In the United States, the number of individuals who are 90 and older is expected to quadruple from 1.9 million in 2010 to 7.6 million individuals in 2050. Aging into the later years is often accompanied by time spent with mobility and cognitive impairments, multiple morbidity and disability in activities of daily living. Impairments in either the physical or mental domains contribute to declines in one’s intrinsic capacity to maintain high functioning and wellbeing.The pace of biological aging differs across older adults of the same chronological age. Faster biological aging is linked to greater accumulation of multiple morbidities, lower physical and cognitive functioning, and earlier death. Epigenetic age, a biomarker of aging, is a composite measure of DNA methylation across specific cytosine-guanine dinucleotide sites (CpG sites) associated with chronologic or phenotypic age. Epigenetic age acceleration, as measured by epigenetic clocks, is postulated to identify whether individuals are aging faster or slower when compared to their chronologic age. There have been no prospective studies that have examined the relationships between genome wide DNA methylation or epigenetic age acceleration with multiple morbidity at a specified older age or survival to 90 years of age with intact mobility and cognitive functioning. The first chapter of this dissertation reviews the epidemiological evidence on DNA methylation and epigenetic clocks and their relationships with longevity, mobility, cognitive functioning and multiple morbidity. The second chapter assesses the associations between four DNA methylation clocks that measure epigenetic age acceleration and exceptional longevity, defined as survival to age 90 with intact mobility and cognitive function. The third chapter utilizes an epigenome wide association study to identify specific CpG sites and regions that are associated with exceptional longevity. The fourth chapter evaluates associations between the four DNA methylation clocks and multiple morbidity among women when they reach 90 years of age. The final chapter summarizes and integrates the key findings from this dissertation and highlights future directions for research on the potential of DNA methylation and epigenetic clocks to expand knowledge of the etiology of healthy longevity and to identify targets for intervention to improve healthy aging.

Background: Black and Hispanic older adults with arthritis are less likely to receive total knee arthroplasty (TKA) than non-Hispanic Whites despite the effectiveness of the procedure. Disparities in functional outcomes after TKA have also been recognized. The mechanisms for disparities in utilization and functional outcomes remain unknown.

Methods: Data from 102,767 women enrolled in the Women’s Health Initiative who had linked Medicare claims data were used to evaluate racial/ethnic disparities in TKA utilization and outcomes. Chapter 1 evaluated utilization rates after adjusting for need for TKA and medical access factors. Chapter 2 assessed physical function in the decades pre- and post-TKA by race/ethnicity. Chapter 3 evaluated racial disparities in post-TKA rehabilitation.

Results: Chapter 1: Black and Hispanic women were significantly less likely to receive TKA than non-Hispanic White women after accounting for need [Black: HR (95%CI)=0.70 (0.63-0.79); Hispanic=0.58 (0.44-0.77)]. Medical access factors had a modest effect on mediating disparities in utilization [Black: HR (95%CI)=0.75 (0.67-0.88); Hispanic=0.65 (0.47-0.89)]. Chapter 2: Among women who underwent TKA, Black women had significantly worse physical function for the entire decade pre-operatively compared with White women (mean difference=5.8 points, p-value<0.01) and had delayed surgery from the time of onset of mobility limitations. Poorer function pre-operatively largely explained racial disparities in functional outcomes post-TKA. Chapter 3: Black women, who had poorer function pre-operatively than White women, received more intensive rehabilitation services postoperatively. After accounting for differences in need for PT services, Black and White women appear to be equally as likely to continue rehabilitation after discharge home, through receipt of either home health or outpatient services.

Discussion: Neither need nor medical access explained why Black and Hispanic women underutilized TKA procedures compared with White women. Delays in receiving TKAs among Black women contribute to poorer functional improvement postoperatively. A better understanding of the decision-making process among surgical candidates is needed to ensure that women of minority race/ethnicity with arthritis receive appropriate treatment and in a timely manner. Efforts to reduce postoperative functional outcomes should be focused pre-operatively at optimizing physical function for Black and Hispanic women with arthritis and improving timely receipt of TKA once need arises.

