The Effects of the BAILAMOS Dance Program on Brain Functional Connectivity of Older Latinos: An Exploratory Study

Abstract Compared to non-Latinos whites, older Latinos are at higher risk of cognitive impairment and engage in less leisure-time physical activity (PA). Resting-state brain functional connectivity (FC) is a putative biomarker for age-related cognitive decline. PA plays a role in FC of brain networks associated with cognitive decline. Objective: Investigate the effects of the BAILAMOS™ dance program on FC in three brain networks associated with age-related cognitive decline (Default Mode [DMN], Frontoparietal [FPN], and Salience [SAL] networks). Methods: Single-group pre-post design. Ten cognitively intact older Latinos participated in the four-month (2x/week for 60min) BAILAMOS™ dance program with four Latin dance styles. MRI was obtained pre- and post-intervention. FC was analyzed using the resting-state fMRI toolbox (CONN) via pairwise BOLD signal correlations and then converted into z-scores. We performed dependent t-tests, computed Cohen’s d and 95%CI for p < 0.05. Results: Within-FPN FC significantly increased (t(9) = 2.35, p = 0.043, d = 0.70) from pre (M=0.49±0.15) to post-intervention (M=0.59±0.13). In the DMN, we observed moderate effect size changes in the ratio of the FC between-networks by the FC within-networks (Mdiff = 0.10; 95%CI = -0.01; 0.21, p = 0.08, d = 0.64). Conclusions: The BAILAMOS™ program increased within-FPN FC, which is a cognitive-control network related to adaptive control and flexibility. Moderate changes between- vs. within-DMN FC suggest BAILAMOS™ also increased whole-brain DMN integration. Taken together, results might signal that Latin dance can combat the disruption of FC between the DMN and other networks, and within-FPN, which are associated with cognitive decline.

and physical limitations. This held true when partialling out genetic variation that could otherwise explain these relationships. More education also predicted later-life PA (ß=.12, p<.0001) separately from family SES and partially mediated the effect of family SES on PA. While school-level resources (e.g., availability of sports and recreation opportunities) did not predict later life PA, they did associate with adolescent sports participation (ß=.26, p=0.007). Overall, later-life physical activity was influenced by earlier life sports participation and education, with family rearing resources being more important than high school resources. As twin pair correlations suggest gender differences, future research will examine whether family or school resources differentially benefit males or females for later-life physical activity.

RACIAL DISPARITIES IN COGNITIVE FUNCTION: THE ROLES OF CUMULATIVE STRESS EXPOSURES ACROSS THE LIFE COURSE
Ruijia Chen, 1 Jennifer Weuve, 2 Laura Kubzansky, 3 and David Williams, 3 1. Harvard University, Cambridge,Massachusetts,United States,2. Boston University,Boston,Massachusetts,United States,3. Harvard T.H. Chan School of Public Health,Boston,Massachusetts,United States Introduction: Racial disparities in cognitive function have been well-documented in the literature, but factors driving the disparities remain under explored. This study aims to quantify the extent to which cumulative stress exposures across the life course explain Black-White disparities in executive function and episodic memory. Method: Data were drawn from the 2004-2006 wave of the Midlife Development in the United States (MIDUS) and the MIDUS refresher study (N=5,967,5,277 White,690 Black). Cumulative stress exposures were assessed by using 10 domains of stressors (e.g., financial stress, childhood adversity). Cognitive function was assessed using the Brief Test of Adult Cognition by Telephone. Marginal structural models were conducted to quantify the proportion of the effect of race/ethnicity status on cognitive function that can be explained by cumulative stress exposures. Result: Blacks reported higher levels of cumulative stress exposures and lower average levels of executive function and episodic memory than Whites. Cumulative stress exposures explained 8.43% of the disparities in executive function and 13.21 % of the disparities in episodic memory. Cumulative stress exposures had stronger effects on racial disparities in cognitive function in the older age group (age≥ 55 years old) than in the younger age group (age < 55 years old). Conclusion: Cumulative stress exposures explain modest proportions of racial disparities in levels of cognitive function. Interventions that focus on reducing stress exposures or improving coping resources among Blacks may help lessen racial disparities in cognitive function at the population level. Chinese older adults are particularly vulnerable to social isolation due to various barriers they face in developing/ maintaining social networks (i.e. limited English proficiency and transportation barriers) in the U.S. However, the prevalence of social isolation and its potential health consequences in this rapidly growing minority aging population remain poorly understood. To address this knowledge gap, the current study examines the prevalence of social isolation, and the relationship between social isolation and cognitive function among U.S. Chinese older adults. Data were obtained from the Population-based Study of Chinese Elderly in Chicago collected between 2011 and 2013 (N=3,157). A four-item index (including living alone, not married, lack of confidant, and low participation in social activities) was constructed to assess social isolation (range: 0 to 4, a score of ≥2 was used to identify individuals who were most isolated). Cognitive function was measured by five validated instruments (range: -2.8 to 2.0). Nearly 22% of the sample were socially isolated. Multivariable linear regression analysis showed that social isolation accounted for 44% of variance in global cognitive functioning. Chinese older adults with greater levels of social isolation had poorer overall cognitive function (B= -0.05, SE=0.01, p=0.001). Study findings highlight the importance of addressing social isolation in cognitive aging among older Chinese Americans. Culturally tailored interventions facilitating the development of supportive social networks/ support have the potential to mitigate cognitive decline in this population. Future longitudinal studies need to elucidate potential mechanisms underlying the relationship between social isolation and cognitive function. Practice implications will be discussed. Compared to non-Latinos whites, older Latinos are at higher risk of cognitive impairment and engage in less leisuretime physical activity (PA). Resting-state brain functional connectivity (FC) is a putative biomarker for age-related cognitive decline. PA plays a role in FC of brain networks associated with cognitive decline. Objective: Investigate the effects of the BAILAMOS ™ dance program on FC in three brain networks associated with age-related cognitive decline (Default Mode [DMN], Frontoparietal [FPN], and Salience [SAL] networks). Methods: Single-group pre-post design. Ten cognitively intact older Latinos participated in the four-month (2x/week for 60min) BAILAMOS ™ dance program with four Latin dance styles. MRI was obtained pre-and post-intervention. FC was analyzed using the resting-state fMRI toolbox (CONN) via pairwise BOLD signal correlations and then converted into z-scores. We performed dependent t-tests, computed Cohen's d and 95%CI

