Occupational Differences in Metabolic Syndrome Incidence Among Older Workers

Abstract This study investigates whether the incidence of metabolic syndrome (MetS), and its components, differs by occupational group among older workers (45-65 years) and whether health behaviors (smoking, leisure-time physical activity, diet quality) can explain these differences. We analyzed data from older workers (N=23 051) from two comprehensive measurement waves of the Lifelines Cohort Study and Biobank. MetS components were determined by physical measurements, blood markers, medication use, and self-reports. Occupational group and health behaviors were assessed by questionnaires. The association between occupational groups and MetS incidence was examined using Cox regression analysis. Health behaviors were subsequently added to the model to examine whether they can explain differences in MetS incidence between occupational groups. Low skilled white-collar (HR: 1.25, 95% CI: 1.13, 1.39) and low skilled blue-collar (HR: 1.45, 95% CI: 1.25, 1.69) workers had a significantly higher MetS incidence risk during 3.65 years follow-up than high skilled white-collar workers. Health behaviors reduced the strength of the association between occupational group and MetS incidence most among low skilled blue-collar workers (i.e. 10.3% reduction) as unhealthy behaviors were more prevalent in this occupational group. Similar occupational differences were observed on MetS component level. To conclude, MetS incidence in older workers differs between occupational groups and health behaviors only explain a small part of these differences. Health promotion tailored to occupational groups may be beneficial specifically among older low skilled blue-collar workers. Research into other factors that contribute to occupational differences is needed, as well as studies spanning the entire working life course.


TYPOLOGY OF FAMILY RELATIONSHIPS AND 6-YEAR MORTALITY RISK
Mengting Li, 1 and XinQi Dong, 2 1. Rutgers,The State University of New Jersey,New Brunswick,New Jersey,United States,2

. Rutgers University, Rutgers Institute for Health, New Jersey, United States
Durkheim's "family protection" thesis indicated that family could protect individuals from death. However, there are heterogeneities in family types. It remains unclear whether all family types play a protective role in older adults' later life. This study aims to test the relationship between family types and 6-year mortality. Data were derived from a prospective cohort study from 2011 to 2017 of 3,018 U.S. Chinese older adults in Chicago. Family typology was clustered by Latent Class Analysis, including tight-knit (high solidarity and low conflict), unobligated ambivalent (high solidarity and high conflict), commanding conflicted (low solidarity and high conflict), and detached (low solidarity and low conflict). Cox model was used. The result showed that older adults in detached type have higher mortality risk than those in tight-knit type after controlling age, gender, education, income, and medical conditions. Future study could explore the mechanisms through which family types affect mortality risk.

ASSOCIATIONS BETWEEN ULTRA-PROCESSED FOOD INTAKE AND CARDIOMETABOLIC HEALTH AMONG DIVERSE U.S. ADULTS 50 OR OLDER
Tali Elfassy, 1 Filippa Juul, 2 Robert Mesa, 1 Latha Palaniappan, 3 Malathi Srinivasan, 3 and Stella Yi, 2 1. University of Miami,Miami,Florida,United States,2. New York University,New York,New York,United States,3. Stanford University,Stanford,California,United States Asian American (AA) diets are naturally adapted to the NOVA dietary recommendations, favoring minimally processed foods. Yet the relationship between dietary intake and metabolic health, among AAs is largely unknown. We examined the association between ultra-processed foods and cardiometabolic health (obesity, hypertension, high cholesterol, and diabetes), among US adults 50 or older reporting a single ethnicity, using the National Health and Nutrition Examination Survey (2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018). From multivariable adjusted logistic regression models, the highest compared to the lowest quartile of ultra-processed food intake was associated with obesity only, among AAs (OR: 2.15, 95% CI: 1.04, 4.45), followed by non-Hispanic blacks (OR: 1.73, 95% CI: 1.40, 2.14), non-Hispanic whites (OR: 1.41, 95% CI: 1.19, 1.68), and Hispanics (OR: 1.34, 95% CI: 1.08, 1.65). AAs are more likely than other ethnic/racial groups to be obese when consuming ultra-processed foods. Cultural adaptation of current North American-focused dietary recommendations should drive AA preventive dietary recommendations. The prospective association of body weight variability with dementia remains unclear. We aimed to investigate whether long-term variability in body weight is associated with the risk of late-life dementia and to explore their potential temporal relationship using data from a nationwide prospective cohort study of the United States. A total of 5,556 participants free of dementia in 2008 (55.66% women; mean [SD] age, 71.1 [3.1] years) were followed up to 8 years for doctor-diagnosed dementia reported biennially. Body weight variability was assessed as the coefficient of variation utilizing the body weight information collected over 16 years before 2008. Cox proportion hazard model was applied to estimate hazard ratio (HR) of dementia associated with body weight variability. Higher body weight variability was associated with an increased incidence of dementia after controlling for sociodemographic factors, lifestyle, mean body weight, and body weight change. The multi-variable adjusted HR of dementia of the highest quartile of body weight variability was 2.01 (95% CI 1.01-1.87) compared with the lowest. Every 1% increment in variability was associated with a 6.2% higher risk of dementia (HR=1.06, 95%CI 1.04,1.09, p-trend<0.001). Such association was observed for both Alzheimer's disease and other types of dementia, with stronger association observed when body weight variability was assessed closer to dementia assessment.  Aging, 2021, Vol. 5, No. S1 This study investigates whether the incidence of metabolic syndrome (MetS), and its components, differs by occupational group among older workers (45-65 years) and whether health behaviors (smoking, leisure-time physical activity, diet quality) can explain these differences. We analyzed data from older workers (N=23 051) from two comprehensive measurement waves of the Lifelines Cohort Study and Biobank. MetS components were determined by physical measurements, blood markers, medication use, and self-reports. Occupational group and health behaviors were assessed by questionnaires. The association between occupational groups and MetS incidence was examined using Cox regression analysis. Health behaviors were subsequently added to the model to examine whether they can explain differences in MetS incidence between occupational groups. Low skilled white-collar (HR: 1.25, 95% CI: 1.13, 1.39) and low skilled blue-collar (HR: 1.45, 95% CI: 1.25, 1.69) workers had a significantly higher MetS incidence risk during 3.65 years follow-up than high skilled white-collar workers. Health behaviors reduced the strength of the association between occupational group and MetS incidence most among low skilled blue-collar workers (i.e. 10.3% reduction) as unhealthy behaviors were more prevalent in this occupational group. Similar occupational differences were observed on MetS component level. To conclude, MetS incidence in older workers differs between occupational groups and health behaviors only explain a small part of these differences. Health promotion tailored to occupational groups may be beneficial specifically among older low skilled blue-collar workers. Research into other factors that contribute to occupational differences is needed, as well as studies spanning the entire working life course.

