Health-related quality of life by allergy symptoms in elementary school students

Asthma, rhinitis, and atopic dermatitis in childhood are common chronic diseases worldwide including Korea [1, 2]. Globally, in 2013, in the 6-7 year old group, the prevalence of asthma was 11.7%, of rhinoconjunctivitis 8.5% and of atopic dermatitis 7.9%, although there were wide variations in the prevalence and severity according to the world region [1]. These allergic diseases not only increase the economic burden but also affect the quality of life of children, and their parents, respectively [2‐6].

Interest in self-reported health-related quality of life [HRQOL] as an outcome measure in children and adolescents has grown in recent decades in both hospital and outpatient settings [7‐9]. Consequently, a number of instruments have been developed to assess HRQOL in children and adolescents. Generic instruments can be applied to various population groups, allowing comparison of the relative effects of healthcare interventions and the quality of life between healthy and sick children [10]. Assessment tools for HRQOL for adults may not be appropriate for children, especially if the children’s language and cognitive abilities are not yet well enough developed. In addition, there may be some discrepancies between children’s self-report and proxy reports e.g. by parents [11‐14].

The EuroQoL group developed a child-friendly version of the already established EQ-5D 3 level version [EQ-5D-3 L] [15], the EQ-5D for youth [EQ-5D-Y] [16], to assess HRQOL in children and adolescents [17]. The new instrument is simple and easy to self-administer. Previous studies supported the feasibility of EQ-5D-Y as an instrument with low missing or ambiguous answers [6, 18‐24].

In Jelsma’s study comparing EQ-5D-Y and EQ-5D-3 L in children, the EQ-5D-Y led to fewer missing responses than the EQ-5D-3 L and response ranges in the EQ-5D-Y were broader than that in the EQ-5D-3 L, suggesting that the responsiveness in EQ-5D-Y may be better [19]. The validity [6, 18, 20‐23, 25, 26], reliability [20‐22, 26] and responsiveness [26] of the EQ-5D-Y were demonstrated in populations including children and adolescents with chronic diseases including asthma. The Korean version of the EQ-5D-Y was launched in 2015 by the EuroQol group, and HRQOL data for elementary school students obtained using the EQ-5D-Y instrument in Korea was published firstly in 2017 [24].

The purpose of this study was to explore health-related quality of life of elementary school students with allergy symptoms using the EQ-5D-Y and to examine the validity and feasibility of the EQ-5D-Y.

Of 9949 handed out questionnaires, 9117 (91.6%) came back to the researchers. The median school response rate was 95%, with a minimum of 68.8% and a maximum of 100%.

Demographic characteristics and health information of the students are in Table 1. The age of participants ranged from 7 to 13 years, with an average age of 10.2 years (SD, 1.8). Age differed significantly (P < 0.001) between valid responders (mean age = 10.2) and responders who did not complete EQ-5D-Y descriptive system (mean age = 9.0) (data not shown). The biggest proportion of respondents – some 25% – was in 1st grade, whereas each some 15% were in grades 2 through 6. Of the missing respondents, 125 (63.1%) were in the 1st grade. Over the past 12 months, wheezing or whistling in the chest (asthma) were reported by 259 (2.9%), runny or blocked nose (allergic rhinitis) by 2477 (27.8%), and itchy rash symptoms (atopic dermatitis) by 1801 (20.2%).

The missing proportion of the EQ-5D-Y for all respondents and by actual responders of the EQ-5D-Y are shown in the Table 2. Of the 9117 responses, 198 (2.2%) had missing data for at least one dimension of the EQ-5D-Y, and the missing proportion for each dimension ranged from 155(1.7%) in ‘looking after myself’ dimension to 162 (1.8%) in the ‘feeling worried, sad, or unhappy’ dimension. There were 1258 (13.8%) missing responses on the EQ-VAS. There was no invalid answer on the EQ-5D-Y descriptive system or EQ-VAS. Of the respondents of EQ-5D-Y, 3627 (39.8%) were the students with the help of their parents, followed by 3371 (37.0%) parents and 1443 (15.8%), the students themselves. When students themselves responded, the missing proportion was somewhat lower, but not significantly.

