Adherence, Urban Life and Mood Tied to Remission in Rheumatoid Arthritis Cohort

People with rheumatoid arthritis who take their biologic medications as prescribed are far more likely to see their disease go into remission within a year, according to a prospective cohort study conducted in Kyrgyzstan and published in Egyptian Rheumatology and Rehabilitation. Among 135 patients followed for 12 months, 57.8 percent achieved clinical remission, defined as a Disease Activity Score in 28 joints based on C-reactive protein (DAS28-CRP) below 2.6. Baseline medication adherence, measured with the 8-item Morisky Medication Adherence Scale, emerged as the strongest modifiable correlate: each one-point increase in the score was associated with 2.44-fold higher odds of remission (95 percent confidence interval 1.71 to 3.48), after adjustment for disease activity and treatment characteristics.

The study enrolled adults aged 18 or older with moderate-to-severe rheumatoid arthritis who were either starting a new biologic drug or continuing a stable existing regimen. Between January 2024 and January 2026, 144 patients were recruited sequentially during routine outpatient visits; nine were lost to follow-up, leaving 135 with complete data for the primary analysis. The cohort was 84.4 percent female, with a mean age of 47.2 years and a mean disease duration of 9.2 years. At baseline, 44.4 percent were initiating a new biologic and 55.6 percent were continuing therapy. Tumor necrosis factor inhibitors accounted for 71.9 percent of the biologics used, interleukin-6 inhibitors for 18.5 percent, and other agents such as rituximab and abatacept for 9.6 percent.

Adherence at baseline was uneven across the cohort. Fifty-one patients scored in the high adherence range (MMAS-8 of 8 or above), 34 in the medium range (6 to 7), and 50 in the low range (5 or below). Beyond adherence, the researchers collected a broad set of socioeconomic and behavioral variables, including urban or rural residence, education, employment, and monthly income categorized in Kyrgyz som, alongside clinical measures such as the Health Assessment Questionnaire Disability Index, C-reactive protein, rheumatoid factor, and anti-cyclic citrullinated peptide antibody status. Psychological status was assessed with the Patient Health Questionnaire-9 for depressive symptoms and the Perceived Stress Scale-10 for perceived stress. Smoking was self-reported, and physical activity was captured with a categorical measure adapted from the short form of the International Physical Activity Questionnaire.

The secondary outcome was biologic drug survival at 12 months, defined as continued use of any biologic agent without treatment discontinuation, with switches between different biologics allowed. Overall, 83.7 percent of patients demonstrated drug survival: 71.9 percent continued their initial therapy, 11.9 percent switched to an alternative biologic, and 16.3 percent discontinued treatment entirely. Kaplan-Meier analysis revealed a striking gradient by adherence. Drug survival was 98.0 percent in the high adherence group and 94.1 percent in the medium adherence group, but only 62.0 percent among patients with low adherence. The difference between high and low adherers was statistically significant, as was the difference between medium and low adherers, while high and medium groups did not differ significantly, which the authors suggest may reflect a threshold effect.

In the multivariable logistic regression model, which also included baseline DAS28, urban residence was independently associated with remission, conferring 4.14-fold higher odds compared with rural residence (95 percent confidence interval 1.49 to 11.51). Three factors pointed in the opposite direction. Current or former smoking was associated with an 84 percent reduction in the odds of remission relative to never-smokers (odds ratio 0.16, 95 percent confidence interval 0.05 to 0.52). Each one-point increase in baseline DAS28 was associated with a 78 percent reduction in remission odds (odds ratio 0.22), and each one-point increase in the PHQ-9 depression score reduced the odds by 24 percent (odds ratio 0.76, 95 percent confidence interval 0.65 to 0.89). The final five-predictor model showed strong discrimination, with an area under the receiver operating characteristic curve of 0.94 and a Nagelkerke R-squared of 0.700.

