The Association Between Obesity and Long-Term Cognitive Performance in Middle-Aged and Older Adults

Abstract Obesity is a growing epidemic in the United States and has been associated with negative health outcomes such as cardiovascular disease and diabetes. However, an obesity paradox has emerged which suggests that the effects of obesity may vary by age, with older adults potentially seeing a protective effect of obesity. This study examined the effects of overweight and obese status on cognitive performance at baseline and follow-up. It was hypothesized that obese middle-aged adults would perform worse than normal weight peers, but that reverse would be observed in older adults. Data from 701 participants in the Midlife in the United States study were included. Body mass index (BMI) and waist circumference were employed as measures of obesity. Z-scores for executive function, memory, and global cognition were used to quantify cognitive performance. While obese participants tended to perform worse on average than normal weight individuals there were no significant differences in performance between obese and normal weight participants in global cognition (p=.134), executive function (p=.164), or episodic memory (p=.708). Additionally, age did not moderate this relationship. However, there was a significant effect of education on all three domains. When stratified by education, participants with some college or higher, had a significant time*obesity*age interaction (F[3,328]=3.016, p<.05). For the oldest-old participants, executive function scores were higher for obese participants at follow-up compared to normal weight participants, but not at baseline. These findings suggest that level of education may serve as a form of cognitive reserve which compensates for deficits due to obesity.

investigate the association among preoperative patients' expectations, post-operative functional abilities, and satisfaction to functional abilities among older adults undergoing TKA at 6-week after surgery. Methods Participants were 97 older adults who purposely selected based on the inclusion criteria. The data were collected at preoperative and postoperative TKA by using the Hospital for Special Surgery Knee Replacement Expectations Survey and the Knee and Osteoarthritis Outcome Score -function in daily living. The data analysis was performed by using descriptive statistics, paired t-test, and Pearson product moment correlation coefficient. Results Before surgery, patients' expectations to postoperative functional abilities had a high level with the total mean score of 70.21(SD =13.86). At 6-week after surgery, the overall functional ability had a significant improvement (t = -9.229, p = .000). Satisfaction to functional ability also had a high level (Mean ± SD = 71.15 ± 14.73). Patients' expectations to functional abilities had a significantly low positive correlation to postoperative functional ability and satisfaction (r = .273, p < .05; r = .292, p < .01, respectively). A significant moderate positive correlation between functional abilities and satisfaction to functional abilities was observed (r = .603, p < .01). Conclusion Understanding of expectations may be beneficial in gaining knowledge, paving expectations on possible outcomes, and enhancing the quality of care for these populations.

