- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
- ROLE OF THE STUDY SPONSOR
Juvenile idiopathic arthritis (JIA) is arthritis of unknown etiology that starts before the age of 16 years. It is one of the most common rheumatic diseases in childhood and a major cause of childhood disability. Worldwide, 0.07–4.01 per 1,000 children are affected (1). Children with JIA experience functional impairment due to joint manifestations of the disease, morning stiffness, and fatigue (2). There is no definite cure for JIA; treatment is aimed at controlling pain and achieving inactive disease or remission by means of medication, which might have side effects as well (1).
Physical measures alone are not sufficient to assess the impact of JIA on a child's life. The evaluation of health-related quality of life (HRQOL) is essential for a full assessment of the influence of the disease on a child's life. Quality of life is defined by the World Health Organization as “individuals' perceptions of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns” (3). HRQOL is a concept that incorporates measures of physical symptoms, functional status, and disease impact on psychological and social functioning (4, 5).
Some studies suggest that JIA does not negatively affect HRQOL or psychosocial functioning (6–9), whereas other studies report that children with JIA have a lower HRQOL compared to healthy children (10–16) and compared to children with other chronic health conditions (17, 18). Although a large number of studies have assessed HRQOL in children with JIA, only a few studies focused on predictors of HRQOL in children with JIA. Risk factors identified for impaired HRQOL are polyarticular arthritis or extended oligoarthritis (19), short disease duration (15), pain (12, 20), disabilities, and increased disease severity (21, 22).
HRQOL studies in JIA patients are scarce in Europe (14, 15, 23) and include heterogeneous groups of patients based on different age cohorts and with different national health care systems (19, 24). Many studies use proxy reporting (14, 25, 26), whereas other studies have shown that self-reporting appears to be more reliable (20, 27) for evaluating HRQOL. Different HRQOL questionnaires, both generic and disease specific, have been used for children with JIA. Because of the aforementioned aspects, it is difficult to compare the various studies. Our present study therefore aimed to examine the HRQOL of all children and adolescents with JIA attending one of the pediatric rheumatology centers in Amsterdam, The Netherlands, with the use of self-report (ages 8–18 years) or proxy report (ages 6–7 years). We investigated the HRQOL of children with JIA of a broad range of ages (6–18 years) using generic HRQOL questionnaires. We compared the HRQOL of children with JIA to a healthy Dutch youth norm population and to children with other chronic health conditions. In addition, we assessed the proportion of children with JIA and an impaired HRQOL and the predictors of HRQOL.
Significance & Innovations
Health-related quality of life (HRQOL) of Dutch children with juvenile idiopathic arthritis (JIA) is considerably lower than that of the Dutch norm population of healthy children and children with other chronic health conditions.
Approximately half of children with JIA have an impaired HRQOL.
Four important predictors of impaired HRQOL are physical ability, pain, school absence, and subjective burden of medication use.
To improve care, it is important to systematically monitor HRQOL by the use of patient-reported outcomes.
- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
- ROLE OF THE STUDY SPONSOR
Our study shows significant impairment of HRQOL in patients with JIA in nearly all domains, almost independent of pediatrician-reported disease activity status or disease duration. Four factors appear to be strongly related to impaired HRQOL in all patients: functional ability (C-HAQ score), patient-reported pain, school absence, and the subjective burden of medication use.
Both the child age group (6–7 years) and the adolescent age group (13–18 years) reported lower HRQOL compared to their healthy peers; HRQOL scores were equal compared to peers with other chronic health conditions. The 8–12 years age group reported lower HRQOL scores compared to their healthy peers, as well as to the children with other chronic health conditions. In all age groups the largest effects were found for physical and psychosocial functioning. These findings are consistent with previous JIA studies (12, 15, 16, 21). Emotional functioning, however, seems to be less affected than other domains in all age groups.
We tried to gain more insight into HRQOL outcomes by using predictive statistical models, but with the use of linear multivariate regression analyses causality cannot be proven. In other studies, physical disability also appears to be strongly related to HRQOL (14, 16, 22, 39). Besides, physical activity is generally impaired in adolescents with JIA (40). Individualized training programs may significantly improve the physical ability of the adolescent (41) and therefore positively influence a child's HRQOL.
As well as physical disability, pain is also a very important predictor for HRQOL as compared to other studies (12, 14, 16, 22). The relief of pain and pain management are therefore important factors to deal with in the care of children with JIA (42–45). The physician-reported VAS scores of disease activity appear to be less directly related to HRQOL, whereas functional ability as measured by the C-HAQ and VAS pain, all reported by the patient, are obviously strongly related (46).
In our study, another important predictor of impaired HRQOL appeared to be school absence. Children who reported more school absence in the 3 months prior to the consultation showed lower HRQOL and more impaired school functioning than those children with less school absence. Therefore, parents and teachers should be informed about the consequences of school absence to HRQOL for a child with JIA, not only in regard to school functioning but also for the other domains of HRQOL.
The reported subjective burden of medication use is also important in explaining HRQOL. Although receiving medication is not a predictor of impaired HRQOL, when children report the experience of a burden from their medication use, they are at risk for impaired HRQOL. It is important to educate patients and their parents about the (side) effects of medication. This issue should be addressed during consultation, and in complex cases support should be provided, e.g., with a psychoeducational program (47, 48).
When the insights into the medical and nonmedical predictors of HRQOL are increased, health care workers will be enabled to optimize their treatment strategy to improve the effect of treatment on HRQOL (22). The influence of disability and pain on HRQOL has been described in other studies (14) and was confirmed in our findings. From our study, we can add 2 nonmedical predictors to the explanation of HRQOL in children with JIA: school absence and subjective burden of medication use.