Responses with missing values for the EQ-5D-Y descriptive system (198 cases, 2.2%) were excluded from the analysis. The remaining 8919 cases (97.8%) were used for the final analysis.

The proportions of reported problems by dimension and EQ-VAS score by grade (low vs. high grade) are summarized in Table 3. The proportion of reported problems in the 4th dimension ‘having pain or discomfort’ and in the 5th dimension ‘feeling worried, sad, or unhappy’ was higher than in the remaining first three dimensions. The mean VAS score was 91.2 (SD, 11.1) in grade 1st through 3rd, and 88.2 (SD, 13.0) in grade 4th through 6th. A total of 71 distinct EQ-5D-Y health states were reported. Among EQ-5D-Y health states, ‘11111’ (i.e., full health) states accounted for 82.0%, followed by the ‘11121’ and ‘11112’ health states at 5.0 and 4.8%, respectively (data not shown).

Comparisons of reported problems and VAS scores on the EQ-5D-Y by allergic symptoms are shown in Table 4. There were significant differences in the problem reporting proportion in all EQ-5D-Y dimensions between groups with and without allergic symptoms. There was a particularly large difference in problem reporting rates in ‘having pain or discomfort’ and ‘feeling worried, sad or unhappy’ dimensions compared with the other dimensions of the EQ-5D-Y. Among the allergic symptoms, the group experiencing symptoms of wheezing or whistling in the chest reported significantly more problems than the group without it. The problem reporting rate in the group experiencing wheezing or whistling in the chest in the respective dimensions was 21.1% in ‘having pain or discomfort’ and 18.8% in ‘feeling worried, sad or unhappy’. The average of EQ-VAS score for children experiencing wheezing was 84.5, for runny or blocked nose 88.8, and for itchy rash 88.6, respectively. Students with allergic symptoms had a lower EQ-VAS scores compared to those without. There were similar patterns of problem reporting of the EQ-5D-Y by the diagnosis and treatment of allergic diseases (data not shown).

Finally, the proportions of reported problems and VAS scores on the EQ-5D-Y by the number of allergic symptoms are shown in Table 5. As the number of allergic symptoms increased, the problem reporting rate tended to increase in all dimensions. In particular, the problem reporting rates in having pain or discomfort’ and ‘feeling worried, sad or unhappy’ dimensions increased significantly. As the number of allergic symptoms increased, the EQ-VAS score decreased significantly.

This study compared HRQOL according to allergy symptoms in Korean elementary school students and focused on the importance of school-based allergic disease management. Our study is a large population-based sample study. This study showed that the use of EQ-5D-Y to measure the quality of life of allergic children in elementary school students was feasible and showed construct validity with allergic symptoms.

We compared the quality of life measured with EQ-5D-Y in students with allergic symptoms versus those without. As expected, the quality of life measured with EQ-5D-Y was significantly lower in students with allergic symptoms than in those without, and the quality of life tended to decrease as the number of allergic symptoms increased. Raven-Sieberer et al. [20] indicated that the EQ-5D-Y might not be very useful for discriminating between respondents in the general population. However, our results indicate that EQ-5D-Y can be useful in distinguishing the quality of life of respondents in the general population.

