This article has Open Peer Review reports available.
Risks of hospitalization and drug consumption in children and young adults with diagnosed celiac disease and the role of maternal education: a population-based matched birth cohort study
© Canova et al. 2015
Received: 5 November 2015
Accepted: 23 December 2015
Published: 5 January 2016
Celiac disease (CD) may affect healthcare use in children and young adults. Socio-economic factors may act as a confounder or effect modifier. We assessed such hypotheses in a population-based birth cohort of young celiac subjects and references matched by maternal education.
The cohort included all newborns recorded in the Medical Birth Register of Friuli-Venezia Giulia Region (Italy) between 1989 and 2011. CD incident cases were identified through pathology reports, hospital discharges and copayment exemptions and matched with up to five references by sex, year of birth and maternal education. Cox regression models were used to estimate Hazard Ratios (HRs) for major causes of inpatient diagnosis and drug prescription occurring after diagnosis in CD patients compared to references, stratifying by time of first event and maternal education.
We identified 1294 CD cases and 5681 references. CD cases had a higher risk of hospital admission for any cause (HR: 2.34; 95 % CI 2.08-2.63) and for all major ICD9-CM categories except obstetric complications, skin and musculoskeletal diseases, and injuries and poisoning. Prescription of all major ATC drug categories, except dermatologicals and genito-urinary medications, was significantly increased in CD subjects. For most outcomes, HRs were highest in the first year after CD diagnosis but remained significant after five or more years. HRs were similar across different categories of maternal education.
Diagnosed CD subjects had a higher risk of hospitalization and medication use compared to the general population, even five or more years after diagnosis, with no effect modification of maternal education.
Celiac disease (CD) is an immune-mediated small intestinal enteropathy triggered by gluten affecting approximately 0.5-1 % of the European population and frequently diagnosed during childhood .
Originally considered a rare malabsorption syndrome of childhood, CD is now recognized as a systemic disorder potentially affecting many organ systems beyond the gastrointestinal tract . The clinical picture is variable and several complications and comorbidities can occur, including consequences of malnutrition (e.g., low bone mineral density, anemia), infections, neurological and autoimmune diseases (e.g. type I diabetes, thyroid disorders), [2, 3] and cancers such as lymphoma . The only available treatment for CD is a life-long gluten-free diet. This dietary regimen is effective in controlling symptoms, however it is burdensome for the patient and his/her family and cannot prevent many complications and comorbidities of CD . Moreover, children with CD, even if adherent to the gluten-free diet, often suffer from functional gastrointestinal disorders  and psychological disturbances  that may lead to impaired quality of life  and greater healthcare needs.
Few studies have investigated and reported an increased use of healthcare services and drugs and associated costs in adults with CD compared to the general population; [8–11] specific data regarding children and adolescents with CD are scarce. Furthermore, only one study  has so far assessed the long-term persistence of increased healthcare use after CD diagnosis.
Healthcare utilization pattern is affected by educational and socio-economic factors, which may reflect socio-economic differences regarding both morbidity and access to healthcare services. Currently, social determinants are emphasised as a major source of health inequalities within developed countries . As far as we know, the impact of socio-economic status on healthcare utilization associated with CD has never been investigated.
Using a population-based birth cohort of subjects born between 1989 and 2011 and followed-up until the end of 2012, we examined the risk of first hospitalization and drug consumption at different time intervals from diagnosis in CD children, adolescents and young adults compared to reference individuals matched on age, gender and maternal education level. This study had three objectives: i) to assess whether children and adolescents with CD use more healthcare resources after diagnosis compared to subjects without CD of similar age, sex, and maternal education; ii) to assess whether the possible increased healthcare use is limited to the short period after CD diagnosis or persists long-term; iii) to assess whether the risk of healthcare use in CD subjects compared to unaffected subjects is modified by maternal education.
Setting and study population
We used a matched cohort design. The underlying study population consisted of 213 635 subjects (excluding abortions, stillbirths, and neonatal deaths within 30 days of birth) born in Friuli-Venezia Giulia (FVG) Region in Italy in the period 1989-2011 identified using the regional Medical Birth Register that contains socio-demographic data on the parents, details of the pregnancy, labour, and delivery, and the newborn at birth.
