A strict and lifelong commitment to a gluten-free diet (GFD) remains one of the most challenging issues in children with coeliac disease. The present study aimed to record compliance rates and investigate the connection between dietary compliance and demographics, disease-related factors and parental knowledge.
Parents of 90 Greek children diagnosed with coeliac disease were recruited from the outpatient gastroenterology clinic of a children’s hospital in Athens, Greece. Dietary compliance and a range of demographic and clinical data were obtained from parents through a specially constructed questionnaire. Further data included parental perceived and actual knowledge about coeliac disease and GFD. Compliant and noncompliant groups were compared for measured factors and a multivariate approach was followed to elicit independent effects of compliance determinants.
Overall, 44.4% of children with coeliac disease were reported to be compliant to a strict GFD. A 1-year increase in the age of the child was associated with 15% lower odds of adhering to a strict diet after adjusting for other variables (odds ratio (OR)=0.85, 95% CI: 0.75–0.96). Parental perceived knowledge was also independently and significantly associated with dietary compliance (OR=3.3, 95% CI=1.1–9.8). No statistically significant correlation emerged between dietary compliance and other clinical or demographic variables.
Low compliance rates to GFD were observed in children with coeliac disease. Information based on children’s age and perceived parental knowledge can be used to develop risk profiles that health care professionals can utilise to identify children likely to be noncompliant and thus adjust their counselling strategy accordingly.
Coeliac disease is a systemic, immune-mediated disorder triggered by dietary exposure to gluten-containing cereals such as wheat, rye and barley in genetically predisposed individuals.1 Coeliac disease can present at any age and is characterised by the occurrence of a variable combination of clinical symptoms (either typical or atypical), positive serologic testing, abnormal small bowel histology and specific HLA haplotypes.2 According to recent epidemiological studies in European and the US populations, coeliac disease is a common disorder with a prevalence of ∼1%.3 More interestingly, a significant increase in the number of children diagnosed with coeliac disease has been observed in epidemiological studies over the last few years, with more new cases being diagnosed later in childhood and with less typical clinical presentations.4, 5
A strict, lifelong commitment to a gluten-free diet (GFD) still remains the only available treatment for children with coeliac disease. There is sufficient evidence to suggest that compliance with a strict GFD is associated with a decrease in the mortality rates of the disease6, 7 and it can also have a role in preventing some of the long-term complications, including osteoporosis and coeliac-related malignancies.8 Despite its obvious benefits, commencing and following a strict GFD for life remains one of the most challenging and intriguing issues in the paediatric population. This fact is illustrated by the findings of several observational studies, according to which, adherence levels in children with coeliac disease vary considerably between 32 and 81%.9, 10, 11, 12
In an attempt to account for these low levels of dietary compliance, a considerable number of published studies have suggested a connection between adherence to GFD and a range of socioeconomic, demographic and clinical factors, including age of children, parental education, number of siblings, positive family history, society membership and presence of symptoms.9, 12, 13, 14 Moreover, the role of parental knowledge has been investigated in some studies and a connection with dietary compliance of children has been proposed.15, 16 However, the results are quite conflicting and vary considerably between different studies. On top of that, the main weakness associated with much of the literature on dietary compliance is that a univariate approach is followed and the effect of each factor on adherence is investigated separately without taking into consideration the effect of other confounding factors, thus, resulting in clinically and methodologically less meaningful results.
The aim of the current study was to record compliance rates and to further extend current knowledge on the epidemiology of coeliac disease in the paediatric population as well as to investigate how socio-demographic variables, clinical features of the disease and parental knowledge impact independently on children's dietary adherence. To the best of the authors’ knowledge, this is the first study to perform a multivariate analysis and consider the joint effect of potential variables on dietary compliance in children with coeliac disease.
