Frailty Worsens Side Effects of Rheumatoid Arthritis Treatment

Patients starting targeted therapies for rheumatoid arthritis (RA) were more likely to develop serious infections and other adverse events if they were considered frail, insurance claims data indicated.

Rates of serious infections were 50% higher among frail RA patients compared with those in more robust condition, and hospital admissions occurred 40% more often, after adjustment for multiple baseline factors, according to Namrata Singh, MD, MSci, of the University of Washington in Seattle, and colleagues.

And it didn’t seem to matter whether patients were taking tumor necrosis factor (TNF) inhibitors, other biologics (such as interleukin blockers), or Janus-associated kinase (JAK) inhibitors, the researchers reported in Arthritis Care & Research.

“This underscores the need for rheumatologists to understand the concept of frailty, best ways to measure it, and help their patients in minimizing the burden of adverse health outcomes,” Singh and colleagues wrote.

In explaining the study’s rationale, the group noted that frailty is more common among RA patients versus people of a similar age in the general population, and not just among the elderly. “In fact, patients with RA had prevalence of frailty at 13% in one study whereas it’s usually 4-11% in geriatric cohorts that are at least 10 years older,” the researchers observed. However, the extent to which frailty may worsen infection risk — an issue with all targeted (as well as older-line) therapies for RA — has not been studied before.

Singh and colleagues drew on the MarketScan databases of private insurance claims, focusing on people with RA diagnoses who began treatment with TNF blockers, other biologic drugs, or JAK inhibitors from 2008 to 2019. They ultimately identified 57,980 such patients with sufficient data, of whom 6% were classified as frail through a validated tool that uses claims data delineating such factors as disability, mobility impairment, recurrent falls, hospital days, and skilled nursing facility days. Individuals with a score of at least 0.2 on this index 12 months prior to starting targeted treatment were considered frail.

Mean patient age was 48.1 (SD 10.1) and about three-quarters were women. Some 83% started a TNF inhibitor, 14% initiated a non-TNF-targeted biologic, and 3% began a JAK inhibitor. Follow-up averaged 1 year for the nonfrail and about 9 months for the frail patients.

Frail patients were only slightly older on average (mean 51.5) than the others. As one might expect, they had more comorbidities — more than half had Charlson scores of 3 or more, compared with just 12% of the nonfrail. On the other hand, the frail patients were generally not in truly terrible condition: their median frailty score was 0.24 with an interquartile range of 0.21-0.25, whereas the scale runs from 0 to 1. The frail patients also received TNF inhibitors at a lower rate (71% vs 84% of the nonfrail), with non-TNF-targeted biologics making up the difference.

Raw data showed that rates of serious infections were more than 4 times as common among the frail — with an eightfold increase in patients starting JAK inhibitors. After adjusting for such factors as comorbidity burden, healthcare utilization prior to starting targeted therapy, age, sex, disease duration, and use of other drugs, the degree of increased risk shrank considerably. But it remained statistically significant, with hazard ratios of 1.5 for TNF inhibitors, 1.2 for non-TNF-targeted biologics, and 3.2 for JAK inhibitors. A similar pattern was seen for incidence of any infection after starting a targeted therapy.

Singh and colleagues also looked at hospital admissions after drug initiation. These, too, were significantly increased with frailty, by factors of 1.2-1.5 depending on which class of drug was started. As with infections, JAK inhibitors associated with the biggest increase in risk.

Limitations to the study included the reliance on administrative data, both for the main analyses and to define frailty. Consequently, “frail” patients’ actual physical performance was not measured. It’s also possible that genuinely frail individuals were not classified as such because they were fortunate enough to avoid falls and healthcare encounters.

Please enable JavaScript to view the

comments powered by Disqus.