Investigations into the link between vitamin D and cardiovascular disease (CVD) have yielded inconsistent results. The vitamin D metabolite ratio (VMR), the ratio of 24,25(OH)2D to 25(OH)D, has shown stronger associations with fracture and mortality than 25(OH)D alone. Our study assessed the association between the VMR and CVD outcomes. We evaluated a cohort of 6,313 participants from the Multi-Ethnic Study of Atherosclerosis (MESA), without pre-existing CVD, over 15 years. Utilizing Cox regression, we examined the associations of both the VMR and 25-hydroxyvitamin D [25(OH)D] with various cardiovascular events. Over the study, 800 participants developed CVD, including conditions such as myocardial infarction, resuscitated cardiac arrest, stroke, coronary heart disease death, and stroke death. Heart failure (HF) was observed in 398 participants, and 413 experienced cardiovascular mortality. Models were adjusted for factors including demographics, lifestyle, clinical conditions and medications, biomarkers, and kidney function. Participants averaged 62 years (range 44-84), with 53% females. The mean (SD) 25(OH)D level was 22.7 (11.0) ng/mL, and the mean VMR was 15.2 (5.0). In fully adjusted models, a two-fold increase in VMR was associated with a 24% reduction in incident CVD (HR: 0.76, 95% CI: 0.65-0.88). However, there was no association between the VMR and HF (0.98, 0.78-1.24), or cardiovascular mortality (0.96, 0.77-1.21). 25(OH)D was not significantly associated with any CVD outcome. In a diverse cohort, VMR was significantly associated with reduced incident CVD, but not HF or cardiovascular mortality. The results suggest that VMR may provide greater insight into vitamin D metabolism, compared with 25(OH)D alone.

Background: Regular physical activity (PA) reduces the risk of new chronic diseases, slows the progression of prevalent chronic diseases, and promotes other health benefits. An older individual’s level of energy expenditure while performing an activity (absolute intensity) may be discordant with their level of exertion relative to their maximal possible effort which expresses the very same amount of absolute effort as a percent of the individual’s maximum capacity (relative intensity). VO2max (maximal oxygen uptake (mL/kg/min)), the gold standard measurement of cardiorespiratory fitness, can be used to normalize absolute activity; if the VO2max of an individual is known and their instantaneous oxygen uptake (VO2) can be measured, ((VO2 / VO2max)*100) gives the percent of their maximal exercise capacity (i.e., their relative intensity). Methods: Aim 1 assessed the performance of previously published VO2max prediction equations in relation to measured VO2max and recalibrated the equations. Aim 2 developed new machine-learned (ML) VO2max prediction algorithms. In the final aim, daily hours spent in light intensity and moderate-to-vigorous PA (MVPA) were calculated on the absolute scale (accelerometer-measured PA) and relative scale (accelerometer-estimated VO2 / predicted VO2max using Aim 2’s algorithms). The associations between absolute and relative intensity PA, total mortality, and incident major cardiovascular disease (CVD) were quantified. Results: In Aim 1, the prediction equations yielded root mean squared error (RMSE) values ranging from 4.2 to 20.4 mL/kg/min and from 3.9-4.2 mL/kg/min after recalibration. The newly developed ML algorithms in Aim 2 yielded RMSE values ranging from 2.9-4.4 mL/kg/min. In Aim 3, on each PA measurement scale (relative and absolute), an increase in light intensity and MPVA reduced the risk of both outcomes. On the absolute scale, MVPA was more strongly associated with both outcomes than light intensity, but the opposite occurred on the relative scale. Conclusion: The PA intensity paradigm should keep shifting towards recommendation of more movement, regardless of intensity, and placing greater emphasis on relative light intensity (37% - 46% of maximal capacity) as modifiable behavioral targets that are more easily achieved, reduce risks of death and CVD, and promote healthy aging.

Background: In the United States, the aging population is rapidly growing. By 2060, it is expected that 12 million women will be ages 85 years and older. However, determinants of longevity and healthy aging in women are not fully understood. This dissertation had three objectives: 1) Determine whether ages at menarche and menopause and reproductive lifespan were associated with survival to age 90, termed “exceptional longevity;” 2) Determine whether genetic factors associated with longevity in prior studies among populations of European descent were associated with survival to ages 85 and 90 and healthy aging in white, African-American, and Hispanic women; and 3) Determine whether accelerometer-measured and self-reported sedentary time were associated with leukocyte telomere length (LTL), a purported biomarker of aging, among older women.

Methods: Three studies were conducted among participants from the Women’s Health Initiative, a longitudinal study investigating major determinants of chronic diseases in postmenopausal women. Study one was a prospective study among 16,251 women who had potential to survive to age 90 as of August 29, 2014. Study two was a prospective study among 11,154 women who could survive to age 85 as of August 29, 2014 and used genetic data from multiple genome-wide association studies. Study three was a cross-sectional study among 1,481 women with information on either accelerometer-measured or self-reported sedentary time. All studies consisted of racially diverse samples.