In the EQ-5D-Y descriptive system, we found a high prevalence of problems reported of ‘having pain or discomfort’ and ‘feeling worried, sad or unhappy’ dimensions in all allergic symptoms. Students with wheezing symptoms reported the most problems in all dimensions of EQ-5D-Y, and also the lowest VAS scores. Thus, among the investigated allergic diseases, asthma appears to have a major effect on the HRQOL of children. Trends in the EQ-5D-Y problem reporting rate and EQ-VAS score in this study were similar depending on whether the allergic diseases were diagnosed or not, although not shown in the results. Covaciu’s study reported that allergic diseases, and particularly asthma, significantly affect children’s HRQOL. Frequent wheezing is associated with a high prevalence of problems of ‘pain or discomfort’ and ‘anxiety or depression’ dimensions (26.7 and 23.3%, respectively) [5]. This is similar to our findings, although there is a difference in the quality of life of children assessed by parents using the EQ-5D-3 L. Coviciu et al. reported that allergic comorbidity was associated with the EQ-VAS score [5]. In a Swedish study evaluating the quality of life of patients with asthma with EQ-5D-Y, the problem reporting rates of dimensions of ‘doing usual activities’ was 20%, ‘having pain or discomfort’ was 33.3%, and ‘feeling worried, sad or unhappy’ was 20%, and the VAS average score was 80.7 [6]. Compared with our study, the problem reporting rate is higher in the ‘doing usual activities’ dimension (20% versus 5.9%) [6]. The subjects of the study in Sweden were recruited from the clinics of Stockholm, so the asthma symptom of the subject is likely to be worse than the school-based survey [6]. Our findings are consistent with studies that reported adversely affected quality of life in patients with allergic rhinitis [4] or atopic dermatitis [3, 30].

Of the total respondents, 2.2% did not answer at least one EQ-5D-Y question, and 13.8% did not respond to the EQ-VAS. A multi-national study [20] reported that the proportion of missing responses in the EQ-5D-Y dimensions ranged from 0% in Spain and Italy to 2% South Africa, and non-responders of the EQ-VAS ranged from 0% in Italy to 9% in Germany. Our study is a survey done at the student’s home with written instruction, while the multi-national study was mainly conducted at the school, and in Italy, the researchers provided assistance if needed [20]. Our study was not a face-to-face survey, so we recorded the person who wrote the EQ-5D-Y. More than half were written by students with or without assistance. However, 37% of respondents are parents, and the lower the grade level, the higher the percentage of parents answering was. Jelsma also reported more missing responses in the lower grades [19]. Parents’ and children’s thoughts may be different in evaluating the quality of life of a child [11‐13]. Therefore, the researcher would be wise to assist the elementary school students, especially in the lower grades, to reduce the missing values and improve the completeness of the data.

The strength of our research is that we have covered a large general population. However, there are several limitations of this study. First, our study is a cross-sectional questionnaire survey without clinical examination, so there may be a limitation on the accuracy of diagnosis. However, in previous allergic diseases prevalence studies in elementary school students in Korea, prevalence rates over the past 12 months of symptoms of asthma ranged from 4.8 to 6.5%, allergic rhinitis from 32.9 to 38.5%, and atopic dermatitis from 14.5 to 16.7% [27, 31]. These findings were not much different from our study. Second, our findings include 37% of parents’ responses to EQ-5D-Y. We conducted the same analysis with students only if they wrote their own and with help. The results were not significantly different from those for the whole. In some cases, the p-value slightly increased, but still remained significant. Third, our study could not evaluate the convergent validity, responsiveness and reliability of the EQ-5D-Y due to limited questionnaires and the study design without retest. Further studies are required to examine the other psychometric properties such as test–retest reliability, and responsiveness of the Korean EQ-5D-Y for elementary school students with a wider range of clinical conditions.

In conclusion, the HRQOL was lower in the students with allergic symptoms compared those without, and the HRQOL became lower as the number of symptoms increased. The EQ-5D-Y instrument is acceptable in terms of feasibility and construct validity in assessing the quality of life of elementary school students in Korea. However, it would be beneficial for the researcher to assist the elementary school students, especially in the lower grade, when they fill out the questionnaire to reduce missing values and improve the completeness of the data. EQ-5D-Y would be a useful tool for measuring the quality of life of students with allergic symptoms.