Patients and Methods
The current study took place in the Gastroenterology Unit of the First Department of Paediatrics of University of Athens, ‘Agia Sophia’ Children’s Hospital during a 6-month period from January 2012 to July 2012. ‘Agia Sophia’ is the largest children’s hospital in Athens, and its gastroenterology outpatient clinic is a reference centre for the diagnosis and follow-up of children with coeliac disease from all regions of central and southern Greece. Inclusion criteria for the study were parents of children diagnosed with coeliac disease at least 1-year prior to the start of the study. Children aged from 2 to 18 years old and coeliac disease was defined according to the revised criteria proposed by ESPGHAN.2 Exclusion criteria included parents of non-Greek nationality and the coexistence of other chronic disease in the child (for example, diabetes and Crohn’s disease). This study was approved by the Ethics Committee of the Hospital and informed consent was given by all parents who agreed to participate in the study. Moreover, all parents were provided with contact details and were encouraged to attend the clinic for medical and dietary counselling. Parents of all children who were registered at the gastroenterology outpatient clinic and fulfilled the above criteria were invited to participate. One hundred and sixteen families were contacted and informed by phone and all agreed to participate in the study. Finally, a total number of 90 parents completed and returned the questionnaire by email (n=42), by post (n=22) or during their attendance at the clinic (n=26). The overall response rate was 77.6%.
Materials and methods
All eligible parents were asked to complete a questionnaire consisting of three sections. The questionnaires were available in both print and electronic format and included detailed written instructions. In the first section of the questionnaire, a set of demographic questions and questions about coeliac society membership and health insurance coverage for gluten-free products were asked. The second part of the questionnaire intended to obtain clinical information relating to coeliac disease, including clinical manifestation, age of diagnosis, delay between onset of symptoms and diagnosis, family history and duration of GFD. Clinical manifestation of coeliac disease was classified, according to literature,17 as classical (when at least one of the following was present: diarrhoea, distension, weight loss, anorexia and irritability) or atypical (all other cases). The third part of the questionnaire evaluated paediatric adherence to GFD and parental knowledge. In order to assess dietary adherence, two equivalent items were used, in which parents were asked to evaluate the level of their child’s overall adherence to GFD on a 7-point Likert scale (item (1): ‘Overall, how would you evaluate your child’s adherence to gluten-free diet?’—‘not at all/very strict’, item (2): ‘My child is following a strict gluten-free diet.’—‘strongly disagree/strongly agree’). These items were pilot tested in a sample of 15 parents prior to the start of the study and were shown to be significantly correlated with results from serology testing. Scores obtained from the two items demonstrated high internal consistency (Cronbach’s α =0.86).
In the current study, a dichotomous definition of dietary compliance was used (compliance, noncompliance). A child was classified as compliant only if the maximum score of reported compliance (score 7) was obtained in both the items. In all other cases, the child was classified as noncompliant. This binary definition of compliance has also been used in previous studies.11, 12, 15 In the case of 26 parents who completed the questionnaire during their attendance at the clinic, current results from the child’s serologic testing (serum endomysial (EMA) and antitissue transglutaminase (t-TG) IgA antibodies) were also recorded in order to evaluate the validity of the parental reported compliance. Serologic test was classified as either negative (when both antibodies were negative) or positive (all other cases). When the results of children’s serologic testing were compared with parental reported compliance status in the subsample of 26 children whose parents participated in the study while attending the clinic as part of the follow-up programme, a highly significant correlation was observed (χ2(1)=11.8, P=0.001, Cramer’s V=0.68).
Two dimensions of parental knowledge were measured: perceived and actual knowledge. With regard to perceived knowledge, parents were asked to evaluate their own level of knowledge on GFD and coeliac disease on a 7-point Likert scale, with higher scores indicating higher levels of perceived knowledge. According to their score, parents were further classified into two groups: those with high perceived knowledge (score 7) and those with moderate-to-low perceived knowledge (all other scores). Actual knowledge was assessed with the aid of two tests: a handling menu test and a coeliac disease knowledge test. In the first test, parents were asked to choose the gluten-free products from a given list of 12 commonly found foods and beverages. Selection of foods was made by a panel of experts in coeliac disease and was based on food lists provided by the Greek Coeliac Society. The coeliac disease knowledge test was a shortened version of a questionnaire previously used in the same clinic with established reliability.11 Scores from the coeliac knowledge test were transformed on a 100-point scale. Coeliac knowledge questionnaire was mailed to all participants who responded by email after a 2-week period and an excellent test-retest reliability was obtained for all items of the questionnaire (k=0.71–0.89).
Data were analysed using IBM SPSS software, v.20 (SPSS Inc, Chicago, IL, USA). Continuous variables are presented using the mean and the standard deviation (±s.d.) for normally distributed data or the median and the interquartile range for skewed distributions. Categorical variables are summarised as absolute (n) and/or relative (%) frequencies. Dietary compliance was the main outcome of the study and was analysed as binary variable. For all statistical analyses, a significance level of 5% was used. Associations between categorical variables were tested using the χ2 test. For the associations between continuous variables, the independent sample t-test or the Mann–Whitney U test was utilised depending on assumptions being met. In the logistic multivariate analysis, results were presented as adjusted odds ratio (OR) and 95% confidence interval (95% CI).