Italy has a tax-based National Health Service, which provides universal coverage to all Italian and European Union citizens resident in the country regardless of income, and it is organized at a regional level. FVG Region is covered by a regional integrated healthcare system developed in the 1980s with the goal of automatically collecting and pooling data on healthcare funded by the National Health Service using a unique regional identification code that has been described in detail previously .
The following Regional health data were used in this study (up to 2012): mortality records; pathology reports from all the pathology departments in the FVG Region, coded in the Systematized Nomenclature of Medicine (SnoMed); hospital discharge records collected during episodes of inpatient care occurring within the FVG Region or in Italy outside the Region with up to six diagnostic codes in the International Classification of Diseases ninth revision Clinical Modification (ICD9-CM); exemption from healthcare copayment coded in a national coding system; drug prescription records coded in the Anatomical Therapeutic Chemical (ATC) Classification System (available from 1995).
Ethics, consent and permissions
The study was approved by the Institutional Review Board of the University of Padua (Italy). Patients’ data were completely anonymized and de-identified prior to analysis. Therefore, no informed consent and no Ethics Committee approval were required.
We defined CD as having at least one of the following : a pathology report of villous atrophy (SnoMed codes D6218, D6318, M58, M58005, M58006, and M58007); a hospital discharge record with an ICD9-CM code 5790 in any diagnosis; an exemption from health-care copayment with code I0060 according to the Italian national coding system. In Italy, all patients with CD can obtain clinical tests and gluten-free food free of charge from the National Health Service, provided they have a CD diagnosis certified by a specialist on the basis of a biopsy report. CD onset was defined as the earliest date identifiable in the three sources of information (pathology, hospital and copayment exemption).
Risks of first hospital admission according to ICD9-CM classification after date of diagnosis in CD subjects compared to matched referencesa
REFERENCES (n = 5681)
CD (n = 1294)
HR (95 % CI)
Infectious and parasitic diseases
Endocrine. metabolic. immunity dis.
Dis. of the blood and blood–forming organs
Dis. of the nervous system and sense organs
Dis. of the circulatory system
Dis. of the respiratory system
Dis. of the digestive system
Dis. of the genito–urinary system
Complications of pregnancy and puerperium
Dis. of skin and subcutaneous tissue
Dis. of the muscoloskeletal tissue
Congenital anomalies/Perinatal conditions
Injuries and poisoning
Factors influencing health status
For each case with a CD diagnosis up to five reference individuals were selected from subjects in the Medical Birth Register alive on the date of CD diagnosis (index date), matched on gender, year of birth and maternal education levels (up to middle school; secondary school; university level). Matching was implemented to control for confounding and improve statistical efficiency and to be able to attribute the beginning of follow-up (index date) that corresponds to CD diagnosis to reference individuals. Follow-up started at 30 days after the index date (to exclude hospitalizations and prescriptions concomitant to CD diagnosis) until the earliest of: end of the study (December 31, 2012), death, migration out of the area, or first hospitalization for each group of causes/first drug prescription for each ATC level. Concerning drug prescription outcomes, analyses were restricted to subjects with index date ≥1995 (because drug prescription data are available from that year). Cox regression models were used to estimate the Hazard Ratios (HRs) and 95 % confidence intervals (CI) for each of the healthcare use outcome in patients with CD compared to reference individuals within the same stratum. The assumption of proportional hazards was investigated by studying graphs over the log cumulative hazards function and the Shoenfeld residuals, and verified by a global test of rho. Analyses were also stratified by time of first hospital admission/drug prescription, looking at outcomes occurred between 30-365 days and >365 days after the index date. For the latter analyses only subjects with at least one year of follow-up (alive and resident in the area one year after the index date and without any hospitalization/drug prescription for each ICD9-CM/ATC category within one year from the index date) were considered. In order to assess the possible long-term effect on healthcare use in CD individuals compared to references, we examined the risk of hospitalization and drug consumption five or more years after diagnosis. For this analysis follow-up started five years after the index date and only subjects alive and resident in the area at this time were included. In this analysis, we included subjects with hospitalization/drug prescription within five years from the index date.