SOCIAL DETERMINANTS OF UNMET HEALTHCARE NEEDS: COMPARISON BETWEEN MIDDLE-AGED AND OLDER PEOPLE Sungjae Hong, Emory University, Decatur, Georgia, United States
This study aims to 1) examine unmet healthcare needs by age groups and 2) compare the social determinants of unmet healthcare needs between older and middle-aged people in Korea. This study employed a nationally representative dataset of the 2017 Healthcare Service Experience Survey in Korea. Unmet healthcare needs consisted of three categories by healthcare type: 1) clinic visitation, 2) treatment, and 3) medication. Independent variables include demographic, socioeconomic, and health status. Logistic regression models were estimated to reveal the social determinants of unmet healthcare needs of older (age≥65; N=2,178) and middleaged (age 40~64; N=5,062) people. There was a positive gradient of unmet healthcare needs prevalence by age group, having the highest prevalence among older people (10.8%). While older people living alone were 1.70 times more likely to report any of unmet healthcare needs, there was no significant relationship between the two among middle-aged people. In addition, the effect of chronic disease morbidity on the probability of unmet healthcare needs was stronger among older people than it is among middle-aged people (OR=3.50 and 2.90, respectively). In contrast, the effect of household income was weaker than it is among middle-aged people (OR=1.73 and 2.95, respectively). The gradient of unmet healthcare needs by age group asks gerontologists and public healthcare scholars to focus on older people regarding unmet healthcare needs. Also, the difference between middleaged and older people on social determinants of unmet healthcare needs implies different psychosocial pathways of unmet healthcare needs between the two age groups. The aim of the current study is to provide comprehensive health care to older adults by assessing physical and mental health in a geriatric primary care setting, including evaluation of both subjective and functional health literacy. Health literacy is vital to understanding medical information and making subsequent decisions based on this information. Knowledge of patient health literacy may be particularly important for care providers, as it can provide guidance on how to best communicate with the patient (Nouri & Rudd, 2015). It may be particularly important to monitor health literacy within older adults, as several studies (e.g., Kobayashi et al., 2015) have shown that health literacy decreases with mild cognitive impairment. Approximately 250 patients (mean age = 76; 74% female; 16% African American) attending an interdisciplinary geriatrics clinic in West Alabama have been recruited to take part in a variety of behavioral health screenings. The current study assessed subjective health literacy using questions developed by Chew, Bradley, and Boyko (2004) and functional health literacy using the Newest Vital Sign (Weiss et al., 2005). While there was a significant correlation between subjective and functional health literacy (r = .43, p < .001), 81% of patients reported adequate subjective health literacy, while only 41% demonstrated adequate health literacy on a functional screening measure. Based on these findings, self-reported health literacy may not necessarily be reflective of performance on more functional measures. Given the potential consequences of overestimating health literacy, this represents a serious barrier to patient care.

THE ASSOCIATION BETWEEN OBESITY AND LONG-TERM COGNITIVE PERFORMANCE IN MIDDLE-AGED AND OLDER ADULTS Andrew Fiscella, and Ross Andel, University of South Florida, Tampa, Florida, United States
Obesity is a growing epidemic in the United States and has been associated with negative health outcomes such as cardiovascular disease and diabetes. However, an obesity paradox has emerged which suggests that the effects of obesity may vary by age, with older adults potentially seeing a protective effect of obesity. This study examined the effects of overweight and obese status on cognitive performance at baseline and follow-up. It was hypothesized that obese middle-aged adults would perform worse than normal weight peers, but that reverse would be observed in older adults. Data from 701 participants in the Midlife in the United States study were included. Body mass index (BMI) and waist circumference were employed as measures of obesity. Z-scores for executive function, memory, and global cognition were used to quantify cognitive performance. While obese participants tended to perform worse on average than normal weight individuals there were no significant differences in GSA 2020 Annual Scientific Meeting Innovation in Aging, 2020, Vol. 4, No. S1 performance between obese and normal weight participants in global cognition (p=.134), executive function (p=.164), or episodic memory (p=.708). Additionally, age did not moderate this relationship. However, there was a significant effect of education on all three domains. When stratified by education, participants with some college or higher, had a significant time*obesity*age interaction (F[3,328]=3.016, p<.05). For the oldest-old participants, executive function scores were higher for obese participants at follow-up compared to normal weight participants, but not at baseline. These findings suggest that level of education may serve as a form of cognitive reserve which compensates for deficits due to obesity.