In the majority of cases, questionnaires were completed by mothers (85.6%). Children ranged in age from 26 months to 17.4 years (median=12.1 years, interquartile range=8.9–15.1 years), while, a slight predominance of the female gender was observed (65.6%). Overall, 91.1% of families were members of the Greek Coeliac Society and 85% were fully covered by their health insurance for the cost of GF manufactured products. Most children presented with a variety of symptoms (see Table 1) and 22.2% of children were found to present with atypical symptoms before the onset of the disease. Of note, in 10% of the cases, coeliac disease was diagnosed in completely asymptomatic children, while in 14.4%, coeliac disease was present in a first-degree relative. On an average, children were diagnosed at the age of 6.4±4 years. It is noteworthy that in one out of five children, diagnosis was set during the first 2 years of life, while more than half of the children (54.4%) were diagnosed after the age of 5. Delay between onset of symptoms and diagnosis ranged from 1 month to 11 years, with a median value of 10 months. Children followed the GFD for a median time of 4 years (interquartile range=2–7.1).
Most parents believed they were sufficiently informed about their child’s disease. As far as their ability to handle a gluten-free menu is concerned, 87.8% of parents managed to correctly choose all six gluten-free products from the list of 12 products; however, about one-third of parents (32.2%) made at least one mistake by either not choosing a gluten-free item or mistakenly choosing a gluten-containing item as appropriate for their child. Concerning parental knowledge on coeliac disease, most parents (48.9%) made at least one mistake, while 43.4% answered all the questions correctly.
Adherence to GFD and bivariate associations
From the 90 parents who answered the questionnaire, 40 reported that their child followed a very strict GFD, thus, giving a compliance rate of 44.4%.
From all demographic variables, only age of children was found to be statistically significantly correlated with compliance (U=679.5, z=−2.6, P=0.009). In fact, preschool children (2–4 years) demonstrated the highest level of dietary compliance (66.7%). One out of two children aged from 5 to 12 years followed a strict diet, while adolescents (13–18 years) showed the lowest levels of compliance at 34.2%. Additionally, a statistically significant correlation emerged between compliance and coeliac society membership (χ2(1)=3.7, P=0.04, Cramer’s V=0.21). No statistically significant relationship was found between compliance and clinical variables.
When the two dimensions of parental knowledge were compared between the two compliance groups, a statistically significant difference was observed in the perceived knowledge score between parents of compliant and noncompliant children (U=1265, z=2.45, P=0.01). Parents of compliant children demonstrated higher coeliac knowledge score and were more likely to make no mistake in the menu handling test; however, none of the differences were statistically significant. In the case of the menu handling test, a marginal correlation emerged (P=0.07). Details for bivariate comparisons are given in Tables 2 and 3.
A multiple logistic regression analysis was performed in order to examine the extent to which variables identified by the bivariate associations as significant, contribute independently to children’s dietary compliance. Age of children, coeliac society membership and level of perceived parental knowledge were entered as independent variables in the regression model. Results showed that age of child (P=0.01) and parental level of perceived knowledge (P=0.03) were independently and significantly associated with dietary compliance.
In fact, a 1-year increase in the age of the child was associated with 15% lower odds of following a strict GFD after adjusting for other variables (OR=0.85, 95% CI=0.75–0.96). In terms of perceived parental knowledge, for parents who thought they were highly informed about the disease and the gluten-free foods, their child was 3.3 times as likely to follow a strict diet compared with parents with moderate to low perceived knowledge, after controlling for other variables (OR=3.3, 95% CI=1.1–9.8). Finally, no statistically significant independent correlation was found between membership to the Coeliac Society and dietary compliance (OR=6.89, 95% CI=0.66–71.83), after the effect of age and perceived knowledge was taken into consideration. Details of the logistic regression are summarised in Table 4.