In order to evaluate a possible modifying effect of socioeconomic status, analyses were subsequently stratified by maternal education levels.
We identified 1294 subjects with CD and 5681 reference individuals. Biopsy confirmation was available for 920 (71 %) CD individuals; 179 (14 %) were instead identified by hospital discharges and copayment records, 112 (9 %) only by hospital discharge records and 83 (6 %) only by copayment records. Mean age at diagnosis (age at index date for references) was 7.2 ± 5.3 years (min 0, max 23.8 years). Analyses of drug prescriptions were restricted to 1215 CD subjects diagnosed after 1995 and 5366 matched references. 1166 CD cases (5100 references) were alive and resident in the Region one year after the index date and 777 (3316 references) 5 years after.
Risks of first drug prescription according to first level of Anatomical Therapeutic Chemical (ATC) Classification after date of diagnosis in CD subjects compared to matched referencesa
REFERENCES (n = 5366)
CD (n = 1215)
HR (95 % CI)
Alimentary tract and metabolism
Blood and blood forming organs
Genito–urinary system and sex hormones
Systemic hormonal preparations. excluding sex hormones and insulins
Antiinfectives for systemic use
Antineoplastic and immunomodulating agents
Antiparasitic products. insecticides and repellents
Risks of first hospital admission according to ICD9–CM classification after date of diagnosis in CD subjects compared to matched references, stratified by first outcome occurrence (30–365 days after the index date/>365 days after the index date)a
First outcome occurrence
>1 year after index dateb
HR (95 % CI)
HR (95 % CI)
Risks of first drug prescription according to first level of Anatomical Therapeutic Chemical (ATC) Classification after date of diagnosis in CD subjects compared to matched references, stratified by first outcome occurrence (30–365 days after the index date/>365 days after the index date)a
First outcome occurrence
>1 year after index dateb
HR (95 % CI)
HR (95 % CI)
Among CD cases, 160 (12.4 %) were born to graduated mothers, 648 (50.1 %) to mothers with secondary school education and 471 (36.4 %) to less educated mothers (up to middle school) (15 cases had missing information on maternal education); by study design, references showed similar distributions, because each CD case was matched to references with the same maternal education.
In the analysis stratified by maternal education level, we found no clear modifying effect of maternal education (Additional file 3: Table S3, Additional file 4: Table S4). The risks of being hospitalized for any cause among CD subjects compared to their references born to mothers with similar maternal education were: HR 2.64 (95 % CI 1.79-3.89) for university level, HR 2.33 (95 % CI 1.97-2.76) for secondary school education and HR 2.30 (95 % CI 1.92-2.75) for primary/middle school education. Neither were there any differences in HRs for specific groups of causes according to maternal education although the low number of subjects with graduated mothers did not allow precise or computable estimates in that group (Additional file 3: Table S3). Similar risks were also seen for the majority of drug prescription categories in the groups of subjects stratified by maternal education (Additional file 4: Table S4).
In this large population-based birth cohort, we followed-up 1294 children and adolescents with CD and 5,681 peers without CD, matched by socio-demographic characteristics, for hospitalization and drug prescription after CD diagnosis. Both hospitalization and drug consumption were increased in CD cases in most disease categories and a wide range of drug classes.
A small number of previous studies have reported a greater use of medical services and drugs in CD patients compared to controls after diagnosis, [9–11] but only one included children and adolescents and provided a long follow-up . To our knowledge, this is the first study on the long-term healthcare impact of paediatric CD conducted in Southern Europe. Healthcare use is influenced by culture and type of healthcare system, therefore it is worthwhile to conduct similar studies in different settings. Overall the results of the present study are consistent with current knowledge that childhood CD is associated with multi-organic complications and comorbidities and also with non-specific health complaints [2, 3].
