Musculoskeletal pain and its effect on daily activity and behaviour in Icelandic children and youths with juvenile idiopathic arthritis: a cross-sectional case-control study

A quantitative case-control study with cross-sectional design was used.

This study took place in Iceland and was carried out as part of a larger study in collaboration with the paediatric rheumatology team at the National University Hospital of Iceland. 

Data collection took place at the National University Hospital of Iceland, from October 2019 to March 2020. Information on subtypes of JIA and age of the group with JIA was obtained from the hospital medical record system, and on the age of the control group from the Registers Iceland. The legal guardians of the children with JIA were contacted by telephone and the study was presented to them. An introduction letter was then emailed to them, and the recipients were asked to reply if their children were willing to participate. If there was no response to the email, a researcher followed up with a phone call and encouraged participation.

The legal guardians of the children in the sample for the control group were texted a request to participate in the study, and they asked to reply if their children were willing to participate. A reminder notice was sent to non-respondents 1 week after the initial message. If no response had been made after several days, a researcher called the parents, encouraging them to participate. Snowball sampling was used in the last stage of data collection to obtain a large enough control group.

The children came to the hospital. The children with JIA underwent a medical examination by a paediatric rheumatologist, including an evaluation of the disease status with JADAS27. The height and weight of all the children were measured, and they answered a survey using the SurveyMonkey online survey tool.

The survey included questions on the child’s sex, pain experience and pain frequency during the last 7 days and location of painful body areas, and three short forms of the Patient-Reported Outcome Measurement Information System (PROMIS) questionnaires. The PROMIS measures were developed by the National Institutes of Health in the USA with the aim of assessing the physical, mental, and social health of individuals with chronic diseases. PROMIS instruments are administered as either fixed-length short forms or computer adaptive tests. Regardless of the administration method, each person’s score is placed on the same PROMIS metric [22].

Three short forms which all include questions on pain during the previous 7 days were used in this study.

PROMIS Paediatric Numeric Rating Scale v1.0 – Pain Intensity 1a includes one question on pain intensity. The children rated their pain during the last 7 days on a scale of 0 (no pain) to 10 (worst pain you can think of) [22].

PROMIS Paediatric Short Form v2.0 – Pain Interference 8a consists of eight questions on the consequences of pain on relevant aspects of one’s life. This includes the extent to which pain hinders engagement with social, cognitive, emotional, physical, and recreational activities [22].

PROMIS Paediatric Short Form v1.0 – Pain Behaviour 8a includes eight questions on behaviours that typically indicate to others that an individual is experiencing pain. These actions or reactions can be verbal or nonverbal, and involuntary or intentional [22].

The psychometric properties of the short PROMIS paediatric questionnaires have been studied in several studies, including children with JIA, demonstrating good content validity and high reliability [23‐26].

The children and parents received verbal and written information and the parents signed an informed consent prior to participating in the study. Children over 12 years also signed the consent. Participation was voluntary and anonymous. Participants were informed about their right to withdraw from the study at any time without any consequences. Ethical approval for the study was granted by The National Bioethics Committee (VSN-19-141).

The Microsoft Excel 2017 program (version 15.37) and the Jamovi statistics program (version 1.1.9.0, https://​www.​jamovi.​org) were used to process the data.

Background information and the results of pain were described by mean and standard deviation for frequencies, and proportions for the categorical variables. The raw values for the Pain Interference and Pain Behaviour questionnaires were calculated and converted into T-scores with an interval scale using a calculator called HealthMeasures Scoring Service (https://​www.​assessmentcenter​.​net/​ac_​scoringservice). PROMIS scores have a mean of 50 and standard deviation (SD) of 10 in a reference population. The reference population is the population to which scores are compared to establish normative values (e.g., general population, clinical population). The reference population for the Pain Interference measure is healthy children in the USA and the reference population for the Pain Behaviour scale is children with chronic pain in the USA [27]. A higher T-score indicates more problems in both questionnaires [28]. There was missing data in the pain interference scale. A question of how hard it was for the children to run when they had pain was deleted from the online survey, by mistake. The HealthMeasures Scoring Service uses “Expected A Priori” pattern response scoring methods. These scoring methods use any available non-missing responses for an individual respondent to predict the most accurate score possible for any scores that are missing [28].

Table 2 shows the results of questions about pain presence among the groups, pain frequency, and number of painful body areas during the last 7 days. Significantly more children with JIA experienced pain compared to the control group X²(1, N = 64) = 5.63, p = 0.02. Half (50%) of the children with JIA had recurrent pain, but only 25% of the control group.