Comparisons between the 78 responders and 57 non-responders reinforced these patterns. Responders were more likely to live in urban areas (64.1 percent versus 33.3 percent), be employed (44.9 percent versus 17.5 percent), and report high or middle income (67.9 percent versus 29.8 percent). They also presented with lower baseline disease activity (mean DAS28 of 5.72 versus 6.81), lower functional disability, lower C-reactive protein levels, markedly higher baseline adherence scores (7.15 versus 4.72), lower depression scores (5.82 versus 13.05), and lower perceived stress. Spearman correlation analysis found remission most strongly positively correlated with baseline adherence (rho 0.68) and most strongly negatively correlated with perceived stress (rho -0.62) and depression (rho -0.58).

Depression also tracked closely with treatment persistence. In subgroup analyses of drug survival, 98.7 percent of patients with minimal depressive symptoms (PHQ-9 of 9 or below) remained on therapy at 12 months, compared with 48.0 percent of those with moderate-to-severe depression (PHQ-9 of 15 or above). Urban residents showed superior drug survival compared with rural residents (89.9 percent versus 77.3 percent). Smoking status, by contrast, was not significantly associated with drug survival in unadjusted analysis, which the authors interpret as suggesting that smoking may influence outcomes through other pathways such as adherence or disease activity rather than through discontinuation directly.

The clinical trajectory of the cohort itself was substantial. Mean DAS28 fell from 6.18 at baseline to 3.72 at 12 months, functional disability as measured by the HAQ-DI improved from 1.68 to 0.92, and C-reactive protein declined from 28.4 mg/L to 12.8 mg/L, all statistically significant changes. Interestingly, average adherence declined slightly but significantly over the same period, from an MMAS-8 score of 6.12 at baseline to 5.38 at 12 months. A prespecified sensitivity analysis stratifying patients by baseline treatment status, new initiators versus continuers, produced consistent effect estimates for adherence, depression, and urban residence, though with wider confidence intervals reflecting the smaller subgroup sizes.

The authors frame the work as the first prospective evaluation of socioeconomic, behavioral, and psychological factors associated with biologic treatment outcomes among rheumatoid arthritis patients in Central Asia, a lower-middle-income setting where healthcare access challenges are pronounced. They used a directed acyclic graph constructed a priori to guide covariate selection and explicitly caution that, because the study was observational, the diagram represents a conceptual framework rather than evidence of causal effects. The relationship between disease severity and adherence is likely bidirectional, and with both measured only at baseline, the data cannot establish the temporal direction of that relationship.

Several limitations constrain interpretation. The cohort was overwhelmingly female, a proportion the authors attribute partly to local healthcare-seeking patterns, and they note the findings are most generalizable to female patients. Only 12.6 percent of participants were current smokers, all of them male, so the smoking-remission estimate derives almost entirely from male smokers and carries a wide confidence interval. The adherence effect size is larger than most registry-based estimates, and the authors acknowledge that unmeasured confounding by health literacy or treatment beliefs may have inflated the association. No adjustment for multiple testing was performed in univariable comparisons, and the analysis was complete-case. The authors conclude that these associations warrant confirmation in interventional studies before targeted psychosocial and behavioral screening or support strategies are implemented in resource-limited clinical practice.

Subject of Research: Socioeconomic, behavioral, and psychological factors associated with biologic therapy outcomes in rheumatoid arthritis patients in Kyrgyzstan

Article Title: Socioeconomic and behavioral associations with biologic therapy outcomes in rheumatoid arthritis: a prospective cohort study

Article References: Sakibaev, K., Isakov, U., Zhumabaeva, S., Kulchinova, G., Irisov, A., Taalaibek kyzy, G., Abdimutalib, M., Amegha, N., & Salieva, R. (2026). Socioeconomic and behavioral associations with biologic therapy outcomes in rheumatoid arthritis: a prospective cohort study. Egyptian Rheumatology and Rehabilitation, 53(1), Article 83. https://doi.org/10.1186/s43166-026-00457-2

Image Credits: AI Generated

DOI: 10.1186/s43166-026-00457-2

Keywords: rheumatoid arthritis, biologic therapy, medication adherence, remission, drug survival, prospective cohort, depression, smoking, socioeconomic status, Kyrgyzstan, DAS28-CRP, observational study