Outcome in Juvenile Idiopathic Arthritis

A Population-Based Study From Sweden

Elisabet Berthold; Bengt Månsson; Robin Kahn


Arthritis Res Ther. 2019;21(218) 

In This Article


Study Area and Population

The study area was Skåne, the southernmost county of Sweden, containing 33 municipalities with a total area of 11,027 km2.[19] The population of the area was 1,243,329, 13.2% of the total Swedish population, by December 2010. Children (0–15 years) constituted 17.6% of the total population of Skåne with 219,330 individuals (48.6% females) in December 2010. Of these children, 24% were born outside of Sweden or in Sweden with 2 parents without Swedish heritage.[20]

The healthcare system in Sweden is publicly financed. The care is subsidized for all children until at least 18 years of age and includes preventive standardized controls at a child health center and later at school healthcare, which diminishes the risk of missing a symptom of a disease such as JIA. There are no Swedish pediatric hospitals with exclusively private administration. The national diagnosis register is compulsory and independent of care facility. In the study area, there is 1 university hospital, 7 other hospital-associated pediatric outpatient facilities, and 8 private pediatric outpatient facilities. In Skåne, there are also 165 centers of primary care where children can receive healthcare. The center for pediatric rheumatology in Skåne is located at the University Hospital in Lund and receives patients from primary and secondary care, as well as tertiary referrals from the region and the neighboring healthcare regions.

Case Retrieval

The case retrieval process was a two-step procedure ensuring as close to the total coverage as possible. As the first step, a search for patients diagnosed with JIA between 2002 and 2010 was made in the clinical database at the local hospital register using the International Classification of Diseases (ICD) codes M08-M09 (ICD-10). The search was extended to children up to 18 years of age to secure any referred children with the diagnosis made in another healthcare facility. An additional search for total regional coverage was then made from the diagnosis register at the National Board for Health and Welfare (NBHW) using the same ICD codes recorded as primary as well as secondary diagnosis for outpatient and inpatient visits. Thus, to be included in the initial cohort, a patient only needed a single enrollment for inpatient care or one outpatient visit before 19 years of age, with a diagnosis code for JIA registered at any regional healthcare facility between 2002 and 2010.

Medical records for all the patients found were reviewed in order to establish a diagnosis of JIA according to the 2001 ILAR classification.[4] A patient was included in the study if diagnosed before the 16th birthday while living in Skåne between 1 January 2002 and 31 December 2010. At diagnostic uncertainty, we used a two-part consensus for diagnosis, and if that was not achieved, an additional experienced pediatric rheumatologist was consulted and diagnosis was set in consensus or by voting. The Regional Ethical Review Board for southern Sweden approved the study (2013/192 and 2015/62).

Case Ascertainment and Classification

Due to the uncertainty in interpreting information about symptom debut, we stated the diagnosis date before the 16th birthday as the primary inclusion criterion. We defined diagnosis date as the date when a pediatric rheumatologist or a senior pediatrician with experience in the field of rheumatology coded the arthritis as JIA according to ICD-10. However, data from any years, starting from 2002, before the diagnosis date was included in the study if the patient, for example, had been controlled for "suspected JIA" and the diagnosis later was confirmed. The JIA diagnosis was further classified into subgroups according to the ILAR definition. We continuously reevaluated the diagnosis and classification during the entire study period, considering psoriasis, uveitis, and inflammatory bowel disease as a manifestation of the disease. Due to the retrospective data collection, hereditary information could only be used for classification when stated. The information of uveitis was ascertained based on the information stated in the pediatric review or when the medical review from an ophthalmologic hospital-based facility was available. The presence of RF at one occasion was used as an inclusion criterion for patients who otherwise met the criteria for polyarticular disease, as the local clinical guideline is to only test for RF once if it is present in a polyarticular disease. The presence of one positive RF was however not used as an exclusion criterion in patients with a pattern of oligoarticular disease.

Collection and Recording of Data

For all patients, gender, age at diagnosis, disease onset (debut of symptoms), diagnosis date, presence of uveitis, immunological data on the presence of antinuclear antibodies (ANA), RF, anti-cyclic citrullinated peptide antibodies (aCCP), and HLAB27 were recorded. The following parameters were recorded annually: mean values of hemoglobin (Hb), white blood cell count (WBC), platelet count (PLT), ESR, and CRP; mean height and weight; orthopedic surgical procedures; and swollen and tender joint count. The joint count was recorded as the total number of individually affected joints in the 66/68 joint count index that year. Pharmacological treatment was also recorded annually and included even if only used for shorter periods. We did not register adverse events due to the uncertainty of the data since we only had information from the hospital medical records and not from the records in primary care. We did however actively search for cases of tuberculosis.

The patients were followed until the occurrence of death, migration from the study area, loss to follow-up, or end of the study on 31 of December 2015, whichever occurred first.

Statistical Analysis

We used conventional descriptive statistics such as absolute numbers, median, quartile range, and percentage to describe demographics and clinical outcome. The incidence rate was calculated using the number of incident cases as the numerator and the total pediatric population at risk during the study period as the denominator. For the calculation of age-specific incidence rates, the sum of children in each age during the study period was used as the denominator. Survival analysis on chronic uveitis, as well as the need for joint corrective surgery, was performed plotting a Kaplan-Meier survival curve for the whole group with the date of the first diagnosed uveitis/first surgical intervention as a failure point. We used years during the complete study period without arthritis and uveitis as our outcome measure for inactive disease and have presented it as the number of follow-up years without arthritis or uveitis, divided by the sum of follow-up years in the subgroup. We have excluded the first year of disease (year of diagnosis). Statistical analyses were performed using Statistical Package for Social Sciences software (SPSS 25.0 for Macintosh).