The present study showed that 44.4% of children with coeliac disease were reported to be compliant to a strict GFD. This finding is in line with the range of compliance levels given by several European studies9, 10, 13 and confirms the fact that adherence to a strict diet remains one of the most problematic areas among paediatric patients with coeliac disease. Reported compliance in a previous study conducted by the same clinic in 2010 was 58%.11 It is worth mentioning that dietary compliance, as reported by parents, may result in overestimated compliance levels, especially in the case of teenagers. In the current study, although dietary compliance was analysed as a binary variable, it was measured on a 7-point scale, which is believed to be a better and more valid representation of the true compliance status. On top of that, an effort was made to test the validity of the reported compliance by comparing it with an objective measure (serology testing) in a subclass of the sample and the results showed a highly significant correlation.
In terms of the epidemiological characteristics of coeliac disease, the current study found that only one out of five children with coeliac disease was diagnosed during the first 2 years of life. Moreover, in one-tenth of cases, coeliac disease was diagnosed in completely asymptomatic children, while 22.4% of children presented with atypical symptoms. These findings corroborate with results from recent epidemiological studies worldwide2, 4, 17, 18 and seem to support the emergence of a new ‘epidemiological profile’ of the disease according to which coeliac disease is becoming more likely to present later in life and with less obvious gastrointestinal symptoms.
When dietary compliance was examined in different age groups, teenagers were reported to have the lowest compliance to GFD (34.2%). In fact, age was found to demonstrate a significant negative correlation with dietary compliance even after adjusting for the effect of other confounders. This finding is consistent with a number of other studies9, 12, 19, 20 and can be attributed to the fact that teenagers experience higher levels of peer influence and social networking, while at the same time they are more likely to adopt a more risky health behaviour.21, 22
Another finding of the current study was that compliance rates were the same between symptomatic and asymptomatic children. Even though this finding differs from some published studies,12, 14 it is consistent with the results of a previous study from the same clinic.11 Lack of symptoms may give rise to a less severely perceived clinical picture that can easily lead to dietary rules being bent; however, the presence of disease in another family member, which is the case in screening detected asymptomatic children, may increase familiarity with disease in the family environment, thus, compensating for any expected lack of strictness in the diet followed.
With regard to parental knowledge, parents of compliant children were more knowledgeable about coeliac disease and were better able to choose gluten-free products from the menu compared with parents of noncompliant children, although, no significant differences emerged. However, it is worth underlining that one out of three parents made at least one incorrect choice in the menu handling test. This fact demonstrates the need for clinicians and health care professionals to provide better education and more efficient counselling to parents about gluten-free products. In contrast to actual parental knowledge, perceived knowledge of parents was shown to have a significant independent effect on dietary compliance. This interesting result lends support to the findings of a previous study15 and implies that a clinician can have a key role in reassuring and supporting parents about the adequacy of their knowledge and the level of information required to handle the disease.
Although a significant effort was made to limit potential sources of bias, some limitations should be considered, including the use of parental reported measures of compliance and the inability of the correlational study to verify the causal factors that contribute to dietary compliance. Moreover, despite the fact that a quite high response rate was achieved and a wide coverage of geographical regions was obtained, generalising the findings to the whole Greek paediatric population should be done with some caution. Moreover, it is possible that children of parents who did not respond to the questionnaire may have low compliance levels, thus resulting in an overestimation of the true compliance status. Finally, parental reports of compliance were not compared with simultaneous serological findings in the whole sample of children, but only in a small subclass of them.
In conclusion, the current study has shown that reported compliance to GFD is significantly low in children with coeliac disease. Moreover, the problem of low compliance is becoming more intense, especially among adolescents whose parents think that their level of knowledge is not adequate to manage the disease. Age of the paediatric patient and estimation of the level of perceived parental knowledge are two factors that a clinician can utilise as a ‘prognostic tool’ in order to identify children who run a high risk of being noncompliant to GFD and therefore modify their counselling strategy accordingly.
Hill ID, Dirks MH, Liptak GS, Colletti RB, Fasano A, Guandalini S et al. Guideline for the diagnosis and treatment of celiac disease in children: recommendations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr 2005; 40: 1–19.
Husby S, Koletzko S, Korponay-Szabo IR, Mearin ML, Phillips A, Shamir R et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis of coeliac disease. J Pediatr Gastroenterol Nutr 2012; 54: 136–160.
Rewers M . Epidemiology of celiac disease: what are the prevalence, incidence, and progression of celiac disease? Gastroenterology 2005; 128 (4 Suppl 1), S47–S51.