THE EFFECTS OF THE BAILAMOS DANCE PROGRAM ON BRAIN FUNCTIONAL CONNECTIVITY OF OLDER LATINOS: AN EXPLORATORY STUDY
for p < 0.05. Results: Within-FPN FC significantly increased (t(9) = 2.35, p = 0.043, d = 0.70) from pre (M=0.49±0.15) to post-intervention (M=0.59±0.13). In the DMN, we observed moderate effect size changes in the ratio of the FC between-networks by the FC within-networks (Mdiff = 0.10; 95%CI = -0.01; 0.21, p = 0.08, d = 0.64). Conclusions: The BAILAMOS ™ program increased within-FPN FC, which is a cognitive-control network related to adaptive control and flexibility. Moderate changes between-vs. within-DMN FC suggest BAILAMOS™ also increased whole-brain DMN integration. Taken together, results might signal that Latin dance can combat the disruption of FC between the DMN and other networks, and within-FPN, which are associated with cognitive decline.

CHANGING EDUCATIONAL GRADIENTS IN LIFE EXPECTANCY WITH AND WITHOUT DISEASE AMONG U.S. OLDER ADULTS FROM 2000 TO 2010
Matthew Farina, 1 Phillip Cantu, 2 and Mark Hayward 1

University of Texas at Austin, Austin, Texas, United States, 2. The University of Texas Medical Branch at Galveston, Galveston, Texas, United States
Recent research has documented increasing education inequality in life expectancy among U.S. adults; however, much is unknown about other health status changes. The objective of study is to assess how healthy and unhealthy life expectancies, as classified by common chronic diseases, has changed for older adults across education groups. Data come from the Health and Retirement Study and National Vital Statistics. We created prevalence-based life tables using the Sullivan method to assess sex-specific life expectancies for stroke, heart disease, cancer, and arthritis by education group. In general, unhealthy life expectancy increased with each condition across education groups. However, the increases in unhealthy life expectancy varied greatly. While stroke increased by half a year across education groups, life expectancy with diabetes increased by 3 to 4 years. In contrast, the evidence for healthy life expectancy provides mixed results. Across chronic diseases, healthy life expectancy decreased by 1 to 3 years for respondents without a 4-year degree. Conversely, healthy life expectancy increased for the college educated by .5 to 3 years. While previous research shows increases in life expectancy for the most educated, trends in life expectancy with chronic conditions is less positive: not all additional years are in lived in good health. In addition to documenting life expectancy changes across education groups, research assessing health of older adults should consider the changing inequality across a variety of health conditions, which will have broad implications for population aging and policy intervention.

EXPLORING ACCULTURATION EXPERIENCES OF KOREAN IMMIGRANT OLDER ADULTS THROUGH EXPRESSIVE WRITING
Stephanie Rhee, University of Wisconsin-Green Bay, Green Bay, Wisconsin, United States Korean immigrant older adults residing in areas without well-established Korean ethnic enclaves experience acculturative stress and depressive symptoms due to their lingual and cultural barriers. Expressive writing can be used as a culturally sensitive intervention to help those immigrants disclose their deepest thoughts and feelings related to their immigration and acculturation experiences. This study gathered qualitative data from the author's experimental study using expressive writing for first-generation Korean immigrant adults 60 to 88 years of age residing in Midwestern cities. Participants were instructed to write for 15-20 minutes per day in three consecutive days at their convenience in a comfortable and private setting and asked to return their writings by mail. A total of 22 participants returned their writings: 14 wrote about their past and current stressful experiences related to their immigration and acculturation, while eight wrote about their daily lives. Eight themes emerged from thematic coding processes guided by the grounded theory approach: (1) survival, resilience, hardiness, tenacity, pride; (2) lingual barriers; (3) religious faith; (4) gender difference in roles and values; (5) racial discrimination; (6) traditional strategies of acculturation; (7) family and intergenerational gap; and (8) aging. The themes illustrate the participants' lifelong efforts to shape their unique voices through heartwrenching struggles, haunting wounds, and ethnic pride and resilience. The study findings suggest that culturally relevant programs and services are needed to facilitate social relationships and reduce lingual and cultural barriers for Korean immigrant older adults residing in non-ethnic enclaves.  Michigan,Ann Arbor,Michigan,United States,4. Chapman University,Orange,California,United States,5. University of California,San Francisco,California,United States,6. University of Rochester Medical Center,Rochester,New York,United States,7. Columbia University,New York,New York,United States Having more years of education is independently associated with lower mortality, but it is unclear whether other attributes of schooling matter. We examined the association of high school quality and all-cause mortality across race/ ethnicity. In 1960, about 5% of US high schools participated in Project Talent (PT), which collected information about students and their schools. Over 21,000 PT respondents were followed for mortality into their eighth decade of life using the National Death Index. A school quality factor, capturing term length, class size, and teacher qualifications, was used as the main predictor. First, we estimated overall and sex-stratified Cox proportional hazards models with standard errors clustered at the school level, adjusting for age, sex, composite measure of parental socioeconomic status, and 1960 cognitive ability. Second, we added an interaction between school quality and race/ethnicity.