UP NORTH AND DOWN SOUTH: REGIONAL DIFFERENCES IN PAIN, RELIGIOUS COPING, AND NEGATIVE AFFECT IN OSTEOARTHRITIS
Katherine Cheesman, 1 Patricia Parmelee, 2 and Dylan Smith, 3 1. The University of Alabama, Tuscaloosa, Alabama, United States, 2. University of Alabama,Tuscaloosa,Alabama,United States,3. Stony Brook University,Stony Brook,New York,New York,United States Objective: This research examines regional differences (Northern vs. Southern) in pain, religious coping, and negative affect among African Americans (AA) and non-Hispanic Whites (NHW) over the age of 50 with physician-confirmed knee osteoarthritis (OA). Methods: As part of a larger study of racial/ethnic differences in everyday quality of life with OA, 116 persons were recruited from sites in Alabama (n = 64) and New York (n = 52). Participants completed global measures of pain (PGC Pain Scale) and religious coping (Brief RCOPE); daily variability in pain, coping, and affect was assessed using a daily diary methodology consisting of 4 daily phone calls over 7 days. Site comparisons were conducted using one-way multivariate analysis of covariance (MANCOVA) with covariates of race, sex, education, and marital status. Results: There was a significant multivariate effect of site on pain, religious coping, and affect, F(5, 104) = 3.846, p = .003, Wilk's Λ = .844, partial η2 = .156. Follow-up univariate tests and mean examinations revealed that Southerners reported statistically more daily pain (M = 2.023, SD = .89), religious coping (M = .618, SD = .427), and negative affect (M = 6.556, SD = 2.661) than Northerners (M = 1.810, SD = .719; M = .386, SD = .417; M = 5.865, SD = 1.446). Implications: Results contribute to a growing understanding of how individuals use their religious beliefs to cope with daily pain. (Supported by R01-AG041655 D. Smith and P. Parmelee, PIs.)

ASSESSMENT OF A POTENTIAL UTI-USING VIDEOS TO ANALYZE A SIMULATED NURSING HOME EXPERIENCE
Donna Owen, 1 Alyce Ashcraft, 2 Kyle Johnson, 3 and Huaxin Song, 3 1. Texas Tech University Health Sciences Center,Houston,Texas,United States,2. Texas Tech University Health Sciences Cener,Lubbock,Texas,United States,3. Texas Tech University Health Sciences Center,Lubbock,Texas,United States In the past, urinary tract infections (UTIs) have been attributed to poor hygiene, the natural course of aging, an unfortunate corollary of nursing home (NH) residence, and a condition routinely treated empirically with antibiotics. Recent UTI management consensus statements foretell a very different future and support the need to consider all UTIs in the NH as complex infections. Improving assessment capabilities of the NH nurse workforce is essential for improving quality of care. This study aimed to determine how, using simulation, licensed vocational nurses (LVNs) integrated a mobile decision-support app (MDS-app) into assessment of a NH resident with a potential UTI. The MDS-app directed the LVN to examine or question the resident (mannequin) to identify signs and symptoms developed as part of a simulated clinical scenario. MDS-app items were based on UTI practice guidelines. A descriptive, participant observation design (video-taped) was used with ten practicing LVNs. An observation checklist was used to examine audiovisual recordings and included frequency of verbal interaction (17.9+/-7.2), and eye contact (10.6+/-4.1). Participants (47%) were "glued to" the MDS-app without making resident eye contact or touching residents during the assessment. 60% of participants deviated from the app to ascertain urine odor and color; irrelevant symptoms for UTI diagnosis. Assessments required 11.20 (+/-4.67) minutes to complete. The MDS-app provided LVNs with needed focus on data driven by guidelines and not individual LVN preferences. Training LVNs should focus on integration of communication, assessment skills, and MDS-app use for evidence-based data collection as a basis for UTI treatment decisions. Mealtime difficulties are common in residents with dementia, leading to negative outcomes. Interaction with staff are critical to engage residents in eating. This study characterized dyadic verbal interactions (descriptive statistics), and relationships among verbal behaviors and between verbal behaviors and individual characteristics (bivariate analyses). This secondary analysis of 110 videotaped mealtime observations involved 25 residents and 29 staff (42 unique dyads) in 9 nursing homes (NH). Verbal behaviors (utterances) were coded using the Cue Utilization and Engagement in Dementia mealtime video-coding scheme, addressing 8 positive behaviors and 4 negative behaviors. Staff spoke three times more frequently (76.5%) than residents (23.5%). Nearly all staff utterances were positive (99.2%). 85.1% of residents' utterances were positive and 14.9% negative. Staff positive utterances were associated with staff negative utterances (p=.02), and resident positive