Telega G, Bennet TR, Werlin S . Emerging new clinical patterns in the presentation of celiac disease. Arch Pediatr Adolesc Med 2008; 162: 164–168.
Hoffenberg EJ, MacKenzie T, Barriga KJ, Eisenbarth GS, Bao F, Haas JE et al. A prospective study of the incidence of childhood celiac disease. J Pediatr 2003; 143: 308–314.
Logan RF, Rifkind EA, Turner ID, Ferguson A . Mortality in celiac disease. Gastroenterology 1989; 97: 265–271.
Metzger MH, Heier M, Maki M, Bravi E, Schneider A, Lowel H et al. Mortality excess in individuals with elevated IgA anti-transglutaminase antibodies: the KORA/MONICA Augsburg cohort study 1989–1998. Eur J Epidemiol 2006; 21: 359–365.
Holmes GK, Prior P, Lane MR, Pope D, Allan RN . Malignancy in coeliac disease—effect of a gluten free diet. Gut 1989; 30: 333–338.
Ljungman G, Myrdal U . Compliance in teenagers with coeliac disease—a Swedish follow-up study. Acta Paediatr 1993; 82: 235–238.
Mayer M, Greco L, Troncone R, Auricchio S, Marsh MN . Compliance of adolescents with coeliac disease with a gluten free diet. Gut 1991; 32: 881–885.
Roma E, Roubani A, Kolia E, Panayiotou J, Zellos A, Syriopoulou VP . Dietary compliance and life style of children with coeliac disease. J Hum Nutr Diet 2010; 23: 176–182.
Chauhan JC, Kumar P, Dutta AK, Basu S, Kumar A . Assessment of dietary compliance to gluten free diet and psychosocial problems in Indian children with celiac disease. Indian J Pediatr 2010; 77: 649–654.
Fabiani E, Catassi C, Villari A, Gismondi P, Pierdomenico R, Ratsch IM et al. Dietary compliance in screening-detected coeliac disease adolescents. Acta Paediatr Suppl 1996; 412: 65–67.
Fabiani E, Taccari LM, Rδtsch I-M, Di Giuseppe S, Valentino Coppa G, Catassi C . Compliance with gluten-free diet in adolescents with screening-detected celiac disease: a 5-year follow-up study. J Pediatr 2000; 136: 841–843.
Anson O, Weizman Z, Zeevi N . Celiac disease: parental knowledge and attitudes of dietary compliance. Pediatrics 1990; 85: 98–103.
Jackson PT, Glasgow JF, Thom R . Parent’s understanding of coeliac disease and diet. Arch Dis Child 1985; 60: 672–674.
Roma E, Panayiotou J, Karantana H, Constantinidou C, Siakavellas SI, Krini M et al. Changing pattern in the clinical presentation of pediatric celiac disease: a 30-year study. Digestion 2009; 80: 185–191.
Steele R . Diagnosis and management of coeliac disease in children. Postgrad Med J 2011; 87: 19–25.
Pietzak MM . Follow-up of patients with celiac disease: achieving compliance with treatment. Gastroenterology 2005; 128 (4 Suppl 1), S135–S141.
Mearin ML . Celiac disease among children and adolescents. Curr Probl Pediatr Adolesc Health Care 2007; 37: 86–105.
Cinquetti M, Micelli S, Zoppi G . Adolescents and celiac disease: psychological aspects. Pediatr Med Chir 1997; 19: 397–399.
Kokkonen J, Viitanen A, Simila S . Coping with a coeliac diet after adolescence. Helv Paediatr Acta 1989; 43: 261–265.
The authors declare no conflict of interest.
About this article
Cite this article
Charalampopoulos, D., Panayiotou, J., Chouliaras, G. et al. Determinants of adherence to gluten-free diet in Greek children with coeliac disease: a cross-sectional study. Eur J Clin Nutr 67, 615–619 (2013). https://doi.org/10.1038/ejcn.2013.54
- gluten enteropathy
- gluten-free diet
- parental knowledge
Adherence to gluten-free diet and follow-up of pediatric celiac disease patients, during childhood and after transition to adult care
European Journal of Pediatrics (2021)
Digestive Diseases and Sciences (2019)
Rate and determinants of non-adherence to a gluten-free diet and nutritional status assessment in children and adolescents with celiac disease in a tertiary Brazilian referral center: a cross-sectional and retrospective study
BMC Gastroenterology (2018)