Cody Schmidt, Kunal Patel, Samantha Delgado, Guy Cozzi, Elisea Avalos-Reyes, W. Cliff Rutter, Rashmi Grover, Lucia Feczko, Will Cavers and Kjel Johnson, CVS Health, Woonsocket, RI
Background/Purpose: Rheumatoid arthritis (RA) symptoms are suspected to be linked with exposure to colder environments. Previous studies on RA and seasonality have focused on RA activity; however, none assess medication adherence. The objective of this study was to assess the impact of seasonal changes on medication adherence among patients beginning subcutaneous or oral RA treatment.
Methods: This is a retrospective cohort study of adult RA patients throughout the United States enrolled in a large health plan taking subcutaneous or oral RA medications between 3/1/2019 and 2/28/2021. Patients were included if they were newly diagnosed with RA (ICD-10 codes M05.X and M06.X). Medication adherence for these patients was tracked for 12 months after starting therapy. Medication classes included: TNF-αinhibitors, Janus Kinase inhibitors (JAK), Interleukin-6 (IL-6) inhibitors, a T-cell blocker, and folate analogs. Medication history was stratified into meteorological seasons for the northern hemisphere: spring (3/1-5/31), summer (6/1-8/31), fall (9/1-11/30), and winter (12/1-2/28). Monthly medication adherence was measured using proportion of days covered (PDC) between the first and last fill of the year. The effect of seasonality on monthly adherence was estimated using linear regression, controlling for age, gender, and other patient demographics. Predicted PDC was calculated using resultant equations; p-values < 0.05 were significant.
Results: In total, 3,710 patients were included in this study; adherence was captured for 39,628 member-months. The average age of the cohort was 56 years (standard deviation (sd) = 14); 2,797 (75%) identified as female. The average monthly PDC was 0.84. Predicted monthly PDC was highest in summer (0.85), followed by spring (0.85), fall (0.84) and lowest in the winter (0.83). Significant differences in PDC by season were found in winter (p < 0.01). Seasonality effects differed by drug class. For TNF-α inhibitors, spring (p < 0.01) and winter (p < 0.01) were associated with lower mean PDCs. None of the season coefficients were statistically significant for JAK inhibitors. Adherence for IL-6 inhibitors (p=0.04), T-Cell blockers (p < 0.01) and folate analog (p=0.01) were the most sensitive to winter seasonal changes. Subsampling by climate region revealed the negative impact of winter on medication adherence was limited to members residing in humid (p=0.00) or cold climates (p=0.01).
Conclusion: Medication adherence decreased in winter months, but the effect of seasonality depended on class and was drug specific. Seasonality influences adherence, but only for specific seasons.
C. Schmidt: CVS Health, 3, 11; K. Patel: CVS Health, 3, 11; S. Delgado: CVS Health, 3, 11; G. Cozzi: CVS Health, 3, 11; E. Avalos-Reyes: AstraZeneca, 11, CVS Health, 3, 11, GlaxoSmithKlein(GSK), 11, Haleon, 11, Johnson & Johnson, 11, Moderna, 11, Novavax, 11, Pfizer, 11, Viatris, 11; W. Rutter: CVS Health, 3, 11; R. Grover: CVS Health, 3, 11; L. Feczko: Baxter, 11, CVS Health, 3, 11; W. Cavers: Amedisys Inc, 11, Baxter, 11, Conmed Corp, 11, CVS Health, 3, 11; K. Johnson: CVS Health, 3, 4, 11, HC Technology Patent, 10.