For most disease and drug categories (in particular endocrine/metabolic/immune and digestive disorders and antineoplastic/immunomodulating agents), the risk of hospitalization or prescription was highest during the first year of follow-up, indicating that CD-related health problems are likelier to occur (or become recorded) shortly after diagnosis, when clinical surveillance is greatest and dietary treatment may have not yet displayed its effects. However, our findings suggest that CD-related health problems can develop also later, since for most conditions the risk of a first event after diagnosis was increased in CD subjects compared to reference individuals also after the first year of follow-up. To assess whether the increased healthcare use in CD persisted long-term, we also examined hospital admissions and drug prescriptions occurring five or more years after diagnosis; subjects who had hospital admissions or drug prescription within five years were included in this analysis. We found that even beyond five years of follow-up, the risk of hospitalization and drug prescription was greater in CD cases compared to reference individuals. Such results are consistent with a previous study conducted in the United Kingdom showing that healthcare costs in patients with CD have a peak around time of diagnosis and decline thereafter, but remain higher than in controls for up to 10 years . Unlike the abovementioned study, we did not focused on costs but rather on the risk of events.
Different factors may explain why diagnosed CD children and adolescents used more healthcare resources than their references even five years after diagnosis. First of all, as in any chronic disorder, the frequent contact with the healthcare system can increase the chance of being diagnosed and subsequently treated for other conditions. In our opinion, however, surveillance bias alone cannot explain the differences we observed. Secondly, CD has been associated with non-specific health complaints such as functional gastrointestinal disorders and depressive and anxiety symptoms that can persist over time despite appropriate treatment [5, 6]. Consistently, when we examined main diagnoses and drug categories in detail, we found that anxiety and somatoform disorders (ICD9-CM 300), personality disorders (ICD9-CM 301), eating, sleep and psychogenic pain disorders (ICD9-CM 307), ill-defined symptoms concerning nutrition and abdomen (ICD9-CM 783 and 789), prescriptions of antacids (ATC A02), psycholeptics (ATC N05), and psychoanaleptics (ATC N06) were all more common among individuals with CD. Young CD patients may also be exposed to a persistent risk of developing complications of nutritional deficiencies due to incomplete remission of gut mucosal abnormalities or to dietary restrictions . This is suggested by our finding that CD subjects had a higher proportion of iron-deficiency anemia (ICD9-CM 280), thrombocytopenia (ICD9-CM 287), leukocytopenia (ICD9-CM 288), and prescription of anti-anemic preparations (ATC B03). Infectious diseases, particularly those provoked by encapsulated bacteria, may also complicate the clinical course of CD in the long term,  as a result of hyposplenism  and perhaps of altered intestinal permeability and nutritional deficiencies. We also observed a greater frequency of hospital admissions for otitis media (ICD9-CM 381), pneumonia (ICD9-CM 481, 482, 485, 486), and urinary infections (ICD9-CM 590, 599) in CD, as well as a higher consumption of systemic anti-infectives (ATC J). Finally, a greater occurrence of several diseases, mainly but not exclusively of autoimmune origin, has been described in patients with CD. Although our study was not designed to estimate the risk of specific associated conditions, we think that some of the reported risks could be partially explained by underlying comorbidities. For instance, the greater proportion of admissions for diabetes (ICD9-CM 250) and prescription of antidiabetic agents (ATC A10) and thyroid drugs (ATC H03) could reflect the acknowledged associations between CD and type I diabetes mellitus and autoimmune thyroid disorders . Similarly, the well-known comorbidity between CD and chromosomal anomalies, such as Down  and Turner syndrome,  may in part explain a higher prevalence of congenital malformations in CD subjects . Several neurologic and psychiatric conditions, such as epilepsy (ICD9-CM 345)  and autism (ICD9-CM 299), [3, 21] have been linked to CD although with inconclusive results, and may explain our finding of a higher risk of hospital admission for neurologic and mental disorders and of prescription of antiepileptic drugs (ATC N03) in CD cases. Concerning digestive disorders, most of the increased risk in CD subjects is likely due to CD itself, especially in the first year of follow-up, since intestinal malabsorption (ICD9-CM 579) was not excluded from the analysis of hospitalization outcomes. Our CD cases also had a high frequency of hospitalizations for Crohn’s disease (ICD9-CM 555) and ulcerative colitis (ICD9-CM 556), both of which have been potentially linked to CD although earlier research has shown conflicting results [22–25].
