These Kids With JIA Fare Poorly for Methotrexate Adherence

Some patients with juvenile idiopathic arthritis (JIA) do worse than others when it comes to taking their methotrexate on schedule, a single-center study indicated.

Among 224 JIA patients treated at Children’s Hospital of Philadelphia (CHOP), younger age, Black race, and living in “low opportunity” neighborhoods were all associated with poorer adherence, according to Dori Abel, MD, MSHP, and colleagues at CHOP.

Whereas more than two-thirds of white patients were considered adherent, based on comprehensive pharmacy dispensing records, two-thirds of those identifying as Black were non-adherent, the researchers reported in Arthritis Care & Research.

As well, only three out of 18 patients living in areas rated as very low on the Child Opportunity Index took their methotrexate mostly on time. And while half of those on public insurance were adherent, that was true for two-thirds of those with private insurance, Abel and colleagues found.

“Given the complexity of factors impacting prescription filling and ultimate medication administration behaviors, a deeper understanding of the barriers and facilitators influencing medication adherence is needed to develop targeted interventions aimed at improving overall adherence and narrowing the disparity in outcomes for children with JIA,” Abel and colleagues wrote.

But they were at least as interested in demonstrating that electronic health records (EHR) could be used to monitor medication adherence, and in that respect the study appeared successful.

“[U]tilization of pharmacy dispense data to assess patients’ adherence to medications offers unique advantages compared to other adherence measurements in that it is an objective, inexpensive tool that allows for large sample sizes and geographic coverage, without relying on self-report,” the group asserted. “We successfully linked aggregated pharmacy dispense data to clinical data available within the EHR, allowing us to investigate associations between adherence, patient characteristics, and disease activity measures.”

For the study, Abel and colleagues pulled CHOP records for all JIA patients with methotrexate prescribed for at least a year during 2016-2023. Their analysis covered those with at least two outpatient visits during the first year; one of those visits had to occur in the first 2 months after the initial methotrexate fill date, and another had to occur during months 10-14.

Dispensing data came from CHOP’s electronic records, which include all pharmacies within the system “and for at least 94% of all other pharmacies,” as claimed by the hospital’s EHR vendor as of 2021. These dispensing records were then used to calculate medication possession ratios (MPRs) — the proportion of days for which patients had methotrexate on hand — for each patient over the initial 1-year treatment period. Patients with MPRs less than 80% were defined as non-adherent.

Overall, among the 224 patients included in the analysis, 36.1% were non-adherent. That was the case for fully half of patients younger than 6, while just over one-third of those age 6-12 were non-adherent. Among those 13 and older, only 23% were non-adherent.

Black patients weren’t the only ones to show lower adherence rates compared with whites. Half of those identifying as Hispanic, and 41% of those classed as “multi-racial/other,” were non-adherent.

In addition to patients’ demographic and socioeconomic parameters, the investigators looked at clinical factors for associations with adherence, and they found a couple. Patients considered adherent had greater changes in affected joint counts relative to non-adherent patients (adjusted difference -0.38, 95% CI -0.74 to -0.01), and this gap was even greater among patients with polyarticular JIA (adjusted difference -1.18, 95% CI -2.23 to -0.13). Nonsignificant trends in the same direction were also seen for percent change in joint count and changes in physician and patient/parent global assessments.

Limitations to the study included its reliance on administrative records and its single-center design. In particular, although Abel’s group appeared satisfied with the pharmacy data, they had to exclude 70 patients out of 294 who were otherwise eligible because the dispensing data were missing or incomplete. Also, about one in six of their patients prescribed methotrexate didn’t have one of the protocol-specified clinic visits during the first year of treatment.

Please enable JavaScript to view the comments powered by Disqus.