Our study shows that approximately half of the children with JIA have an impaired HRQOL as compared to 16% of the general population. This more clinical way of looking at the data is new in the JIA population and provides more insight into the number of children with JIA with a low HRQOL. Based on the results of our study, most patients with JIA are at risk of an impaired HRQOL. Systematic monitoring of HRQOL in children with JIA is therefore warranted (4, 22). By using PROs such as HRQOL questionnaires in daily clinical practice care, pediatricians, psychologists, and nurses can adequately identify the specific problems in the HRQOL domains of children with JIA. Consequently, more tailored advice and referrals can be given to these children. Children with JIA can be provided with HRQOL questionnaires before a consultation. Systematically reporting PROs to the pediatric rheumatologist during the consultation can facilitate communication about HRQOL. The PedsQL is a suitable instrument for systematic HRQOL assessment in daily clinical practice, as this is a questionnaire that is simple, easy to administer, and requires only 5–10 minutes to complete (28, 29). The PROs can easily be expanded with the information about functional ability, pain, the subjective burden of medication, and information on school absence.
Our study has some limitations. First, differences in age, parental country of birth, and the patient's BMI were found between participants and nonparticipants. This is not surprising as older children and adolescents are, in general, less willing to participate in clinical studies, and parents decide whether the younger children should participate. The percentage of parents born in The Netherlands was also significantly higher among the participants. Nevertheless, 20% of the participating parents were born outside The Netherlands, which reflects the population of the Amsterdam region. Both parental country of birth and BMI were not significant factors in contributing to HRQOL among the participating children with JIA. We therefore assumed that the differences in these factors between participants and nonparticipants did not influence our findings.
In our study, data on HRQOL were compared with a Dutch norm population containing a sample of healthy children and children with a chronic health condition (36). These Dutch norm data for the norm population were collected in previous research in children and their parents from Amsterdam and the surrounding regions at 4 elementary schools (3 suburban and 1 urban), 4 high schools (1 rural, 1 suburban, and 2 urban), and 1 school for vocational education (urban) and are considered to be representative for The Netherlands (36).
Due to the lack of data, we could not fully compare the socioeconomic status of the participants and nonparticipants. We could only compare parental country of birth. For the participants, the parental country of birth was most often The Netherlands. In the comparison between the Dutch norm data and our sample, a difference in education was found. An influence of these aspects of socioeconomic status on the HRQOL of the children in our sample cannot be ruled out.
Regarding the comparison of HRQOL of our sample with children with a chronic health condition from the Dutch norm population, some aspects need to be taken into account. Children with a chronic health condition were identified in the norm population based on proxy report. It is plausible that the more severely ill children were not included in the norm population, since the sample for the norm population was collected from regular schools. In addition, the sample sizes of children with chronic health condition are relatively small. The differences in HRQOL between children with JIA and children with other chronic health conditions will be investigated further.
In the univariate analyses, the VAS general well-being and the number of active joints were moderately to highly correlated with HRQOL, but we had to reject these correlations for inclusion in the regression model because of high multicolinearity. Based on previous studies, VAS general well-being and the number of active joints are important factors for a child's HRQOL.
Unfortunately, we were unable to report on the influence of fatigue and sleep problems on HRQOL. For practical reasons, we did not add questionnaires for the evaluation of fatigue and sleep problems to our web-based survey. Recent studies have demonstrated that children with JIA report more fatigue and sleep problems compared to healthy children and that these factors appear to correlate highly with HRQOL (16, 39, 46).
A strong point of our study is that we focused not only on outcomes of HRQOL, but also on the determinants. These determinants can be taken into account during consultations by the health care workers. If indicated, both children and parents can be educated and supported during the developmental trajectory. We used a web site for data collection. Having the participants use the internet to complete the PedsQL questionnaire is valid, reliable, less time consuming, and more efficient compared to a paper and pencil version. Lastly, we used self-reports from the age of 8 years; at this age it is reasonable to assume that children are capable of reporting their own HRQOL (49). Moreover, self-reporting appears to be more reliable as compared to proxy report (20, 27) for evaluating HRQOL.
The best way to evaluate HRQOL is to use a standardized generic instrument with a disease-specific supplement (50). Depending on the availability of validated questionnaires, disease-specific rheumatology questionnaires can be added. It would be interesting, if possible, to add the PedsQL 3.0 rheumatology module (13) to the PedsQL 4.0 generic core scales to evaluate HRQOL.
In the near future, HRQOL studies in children with JIA should be extended so as to investigate the more predictive and resilient factors that have been identified in pediatric psychology research, such as self-management and family and social circumstances. In this study we already found a substantial percentage of explained variance in HRQOL, but there appears to be other important factors that are predictors of HRQOL. By identifying all predictors of HRQOL, the quality of care for children and adolescents with JIA can increase further.
In conclusion, from the data on 152 children with JIA, age ≥6 years, and from consulting with pediatric rheumatologists in Amsterdam, we concluded that the HRQOL of children with JIA is considerably lower than that of the Dutch norm population, as well as compared to children with another chronic health condition. The 4 important predictors of these problems are impaired physical ability, patient-reported pain, school absence, and subjective burden of medication use. To improve patient care in children with JIA, we believe that it is important to systematically monitor HRQOL by the use of PROs (28). Based on our findings, we suggest discussing subjective burden of medication use and school absence during consultations. More risk and resilient factors influencing HRQOL in patients with JIA need to be investigated in future studies.