2,18), and so it may be possible to suggest that the
parents of children in high socio-economic status areas are
more likely to seek advice for minor symptoms of coeliac
disease, such as mild abdominal pain than parents who have
a lower socio-economic status. Equally, it is possible that
children of high socio-economic status are more likely to be
presented for serological screening if a first degree relative
has been diagnosed than in those families from lower socioeconomic status groups.
A second reason for our findings is the hygiene hypothesis. It is believed that early colonisation with a normal
microbacterial flora encourages regulation and maturation
of the immune system (1921). The hygiene hypothesis
suggests that deficiency of these responses of the immune
system in early life increases the susceptibility of all types of
immune responses, including inappropriate responses that
lead to autoimmune disease and allergy. Improved public
health and living conditions have meant that particularly in
the homes of those of a higher socio-economic status group,
the exposure to foreign antigens is much reduced (19,20).
The frequency of gastrointestinal infections seems to have
an effect on the development of coeliac disease, as it alters
the microbacterial flora of the gut. Some studies have noted
a higher coeliac disease rate after enterovirus and adenovirus infection due to the similarities between these bacteria
and the structure of gliadin (21,22); one would therefore
expect the lower socio-economic status groups, who have
higher rates of gastrointestinal infection (23) to have a
higher coeliac disease risk, but they do not. Therefore, we
hypothesise that higher numbers of gastrointestinal infective episodes protect against coeliac disease, as the immune
system remains appropriately regulated, so decreasing the
likelihood of inappropriate activation of the immune system
against a normal antigen such as gliadin.
Whatever the reasons for the strong association we have
found between the prevalence of coeliac disease and high
socio-economic status in our region and future studies to
confirm these findings in other areas of the UK will be
necessary.

Coeliac disease and socio-economic status

CONFLICT OF INTEREST
The authors have no conflict of interests to disclose.

FUNDING SOURCE
The authors have no funding source to disclose.

References
1. Hawkes ND, Swift GL, Smith PM, Jenkins HR. Incidence and
presentation of coeliac disease in South Glamorgan. Eur J
Gastro Hep 2000; 12: 3459.
2. White LE, Merrick VMW, Bannerman E, Wilson DC, Russell
R, Basude D, et al. Trends in paediatric coeliac disease
1990-2009: a UK regional retrospective audit of cases. Arch Dis
Child 2012;97(Suppl 1):A1186. Paper presented at: RCPCH
Spring conference; 2012; Glasgow, UK.
3. Tommasini A, Not T, Ventura A. Ages of coeliac disease. World
J Gastroenterol 2011; 17: 366571.
 IR, Mearin ML,
4. Husby S, Koletzko S, Korponay-Szabo
Phillips A, Shamir R, et al. ESPGHAN Gastroenterology
Committee; European Society for Pediatric
Gastroenterology, Hepatology, and Nutrition.ESPHGAN
Guidelines for the diagnosis of coeliac disease in children
and adolescents: an evidenced based approach. JPGN 2012;
54: 136.
5. George EK, Mearin ML, Franken HC, Houwen RH, Hirasing
RA, Vandenbroucke JP. Twenty years of childhood coeliac
disease in the Netherlands: a rapidly increasing incidence? Gut
1997; 40: 616.
6. Roberts SE, Williams JG, Meddings D, Davidson R, Goldacre
MJ. Perinatal risk factors and coeliac disease in children and
young adult: a record linkage study. Aliment Phamacol Ther
2009; 29: 22231.
7. Ivarsson A. The Swedish epidemic of coeliac disease explored
using an epidemiological approach some lessons to be learnt.
Best Pract Res Clin Gastroenterol 2005; 19: 42540.
8. Kondrashova A, Mustalahti K, Kaukinen K, Viskari H,
Volodicheva V, Haapala AM, et al. Lower socioeconomic
status and inferior hygiene may protect against coeliac disease.
Ann Med 2008; 40: 22331.
9. Olen O, Bihagen E, Rasmussen F, Ludvigsson JF.
Socioeconomic position and education in patients with coeliac
disease. Dig Liver Dis 2012; 44: 4716.
10. Lewis NR. Is the diagnosis of coeliac disease associated with
socioeconomic status? A population based study [PhD thesis].
University of Nottingham; 2007.
11. Braveman PA, Egerter SA, Cubbin C, Marchi KS. An approach
to studying social disparities in healthcare. Am J Public Heath
2004; 94: 213948.
12. National Research Council (US) Panel on Race, Ethnicity, and
Health in Later Life; Bulatao RA, Anderson NB, editors.
Understanding racial and ethnic differences in health in late
life: a research agenda. [online] Washington (DC): National
Academies Press (US), 2004. 4, Socioeconomic Status.
Available from: http://www.ncbi.nlm.nih.gov/books/
NBK24690/ (Accessed on August 8, 2012).
13. Payne RA, Abel GA. UK indices of multiple deprivation - a way
to make comparisons across constituent countries easier.
Health Stat Q 2012; 53: 116.
14. Welsh Government (2008) WIMD 2008 [online] Available
from: https://statswales.wales.gov.uk/Catalogue/
Community-Safety-and-Social-Inclusion/
Welsh-Index-of-Multiple-Deprivation/WIMD-2008/
OverallIndex-by-Area-Domain. (Accessed on July 5, 2012).

2013 Foundation Acta Pdiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 289294

293

Coeliac disease and socio-economic status

Whyte et al.

15. Office for National Statistics: Census 2008 [online]. Available

from: http://www.ons.gov.uk/ons/search/index.html?pageSize
=50&sortBy=none&sortDirection=none&newquery=2008+
wales (Accessed on April 4, 2012).
16. Kozyrskyj AL, Dahl ME, Chateau DG, Mazowita B,
Klassen TP, Law BJ. Evidence-based prescribing of
antibiotics for children: role of socioeconomic status
and physician characteristics. CMAJ 2004; 171:
13945.
17. Wamala S, Merlo J, Bostrom G, Hogstedt C. Perceived
discrimination, socioeconomic disadvantage and refraining
from seeking medical treatment in Sweden. J Epidemiol
Community Health 2007; 61: 40915.
18. Whyte LA, Humphries L, Tuthill DP, Davies IH, Jenkins HR.
Coeliac disease is small intestinal biopsy always necessary for
diagnosis? G82(P) (Abstract only). Arch Dis Child 2012; 97
(Suppl 1): A1100.

294

19. Rautava S, Isolauri E. The development of gut immune

responses and gut microbiota: effects of probiotics in
prevention and treatment of allergic disease. Curr Issues Intest
Microbiol 2002; 3: 1522.
20. Custovic A, Simpson A, Woodcock A. Importance of indoor
allergens in the induction of allergy and elicitation of allergic
disease. Allergy 1998; 53(48 Suppl): 11520.
21. De Palma G, Kamanova J, Cinova J, Olivares M, Drasarova H,
Tuckova L, et al. Modulation of phenotypic and functional
maturation of dendritic cells by intestinal bacteria and gliadin:
relevance for celiac disease. J Leukoc Biol 2012; 92: 104354.
22. DuPont AW, DuPont HL. The intestinal microbiota and
chronic disorders of the gut. Nat Rev Gastroenterol Hepatol
2011; 8: 52331.
23. Simonsen J, Frisch M, Ethelberg S. Socioeconomic risk factors
for bacterial gastrointestinal infections. Epidemiology 2008; 19:
28290.

2013 Foundation Acta Pdiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 289294

Copyright of Acta Paediatrica is the property of Wiley-Blackwell and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.