Although CD is a chronic illness, many of its symptoms and effects can be reversed or controlled with a gluten-free diet . Therefore, one might suppose that with a well-conducted gluten-free diet and an appropriate follow-up, the long-term outcome of CD paediatric patients would be similar to the general population ones. Our data showed that this is not the case, at least not in FVG Region. This issue deserves further investigation to confirm our findings and to assess whether the unfavorable long-term outcome of paediatric CD can be attributed only to complications and comorbidities that are not prevented by the gluten-free diet or also to an inadequate management of CD.
Socioeconomic factors in early childhood have a well-known impact on several health outcomes in adulthood, although evidence regarding early exposure impact on childhood and adolescence health is more limited . Moreover, socio-economic level has been associated with the risk of developing CD in some studies, but results are conflicting [13, 27–32]. For this reason we decided to control for the possible confounding effect of socio-economic status by design matching reference individuals by maternal education. To the best of our knowledge this has never been done before. We also hypothesised that a poorer educational and socio-economic condition may adversely affect health of CD children and adolescents compared to reference individuals within the same socioeconomic stratum, possibly because of a poorer adherence to gluten-free diet. However, the analyses stratified by maternal education showed similar HR of hospital admission and drug prescription across groups. These results, although limited by the low number of events particularly for the hospitalization outcomes and among the most educated mothers, do not suggest a modifying effect of maternal education on the association between CD and healthcare use, but as any novel tested hypothesis they need to be further replicated in other setting/studies.
Strengths of our study besides the matched design include the large population size, the longitudinal design with long follow-up period and the availability of objective data, with no risk of recall bias. The main limitations are due to the use itself of administrative data with the inability to directly verify the presence of CD, and obtain data on compliance with a gluten-free diet. As regards the identification of CD cases, however, we are confident that our data are accurate since 71 % of cases had a biopsy report of villous atrophy, which has been found to be highly specific (positive predictive value 95 %) for CD in a validation study . Of the remaining cases, 20 % had a copayment exemption, which in Italy is based on a physician’s diagnosis. Treatment with a gluten-free diet may result in decreased consumption of health care services and drugs,  although some studies suggested that, even after years of compliance with a gluten-free diet, CD patients have significantly impaired quality of life . Moreover, we were unable to retrieve data on outpatient visits and diagnostic procedures, which would have drawn a more comprehensive picture of healthcare utilization. A further limitation is that we did not exclude from the analyses those subjects already affected by chronic disorders other than CD before the index date. Therefore, the higher healthcare use we observed in CD individuals may in part depend on comorbidities that were already present before the onset of CD. Our study should be therefore regarded as exploratory rather than confirmatory because we did not examine temporal relationships between CD onset and outcomes. The aim of this study was to describe healthcare use pattern after CD diagnosis without inferring causal relationships between CD and associated conditions.
Using a population-based birth cohort matched study design, we showed that young people with diagnosed CD were at increased risk of later hospitalization and medication use compared to unaffected peers from the general population independently from socio-economic status at birth. Greater healthcare consumption may be explained both by non-specific health complaints and by complications and associated conditions of CD. Whatever the mechanism, our findings show that the burden associated with paediatric CD is heavy and persists in the long term, despite the existence of an effective treatment. Further longitudinal studies should be devoted to identify factors associated to a bad prognosis of CD in the paediatric population, in order to improve the management of this disease.
This study was financially supported by the University of Padua with a research grant to Dr. Cristina Canova [F.S. 4.18.01.05].