Results: In study one, the odds of exceptional longevity were elevated among women with later menarche, later menopause, and longer reproductive lifespan. In study two, three variants at APOE were associated with survival to age 90 and healthy aging in white women, and seven variants at a novel locus were associated with survival to age 85 in Hispanic women. In study three, among women at or below the median level of moderate-to-vigorous intensity physical activity (MVPA), higher accelerometer-measured sedentary time was associated with shorter LTL. Among women with higher MVPA levels, sedentary time was not associated with LTL.

Conclusions: Findings suggest that reproductive and genetic factors may be associated with late-age survival and that a high level of inactivity may be associated with short LTL.

Background: Older adults exhibit more fragmented circadian rhythm as they age, and undergo age-related changes in sleep patterns that influence sleep duration, timing, and overall satisfaction. Several challenges exist in conducting inclusive research for promoting healthy sleep and aging over the lifespan. Multiple sleep dimensions exist and complicate synthesis of sleep findings. Sleep health disparities also disproportionately affect the same populations that experience overall health disparities, yet diverse groups are underrepresented. Methods: Study 1 examined prospective associations between self-reported sleep from the Hispanic Community Health Study / Study of Latinos (n=10,640) and actigraphy-derived sleep/circadian measures from Sueño (n=1,808) with multimorbidity at follow-up. Study 2 used an ensemble of machine learning techniques to identify accelerometry-derived sleep and circadian profiles; we then examined associations between sleep-circadian profiles, fall risk, and physical functioning among 4,543 diverse older women in OPACH. Study 3 linked accelerometry data, claims, and genetic data to construct two network models to simultaneously evaluate the relationship between multiple sleep rest-activity rhythm measures with each cognitive outcome (e.g., dementia and Alzheimer’s Disease). Results: In study 1, several sleep and circadian measures were associated with comorbidity at follow-up, and we observed effect modification of these associations by US-born and non-US-born status and duration of residency. In study 2, we identified multiple sleep and circadian profiles using machine learning, and these profiles were associated with greater fall risk and lower physical functioning. In study 3, several circadian measures were indirectly associated with dementia and AD and shared a central hub in the network model. Results from adjusted survival models showed consistent associations with the network model. Conclusion: This analysis enhances the field of life course epidemiology and sleep disparities research, having identified sleep and circadian behaviors as risk factors for disease and worse aging at middle and older ages. This study also highlights study designs that promote inclusive research when performing epidemiological studies on sleep.

Background: As the US population of older women continues to increase, promoting healthy longevity has become a public health priority. Increasing physical activity (PA) and limiting sedentary behavior (SB) can be an effective strategy for facilitating healthy longevity, as it is known to reduce the risk of developing many chronic conditions, manage the progression of current conditions, and helps maintain independence and mobility. However, evidence on the associations between accelerometer-measured PA (i.e., light, moderate-to-vigorous, total, daily steps), SB (i.e., sitting time, mean sitting bout duration), and healthy longevity among older women is lacking.Methods: Aim 1 and Aim 2 use data from the Women’s Health Accelerometry Collaboration (WHAC). Aim 1 quantified the prospective associations between accelerometer-measured PA, SB, and survival to age 90 with intact mobility among women ages 78–89 years at baseline in WHAC. Aim 2 examined the prospective associations between accelerometer-measured PA, SB and all-cause, cancer, and cardiovascular disease mortality among cancer survivors in WHAC. In Aim 3, we characterized the longitudinal changes in accelerometer-measured PA among older women in the Women’s Health Initiative Strong and Healthy (WHISH) trial. Results: In Aim 1, increasing PA of any intensity, taking more steps, and reducing prolonged sitting were associated with increased odds of survival to age 90 with and without intact mobility. Among over 2,500 US cancer survivors examined in Aim 2, accelerometer-assessed PA and daily steps were associated with reduced risk of all cause and cardiovascular disease mortality, while higher levels of SB were associated with increased risks. In Aim 3, we showed that PA measures decreased over time. However, sociodemographic and health status factors were associated with different trajectories of PA over time. Conclusion: These data provide further evidence that more PA of any intensity and shortened prolonged sitting are associated with reducing mortality risk after cancer and healthy longevity among older women. Furthermore, we highlight the variability in PA trajectories over time related to demographic and health-related factors, which could support targeted screening and intervention efforts.