Public Health

Social support and mental well-being may differentially provide risk/protection for late-life cognition among different immigrant generations of U.S. older adults.

Kaiser Healthy Aging and Diverse Life Experiences (KHANDLE) Asian and Latino participants were categorized as 1st-generation arriving age <18 (n = 73); 1st-generation arriving age≥18 (n = 282); 2nd-generation (n = 279); or ≥3rd-generation (n = 174) immigrants. The NIH Toolbox Emotion Battery assessed emotional support, instrumental support and loneliness. The PROMIS© Depression Instrument assessed depressive symptoms. All were standardized (mean = 0; SD=1) to the U.S. adult population and dichotomized at 0. Verbal episodic memory (VEM) and executive function (EF) were assessed up to 4 times (max. years=6.6) using the Spanish and English Neuropsychological Assessment Scales (SENAS). Linear mixed-effect models examined associations of social support and mental well-being with EF or VEM. Separate models tested interactions by immigrant generation. All models adjusted for age, sex, education, race/ethnicity, in-person vs. phone interview and SENAS language (Spanish vs. English).

Participants' mean age was 76 years (SD=6.5; Table 1). 1st-generation immigrants arriving at ages <18 had the lowest instrumental support (mean = -0.18, SD=1.0) and highest loneliness (mean = 0.25, SD=0.93) and depressive symptoms (mean = -0.04, SD=0.80). In mixed effects models, instrumental support and emotional support were not associated with baseline EF (instrumental β=0.02 [95% CI=-0.04, 0.07]; emotional β=0.02 [95% CI=-0.03, 0.08]) or VEM (instrumental β=-0.02 [95% CI=-0.08, 0.05]; emotional β=0.01 [95% CI=-0.06, 0.07]; Table 2). Associations of instrumental support, emotional support, depressive symptoms and loneliness with longitudinal EF and VEM approached null. Compared to generational peers with high instrumental support (ref.) 1st-generation immigrants with low instrumental support trended toward lower baseline VEM (marginal estimate at the mean (MEM)=-0.12 [95% CI=-0.29, 0.04]); 2nd-generation immigrants with low instrumental support trended toward higher baseline VEM (MEM=0.18 [95% CI=-0.01, 0.36]); and there was no difference by level of instrumental support among ≥3rd-generation immigrants (MEM=0.004; 95% CI=-0.23, 0.24]; immigrant generation*instrumental support p-value=0.05). Among those with high depression symptoms (ref=low depression symptoms), EF decline was only observed among ≥3rd-generation immigrants (MEM=-0.07 [95% CI=-0.11, -0.04] Figure 1; depressive symptoms*immigrant generation*years p-value=0.04).

Low instrumental support for 1st-generation immigrants and high depressive symptoms for ≥3rd-generation immigrants may be important intervention targets to protect late-life cognition.