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Mustalahti K, Catassi C, Reunanen A, Fabiani E, Heier M, McMillan S, et al. The prevalence of celiac disease in Europe: results of a centralized, international mass screening project. Ann Med. 2010;42:587–95.View ArticlePubMedGoogle Scholar
- Green PH, Jabri B. Coeliac disease. Lancet. 2003;362:383–91.View ArticlePubMedGoogle Scholar
- Newton KP, Singer SA. Celiac disease in children and adolescents: special considerations. Semin Immunopathol. 2012;34:479–96.View ArticlePubMedGoogle Scholar
- Tio M, Cox MR, Eslick GD. Meta-analysis: coeliac disease and the risk of all-cause mortality, any malignancy and lymphoid malignancy. Aliment Pharmacol Ther. 2012;35:540–51.View ArticlePubMedGoogle Scholar
- Turco R, Boccia G, Miele E, Giannetti E, Buonavolontà R, Quitadamo P, et al. The association of coeliac disease in childhood with functional gastrointestinal disorders: a prospective study in patients fulfilling Rome III criteria. Aliment Pharmacol Ther. 2011;34:783–9.View ArticlePubMedGoogle Scholar
- Mazzone L, Reale L, Spina M, Guarnera M, Lionetti E, Martorana S, et al. Compliant gluten-free children with celiac disease: an evaluation of psychological distress. BMC Pediatr. 2011;11:46. 2431-11-46.PubMed CentralView ArticlePubMedGoogle Scholar
- Altobelli E, Paduano R, Gentile T, Caloisi C, Marziliano C, Necozione S, et al. Health-related quality of life in children and adolescents with celiac disease: survey of a population from central Italy. Health Qual Life Outcomes. 2013;11:204. 7525-11-204.PubMed CentralView ArticlePubMedGoogle Scholar
- Long KH, Rubio-Tapia A, Wagie AE, Melton LJ 3rd, Lahr BD, Van Dyke CT, et al. The economics of coeliac disease: a population-based study. Aliment Pharmacol Ther. 2010;32:261–9.PubMed CentralView ArticlePubMedGoogle Scholar
- Roos S, Wilhelmsson S, Hallert C. Swedish women with coeliac disease in remission use more health care services than other women: a controlled study. Scand J Gastroenterol. 2011;46:13–9.View ArticlePubMedGoogle Scholar
- Ukkola A, Kurppa K, Collin P, Huhtala H, Forma L, Kekkonen L, et al. Use of health care services and pharmaceutical agents in coeliac disease: a prospective nationwide study. BMC Gastroenterol. 2012;12:136. 230X-12-136.PubMed CentralView ArticlePubMedGoogle Scholar
- Violato M, Gray A, Papanicolas I, Ouellet M. Resource use and costs associated with coeliac disease before and after diagnosis in 3,646 cases: results of a UK primary care database analysis. PLoS One. 2012;7:e41308.PubMed CentralView ArticlePubMedGoogle Scholar
- Marmot M, Allen J, Bell R, Bloomer E, Goldblatt P. WHO European review of social determinants of health and the health divide. Lancet. 2012;380:1011–29.View ArticlePubMedGoogle Scholar
- Canova C, Zabeo V, Pitter G, Romor P, Baldovin T, Zanotti R, et al. Association of maternal education, early infections, and antibiotic use with celiac disease: a population-based birth cohort study in northeastern Italy. Am J Epidemiol. 2014;180:76–85.View ArticlePubMedGoogle Scholar
- Penagini F, Dilillo D, Meneghin F, Mameli C, Fabiano V, Zuccotti GV. Gluten-free diet in children: an approach to a nutritionally adequate and balanced diet. Nutrients. 2013;5:4553–65.PubMed CentralView ArticlePubMedGoogle Scholar
- Ludvigsson JF, Olen O, Bell M, Ekbom A, Montgomery SM. Coeliac disease and risk of sepsis. Gut. 2008;57:1074–80.View ArticlePubMedGoogle Scholar
- Di Sabatino A, Rosado MM, Cazzola P, Riboni R, Biagi F, Carsetti R, et al. Splenic hypofunction and the spectrum of autoimmune and malignant complications in celiac disease. Clin Gastroenterol Hepatol. 2006;4:179–86.View ArticlePubMedGoogle Scholar