REMOTE MONITORING ACTIVITY TRAJECTORY-ASSOCIATED WEIGHT LOSS AND FUNCTIONAL ABILITY IN OBESE OLDER ADULTS
John Batsis, 1 Curtis Petersen, 2 Anna Kahkoska, 3 and Karen Fortuna, 2 1. University of North Carolina at Chapel Hill,Chapel Hill,North Carolina,United States,2. Dartmouth College,Lebanon,New Hampshire,United States,3. University of North Carolina at Chapel Hill,University of North Carolina at Chapel Hill/Chapel Hill,North Carolina,United States Functional decline in older adults can often be mitigated by physical activity. As older adults increasing adopt wearable technology, an understanding of how remotely monitored activity is associated with clinical outcomes is needed. Data was analyzed from two cohorts of older adults with obesity (□65 years, BMI (□30kg/m2)) who completed weekly dietary and exercise-based weight loss interventions (n=93). Follow-up time varied between cohorts (n=37: 12-weeks; n=56: 16 weeks). All participants were provided a Fitbit to monitor physical activity. Baseline and follow up weight, 6-minute walk distance, grip strength, and Late Life Function and Disability Instrument (LLFDI) were collected. We used k-means clustering for longitudinal data to identify physical activity trajectories from Fitbit steps at the daily level. Linear regression models tested for differences in each outcome between trajectories, adjusting for age and sex. Baseline characteristics did not vary across cohorts: mean age 72.7±4.5 years, 76.5% were female, and mean BMI was 36.4±5.1 kg/m2. Two physical activity trajectories were identified, termed high and low activity based on differences in mean daily steps (7,476±4,117 vs. 2,960±2,453, p <0.001). Participants in the high activity group experienced a 2.4% reduction in weight (p <0.001) and a 4.74% increase in LLFDI score (p=0.007) relative to the low activity cluster. Other outcomes were not significantly different between trajectories. These results demonstrate the potential for remote monitoring data to elucidate longitudinal trends in weight and functional ability. As such, older adults' use of wearable technology may facilitate improvements in weight and functional ability in the community.

THE ASSOCIATION OF MEAL TIMING WITH BODY COMPOSITION AND CARDIOMETABOLIC HEALTH IN OBESE OLDER ADULTS
Samaneh Farsijani, 1 Nancy W. Glynn, 2 and Anne Newman, 1 1. University of Pittsburgh,Pittsburgh,Pennsylvania,United States,2. University of Pittsburgh Graduate School of Public Health,Pittsburgh,Pennsylvania,United States Objectives: To determine the association between eating window and time of last calorie intake with body composition and cardiometabolic health in obese older adults. Methods: We performed a cross-sectional analysis on 36 communitydwelling, overweight-to-obese (BMI 28.0-39.9 kg/m2) older adults, recruited to participate in a weight loss and exercise trial. Time of food intake were extracted from three 24-hour food recalls. Eating window was calculated as the time elapsed between the first and last food intake. We recorded the time of last calorie intake either from food or drink. Blood glucose, triglycerides, high-density (HDL) & low-density (LDL) lipoprotein cholesterols were measured as markers of cardiometabolic health. Total fat and lean mass were assessed by DXA. Partial correlation was used to determine the relationships between eating window and last calorie intake with body composition and cardiometabolic markers, while controlling for sex, age, and total calorie intake. Results: On average, participants' eating window was 12.0±1.1 hours. Time of last calorie intake in 86% of participants was between 6:00-8:00 PM. After controlling for potential confounders, longer eating windows were associated with higher triglyceride levels (P=0.032) and lower HDL (P=0.035), while no association was observed with the other cardiometabolic markers. We observed negative trends, though not statistically significant, between longer eating windows and greater weight, BMI, and fat mass. No association was observed between time of last calorie intake, body composition and cardiometabolic markers. Conclusions: Our results suggest that timing of food intake may influence cardiometabolic risk and obesity in older adults. We aimed to study the association of long-terms trajectories of body mass index (BMI) with contemporaneous