A chi-square test showed a significant difference between the two subgroups who had pain, in the number of painful areas of the body, as children with JIA had pain in more than one part of the body X²(1, N = 26) = 4.47, p = 0.03.

Table 2 shows the results of from the Patient-Reported Outcome Measurement Information System (PROMIS) questionnaires. Since the data were not normally distributed, a Kruskal-Wallis test was used to compare the results of the groups for each PROMIS questionnaire. The children with JIA had significantly greater pain intensity, H(1) = 6.77, p = 0.009, pain behaviour, H(1) = 7.42, p = 0.006 and pain interference, H(1) = 8.47, p = 0.002 than age-matched peers (Table 2).

Pain was experienced by participants in both groups. Of those who reported pain, 93% of the children with JIA and 90% of their peers had recurrent or persistent pain throughout the week. A Kruskal-Wallis test was used to compare pain intensity between the groups. Other variables were normally distributed and calculated by t-test. No significant difference was found between the groups in pain intensity, H(1) = 2.68, p = 0.102 or pain behaviour, t(24) = 1.99, p = 0.058. However, the children with JIA and pain reported significantly more pain interference on daily activities than their healthy peers with pain t(24) = 2.43, p = 0.023 (Table 3).

A closer look at the children who had experienced pain in the past week, showed that although the children with JIA and pain scored higher on the PROMIS scale than the healthy peers with pain (5.25/10 versus 4.1/10), the difference was non-significant. In terms of pain behaviour, the difference between the groups was non-significant, although it did approach significance. It should be noted that the sample is small which increases the risk for Type 2 error, and while the difference was non-significant, t was in the expected direction, i.e., children with JIA and pain displayed more pain behaviour than their healthy peers. Moreover, the children with JIA and pain reported significantly more pain interference than their healthy peers with pain. The authors of this study are not aware of other studies where pain intensity, pain interference and pain behaviour were compared between groups of children with JIA who reported pain and a group of children from the general population who reported pain. Comparison of these two subgroups in this study indicated different impacts of pain between the two groups of children experiencing pain. However, the subgroups were small but our study’s results might indicate the need for further exploration of the impact of pain on the two groups of children experiencing pain.

The study results should be interpreted with several limitations in mind. This was a cross-sectional observational study; therefore, no causal inferences can be made and there may have been unmeasured confounding factors. The data were based on self-reports, which are inherently subjective and rooted in individual personality, outlook, and context. However, subjective measures do provide us with important information about the way the children perceive their pain independent of its cause. While the use of an objective performance-based measurement including a set of tasks would have given the study an added value few such measurements have been developed specifically for children with juvenile arthritis. No measurement tool includes tasks of a wide spectrum of activities, like walking, running, transferring, stair climbing, getting dressed, writing and other fine motor activities.

The study may be prone to non-response bias, in that participants may potentially have more interest in health, JIA, and physical activity than non-participants, as these topics were all within the scope of a large research project that this study was a part of. Due to the small population in Iceland, there is a rather small number of children with JIA, which increases the risk for type 2 error and makes generalisation of the results to other groups difficult. It should be kept in mind, that the sample only included families who lived within a 100 km radius of the capital region where the only Pediatric Rheumatology Outpatient Clinic is situated. This leaves out 10 children with JIA who belong to the clinic but must travel far for control visits and treatment. This could have biased the results in such a small study. In addition, the small sample makes it impossible to look at differences among those with active and inactive disease, age groups or different subtypes of JIA.

The strengths of the study include the use of the three PROMIS scales, which are short and standardised questionnaires with established psychometric properties. The questions are clear and user-friendly. The use of the questionnaires had the benefit of being time- and cost-efficient and easy to administer. All the questions were on pain and the impact of pain during the past 7 days. When children experience episodes of joint inflammation and pain, it is more valuable to ask them about pain experienced over a week, rather than on the 1 day when they respond to the survey. The response rate of the children with JIA was 58%, without reimbursement. There is no consensus on what a good response rate is in a survey but generally responses of 30% or more is considered high [32]. So, while the study’s sample includes few children, it represents a good portion of the population of 8–17-year-old children with JIA in Iceland.

The results give an overview of pain, pain behaviour and pain impact on children with JIA in Iceland. The control group was similar in age, sex, height, weight, and body mass index (BMI), which avoids biases due to differences between the groups. Finally, this is the first study of pain and impact of pain on children with JIA in Iceland.