Health-related quality of life in children and adolescents with celiac disease: survey of a population from central Italy

Recruitment took place from January 1 2011 to March 31, 2012.

Participants were Caucasian children and adolescents aged 10–18 years residing in Abruzzo (central Italy). They had a diagnosis of CD confirmed by small bowel biopsy according to the protocol of the European Society of Pediatric Gastroenterology, Hepatology and Nutrition, where small bowel biopsy followed by favorable clinical and serological response to GFD is considered sufficient to confirm the diagnosis [1]. Consecutive patients identified by pediatric gastroenterologists in the hospitals of the four Abruzzo provinces (L’Aquila, Pescara, Teramo and Chieti) were contacted by telephone and offered participation in the study. Those suffering from severe or terminal illnesses that might prevent participation were excluded. Participants were divided into groups of 10–20 and guided in filling out the questionnaire by trained psychologists. Written, informed consent was obtained from those of age (18 years) and from parental figures of minors. Strict anonymity was maintained via de-identification of data. The collected data were stored on a password-protected database.

For data collection we adopted the “Canadian Coeliac Health Survey”—a questionnaire developed by the Canadian Coeliac Association and the Medical University of Ottawa used in the study by Rashid et al. [20]—modified and improved based on the results of a pilot study of a group of teenagers followed at San Salvatore pediatric hospital (L’Aquila). Their data were not included in the analysis.

HRQOL was assessed using the Italian language version of the “Questionnaire on the Health Status SF-12 [21]. This questionnaire is constructed to measure 8 different domains: four refer to the area of physical health (Physical Functioning, Role Limitation-Physical, Bodily Pain, General Health) and four to mental health (Role Limitation-Emotional, Vitality, Mental Health, Social Functioning). Two synthesized indexes are derived from these 8 domains: one related to physical health (PCS12), the other to mental health (MCS12). Both summary scores are standardized to have a mean of 50 and standard deviation of 10, higher scores indicating better health perception.

To illustrate the effect of CD on the HRQOL of adolescents we compared our results with Italian normative values from a study conducted by ISTAT on a sample of 61,434 subjects, 3,221 (5.2%) of whom were aged 14–17 years [21]. Our questionnaire also included a set of specific questions about the wellbeing and lifestyle of children and adolescents with CD and their families regarding personal issues, social activities and potential interventions they felt would improve the QoL of celiac patients.

The self-report questionnaire was divided into two sections. One was administered to children and adolescents with CD, with items regarding social and demographic data, clinical symptoms at presentation, previous misdiagnoses, adherence to diet, and QoL and ways to improve it. The other was administered to pediatric gastroenterologists and solicited data on clinical features, to validate and broaden the information reported by patients. Children aged 10–13 years were helped by parents only in completing the sections related to demographic information and clinical features.

The following parameters were used to estimate sample size: sample error = 0.06, event occurrence proportion 0.5 (in the case of maximum variability), probability 1-alpha = 0.95. This yielded a sample size of 136.1.

Continuous variables, presented as mean ± standard deviation (SD), were compared using Student’s t-test. The Kolmogorov-Smirnov test was performed in advance to check the normality of variables. The χ 2 test was used to estimate the association between the categorical variables under study. Wilcoxon’s test and the Kruskal-Wallis test were applied to interval and ordinal variables. Logistic regression analysis was applied to evaluate whether there was a relationship between the HRQOL reduction and the variables examined. A value of p < 0.05 was considered statistically significant. SAS software, version 9.1.3 (SAS Institute, Cary, NC, USA), was used for the statistical analyses.

For this study, 140 individuals with biopsy-confirmed CD were interviewed. All those who were invited to participate agreed to do so. The mean age of patients was 14.2 years (± SD 2.5); the majority were female (n = 110, 78.6%). Participants were divided into two age groups: children, aged 10–13 years (n = 63, 45.0%) and adolescents, aged 14–18 years (n = 77, 55.0%). Their demographic and clinical characteristics are listed and compared in Table 1.

Mean age at diagnosis was 7.7 years (± SD 4.1) (males: 9.5 ± SD 3.6; females: 7.2 ± SD 4.2) with an average latency between symptom onset and CD diagnosis of 14.2 months (± SD 19.3). Fifty participants (35.7%) had received a previous diagnosis: anaemia and irritable bowel syndrome were the more common diagnoses (both 28.0%). Reported symptoms at presentation are listed in Figure 1.

