Parental Perception of Dietary Intervention in Juvenile Idiopathic Arthritis

Introduction

Juvenile idiopathic arthritis (JIA) is the most common rheumatic disease in children. ^1,2 Clinical outcomes are variable among different JIA categories and treatment strategies often differ. The etiology of this chronic rheumatic condition remains unclear. Treatment of JIA has benefited greatly from the development of biological treatments, early diagnosis, and aggressive intervention. ³ However, there is still significant interest from families to explore dietary changes in hopes of obtaining better control of this disease with less use of medications. ^4,5 In one study (n = 52), 35% of patients with JIA reported removing foods from their diets to control joint symptoms and had reported favorable responses. ⁵ Although there have been successful dietary interventions in other pediatric inflammatory diseases, such as the use of a specific carbohydrate diet in inflammatory bowel disease (IBD), ^6,7 no controlled dietary studies have been performed in JIA.

Our study explored the prevalence of dietary interventions in a cross-sectional group of patients diagnosed with JIA and parental report of the responses of joint symptoms.

Materials and Methods

This research project was approved by the Seattle Children's Hospital Institution Review Board (No. 15963). Consent was waived, because no protected health information was collected. The survey included questions regarding demographics, disease course, use of conventional medications, and/or special diets and perceived outcomes of dietary interventions. Commonly used diet types are listed. The survey was constructed in REDCap, a secure, web-based data capture application. ⁸ The survey link was posted to the “Juvenile Idiopathic Arthritis” Facebook group (1800 members) in March 2016 and distributed by the Arthritis Foundation through email (16,000 recipients) on April 20, 2016. Reminders were sent periodically during the study period. Letters, including the survey link, were also sent to 1414 patients with JIA followed in Seattle Children's Hospital in March 2017. The targeted study population was parents of children with JIA diagnosed by rheumatologists. Patients with celiac disease or IBD were excluded. The survey was closed on May 4, 2017. Responses were summarized using descriptive statistics. Categorical variables were summarized using frequencies and proportions, while for time after diagnosis, we calculated median and interquartile range. We calculated 95% confidence intervals for symptom response proportions. Independence of categorical variables was evaluated using chi-square tests. Statistical analyses were performed using Stata version 14 (StataCorp, College Station, TX).

Results

A total of 338 responses were received (1.8% response rate) from all sources with 261 responses completed by parents of children with JIA. Other response was from adult patients and excluded. We also excluded 10 patients not diagnosed by a rheumatologist and 13 with IBD or celiac disease. Demographic and JIA patient characteristics are summarized in Table 1. The majority of patients residing in the United States (97%) were female (73%) and had a polyarticular disease (63%). Eighty-three percent of respondents endorsed the use of a disease-modifying antirheumatic drug (DMARD) and/or a biologic drug. A subset of respondents (n = 51) was queried about postdiagnostic duration of JIA as a question was added during the study period. Median time since diagnosis was 2 years (interquartile range 1–5 years).

Parental perception of response to special diet

Among the three most common special diets, the percentage reporting no change in or worsening joint symptoms ranged from 35% for lactose-free to 48% for gluten-free diets and 50% for anti-inflammatory, whereas improved pain or joint swelling was reported by 50% of anti-inflammatory, 52% of gluten-free and 65% of lactose-free respondents (Fig. 1A). Response rates were further divided into joint involvement by diet (Fig. 1B). For each of the three most common diets (gluten-free, anti-inflammatory, lactose-free), the proportion of children experiencing improved symptoms was higher for those children with <5 joints involved (p < 0.05).

OPEN IN VIEWER

FIG. 1.

Percentage of children with juvenile idiopathic arthritis who had a favorable response to diet per parental report (A) and by number of joints affected (B). There was no statistical significance. CI, confidence interval.

As a proxy for disease activity, we specifically asked, “Did your medication change while on a special diet.” Most parents reported no change in medication during a dietary trial. When looking at medication changes while on a special diet, we found that within the patient group that adopted a gluten-free diet, there was a trend toward escalating antirheumatic medication (for example, from nonsteroidal anti-inflammatory drugs to a DMARD or a biologic). In those patients, 19% escalated medication, 8% reduced use of antirheumatic medication, 73% reported no change, and 6% were unable to recall. An opposite trend was seen with the lactose-free diet, where 20% of patients reported a reduced use of antirheumatic medication, 70% reported no change, 10% could not recall, with none reporting an escalation in antirheumatic therapy. The responses from those on the anti-inflammatory diet were missing.

Discussion

We have completed a survey study to capture parental perception of dietary effect on the joint health of their children with JIA. One third of our parents had already tried a special diet to help their child's disease, and 43% of them did so within the first 6 months of diagnosis, despite a paucity of data supporting dietary therapy. These results underscore the desire of families to seek helpful information on the role of diet in JIA management. There is a need for more research on diets and JIA that will allow rheumatologists to help patients and families make informed dietary decisions.

The most popularly adopted diets included a gluten-free diet, an anti-inflammatory diet, and a lactose free diet. Due to the limited number of responses, comparison among these diets did not appear to be statistically significant although there was a trend toward a greater reported reduction in antirheumatic medications in children on a lactose-free diet. The standardized dietary intervention and randomized control study will be necessary to minimize confounding factors and placebo effects. An objective measurement of joint outcomes needs to be used for a future controlled study to confirm these results.

For each of the most common diets, those with fewer joints involved were more likely to experience improvement in symptoms. Several factors may account for this result. First, those with fewer joints involved may respond better to a special diet. Second, when patients had fewer joints involved, they were more likely to report favorable outcome because of a greater tendency to improve or experience a placebo effect.

Based on the reported rates of parent initiation of special diets in our study, a prospective dietary study would be feasible in JIA. However, factors such as growth or picky eating could impact parents' willingness to make changes in their children's diets. Indeed, a review of dietary interventions in rheumatoid arthritis found weight loss to be a common side effect of dietary interventions meant to treat arthritis symptoms. ⁹ Therefore, close weight and growth monitoring and corresponding corrective measures should be considered in the design of such studies.

There is a paucity of research on how diet impacts the disease status of JIA. While our sample size is limited, our study demonstrates parental interest in diet in JIA. Our respondents had a high prevalence of DMARDs and biologic medication use during the disease course. For that reason, we conclude that the respondents are parents of children with higher disease burden, and thus they are more likely motivated to seek alternative treatment to their children's arthritis. But the parents of children with fewer joint counts were more likely to report improvement with dietary changes.

This study had several limitations making our results exploratory and requiring further validation. First, generalizability is limited because of geographic variations in the response to questions as well as the predominant responses from parents of children with polyarticular disease and a disproportionally overrepresented white population. Second, the response rate was low and families with greater interest in this topic may have been more likely to participate, which may have led to a selection bias. Third, the sample sizes of the entire group and subgroups were small, which reduced the power of statistical analysis. Fourth, questionnaires may lead to recall bias. Lastly, definition of a diet, such as “anti-inflammatory” or “gluten free” may also vary from patient to patient, and the adherence to special diets was not measured, nor was the duration on each diet.

Conclusions

One third of survey respondents from families of children with JIA attempted special diets, and 50% reported improvement in symptoms. Only 26% of survey respondents were willing to participate in a dietary intervention study.