- Elfstrom P, Sundstrom J, Ludvigsson JF. Systematic review with meta-analysis: associations between coeliac disease and type 1 diabetes. Aliment Pharmacol Ther. 2014;40:1123–32.View ArticlePubMedGoogle Scholar
- Marild K, Stephansson O, Grahnquist L, Cnattingius S, Söderman G, Ludvigsson JF. Down syndrome is associated with elevated risk of celiac disease: a nationwide case-control study. J Pediatr. 2013;163:237–42.View ArticlePubMedGoogle Scholar
- Wingren CJ, Agardh D, Merlo J. Congenital anomalies and childhood celiac disease in Sweden. J Pediatr Gastroenterol Nutr. 2012;55:736–9.View ArticlePubMedGoogle Scholar
- Ludvigsson JF, Zingone F, Tomson T, Ekbom A, Ciacci C. Increased risk of epilepsy in biopsy-verified celiac disease: a population-based cohort study. Neurology. 2012;78:1401–7.View ArticlePubMedGoogle Scholar
- Ludvigsson JF, Reichenberg A, Hultman CM, Murray JA. A nationwide study of the association between celiac disease and the risk of autistic spectrum disorders. JAMA Psychiatry. 2013;70:1224–30.View ArticlePubMedGoogle Scholar
- Yang A, Chen Y, Scherl E, Neugut AI, Bhagat G, Green PH. Inflammatory bowel disease in patients with celiac disease. Inflamm Bowel Dis. 2005;11:528–32.View ArticlePubMedGoogle Scholar
- Casella G, D'Inca R, Oliva L, Daperno M, Saladino V, Zoli G, et al. Prevalence of celiac disease in inflammatory bowel diseases: An IG-IBD multicentre study. Dig Liver Dis. 2010;42:175–8.View ArticlePubMedGoogle Scholar
- Virta LJ, Kolho KL. The risk of contracting pediatric inflammatory bowel disease in children with celiac disease, epilepsy, juvenile arthritis and type 1 diabetes--a nationwide study. J Crohns Colitis. 2013;7:53–7.View ArticlePubMedGoogle Scholar
- El-Matary W, Senthilselvan A, Fedorak RN, Dieleman L, Spady D. Celiac disease in children with inflammatory bowel disease: a prospective cohort study. Am J Gastroenterol. 2013;108:455–6.View ArticlePubMedGoogle Scholar
- Spencer N, Thanh TM, Louise S. Low income/socio-economic status in early childhood and physical health in later childhood/adolescence: a systematic review. Matern Child Health J. 2013;17:424–31.View ArticlePubMedGoogle Scholar
- Kondrashova A, Mustalahti K, Kaukinen K, Viskari H, Volodicheva V, Haapala AM, et al. Lower economic status and inferior hygienic environment may protect against celiac disease. Ann Med. 2008;40:223–31.View ArticlePubMedGoogle Scholar
- Roberts SE, Williams JG, Meddings D, Davidson R, Goldacre MJ. Perinatal risk factors and coeliac disease in children and young adults: a record linkage study. Aliment Pharmacol Ther. 2009;29:222–31.View ArticlePubMedGoogle Scholar
- Olen O, Bihagen E, Rasmussen F, Ludvigsson JF. Socioeconomic position and education in patients with coeliac disease. Dig Liver Dis. 2012;44:471–6.View ArticlePubMedGoogle Scholar
- Wingren CJ, Bjorck S, Lynch KF, Ohlsson H, Agardh D, Merlo J. Coeliac disease in children: a social epidemiological study in Sweden. Acta Paediatr. 2012;101:185–91.View ArticlePubMedGoogle Scholar
- Whyte LA, Kotecha S, Watkins WJ, Jenkins HR. Coeliac disease is more common in children with high socio-economic status. Acta Paediatr. 2014;103:289–94.View ArticlePubMedGoogle Scholar
- Zingone F, West J, Crooks CJ, Fleming KM, Card TR, Ciacci C, et al. Socioeconomic variation in the incidence of childhood coeliac disease in the UK. Arch Dis Child. 2015;100:466–73.PubMed CentralView ArticlePubMedGoogle Scholar
- Ludvigsson JF, Brandt L, Montgomery SM, Granath F, Ekbom A. Validation study of villous atrophy and small intestinal inflammation in Swedish biopsy registers. BMC Gastroenterol. 2009;9:19.PubMed CentralView ArticlePubMedGoogle Scholar