In our sample, 39 participants (29.3%) were symptomatic at the time of diagnosis, and 94 (70.7%) were asymptomatic (data missing in 7 cases). A large proportion of symptomatic patients reported gastrointestinal symptoms after ingestion of gluten-containing food. Specifically, 26 (74.3%) patients reported diarrhea, 25 (71.4%) reported abdominal pain and 15 (42.9%) reported bloating and gas.

Almost all (97.9%) patients were advised to follow a GFD by their gastroenterologist. Strict adherence was reported by 122 (87.1%). Of these, 101 (73.2%) found a health benefit. Most found adherence to the diet to be “not difficult” (27.9%), or “a little difficult” (41.4%), while only 16 (11.4%) found adherence to be “extremely difficult” and 27 (19.3%) “quite difficult”. Ninety-eight (70.0%) subjects reported using the CD handbook to learn about the presence of gluten in food; 15.7% never used it and 14.3% did so “only sometimes”. Looking at the two age groups, children (aged 10–13 years) reported better compliance compared to older ones (p = 0.02) and less difficulty in adhering to the GFD (p = 0.01). Strict GFD compliance was reported by 95.2% of children, while 80.5% of adolescents did so. Additionally, 39.7% of children found GFD compliance to be “not difficult”, while only 18.2% of participants aged 14–18 years felt the same. No difference in the use of the CD handbook was found between the two groups.

We assessed the QoL of adolescents with CD and the perception of how the disease influenced their social and personal life. Their mean PCS12 was 54.00 ± SD 5.00 and their mean MCS12 was 49.16 ± SD 10.61. The two summary scores were not normally distributed in each group (both p-values < 0.001).

To compare these data with Italian normative values, we looked only at the adolescent age group of the ISTAT study (14–17 years) [15]. No differences were observed in the physical health area (Table 2), whereas in the mental health area our sample showed a lower MCS12 score (p < 0.001). As shown in Table 2 the same difference was observed in both genders (males: p = 0.03; females: p < 0.001).

We estimated the association between several variables and the scores of the sample on the physical and mental synthesized indexes of the SF-12. No differences were observed between genders, age groups, duration of disease categories, age at diagnosis, diagnostic difficulties (previous misdiagnoses and delays in making the diagnosis) or symptomatology at the time of diagnosis (data not shown). Compliance did not influence SF-12 results either. With regard to the perceived degree of difficulty in GFD compliance, those subjects who reported more difficulties had a lower MCS12 score, but the p-value was not significant (p = 0.08, data not shown).

The results of the binary logistic regression analysis are reported in Table 3. The independent variables analyzed did not significantly affect PCS12 or MCS12.

Most of the items selected to assess the general wellbeing and lifestyle of children and adolescents with CD are summarized in Table 4. When looking at the age groups, no differences were observed in avoiding restaurants or in travel. Conversely, 14.3% of children reported bringing gluten-free food while travelling only “some of the time” or “never”, while only 2.6% of the older ones did so (p = 0.01). Difficulty in finding gluten-free food in stores was reported more frequently by adolescents (p < 0.01), as was difficulty in finding high-quality food (p = 0.03). On the other hand, children felt left out of activities more often than their older counterparts (p = 0.03). Gender did not influence most of the items, but girls felt more embarrassed about bringing gluten-free food to parties (p = 0.01, data not shown) and felt left out of activities more often (p = 0.02, data not shown). Division of the sample into symptomatic and asymptomatic patients yielded no difference in most of the items, except that symptomatic children and adolescents felt that they were not invited out for meals because of CD more often than asymptomatic ones (p < 0.01, data not shown), and worried more about staying in hospital because of CD (p < 0.001). Conversely, asymptomatic children and adolescents felt more often that they could be healthy without following a GFD (p = 0.01, data not shown). The analysis of age at diagnosis groups (0–6 years, ≥ 7 years) revealed no differences in the above items (data not shown).

Finally, participants were asked to choose from a list (not reported here) two items that could improve QoL for themselves and their families. The most frequent element chosen was “gluten-free choices in restaurants”, which was selected by 68.8% of participants; “gluten-free choices in supermarkets” was chosen by 36.0% and “informative campaigns in all social contexts” (e.g. schools, restaurants) by 36.6%; 32.9% of patients selected “better labelling”; 22.8% chose “earlier diagnosis”, and only 2.9% chose “better dietetic counseling”.