Maria Vassilaki1, Stephanie Duong1, Chanakya Kodishala1, Rabia Javed1, Michelle Mielke2, John Davis1, Cynthia Crowson3 and Elena Myasoedova1, 1Mayo Clinic, Rochester, MN, 2Wake Forest University School of Medicine, Winston-Salem, NC, 3Mayo Clinic, Eyota, MN
Background/Purpose: Instrumental activities of daily living (iADLs) are complex abilities necessary for independent living involving higher cognitive functions and decline earlier than basic (b) ADLs in the dementia course. The cognitive function and iADLs association is not thoroughly understood; motor and physical limitations could interfere with bADLs and iADLs. Patients with rheumatoid arthritis (RA) are at increased risk for functional limitations. The study examined whether having RA modifies the association between cognition and iADLs.
Methods: The study included 104 RA cases and 312 participants without RA (≥40 years old; mean (SD) age: 75 (10.4) years; 33% male; median follow-up: 3.01 (IQR: 1.2, 6.6) years) of a prospective study of cognitive aging in community-dwelling adults, matched 1:3 for age, sex, education, and cognitive status at cognitive aging study baseline. RA cases satisfied the 1987 ACR classification criteria. iADLs were assessed by the informant-based Functional Activities Questionnaire (FAQ), a 10-item scale (range 0-30, a higher score indicates more functional difficulties). We used Kruskal-Wallis rank-sum and Fisher's exact tests to compare the baseline characteristics between the groups. We used linear mixed-effects models to examine the association between cognitive performance (global cognitive z-score; based on assessment with nine neuropsychological tests) and FAQ score adjusted for age, sex, education, and practice effects. Potential effect modification by RA was examined using an interaction term between cognitive z-score and RA status (yes/no). Estimated coefficients (b) and standard errors (SE) are presented.
Results: Participants with vs. without RA did not differ in age, sex, years of education, or cognitive status, as expected. Informants were less likely to report normal activity for participants with RA vs. without RA in two FAQ items at baseline, i.e., "assemble tax records, business affair, or papers ", (82% vs. 92% normal activity; p=0.024) and "shopping alone for clothes, household necessities, or groceries," (89% vs. 95% normal activity; p=0.028). There was a borderline statistically significant three-way (i.e., cognitive z-score*time*RA) interaction (p=0.07); we observed that the increase in FAQ score (i.e., worse iADLs) per unit decrease in cognitive function might be greater in participants with RA than without RA; i.e., for every 1-unit decrease in global cognitive z-score, the difference in the annualized rate of change from baseline in the FAQ total score was b (SE)=0.22 (0.05) (p< 0.001) in the group with RA, and b (SE)=0.07 (0.03) (p=0.006) in the group without RA.
Conclusion: There is an inverse association between cognitive measurement and FAQ score over time that could be adversely affected by RA status. Further examination will be helpful to improve our understanding of the implications of these findings on the interpretation of the FAQ results in patients with RA and possible cognitive impairment (CI), especially as RA has been associated with CI in previous reports.
Disclosures: M. Vassilaki, F. Hoffmann-La Roche, Biogen, F. Hoffmann-La Roche, Avobis Bio, LLC., Amgen, Abbott Laboratories, Johnson and Johnson, Medtronic, AbbVie; S. Duong, None; C. Kodishala, None; R. Javed, None; M. Mielke, LabCorp, Biogen; J. Davis, Pfizer; C. Crowson, None; E. Myasoedova, None.
