Interpretable machine learning for high-dimensional trajectories of aging health

We have built a computational model for individual aging trajectories of health and survival, which contains physical, functional, and biological variables, and is conditioned on demographic, lifestyle, and medical background information. We combine techniques of modern machine learning with an interpretable interaction network, where health variables are coupled by explicit pair-wise interactions within a stochastic dynamical system. Our dynamic joint interpretable network (DJIN) model is scalable to large longitudinal data sets, is predictive of individual high-dimensional health trajectories and survival from baseline health states, and infers an interpretable network of directed interactions between the health variables. The network identifies plausible physiological connections between health variables as well as clusters of strongly connected health variables. We use English Longitudinal Study of Aging (ELSA) data to train our model and show that it performs better than multiple dedicated linear models for health outcomes and survival. We compare our model with flexible lower-dimensional latent-space models to explore the dimensionality required to accurately model aging health outcomes. Our DJIN model can be used to generate synthetic individuals that age realistically, to impute missing data, and to simulate future aging outcomes given arbitrary initial health states.

Fundamental Clock of Biological Aging: Convergence of Molecular, Neurodegenerative, Cognitive and Psychiatric Pathways: Non-Equilibrium Thermodynamics Meet Psychology

In humans, age-associated degrading changes, widely observed in molecular and cellular processes underly the time-dependent decline in spatial navigation, time perception, cognitive and psychological abilities, and memory. Cross-talk of biological, cognitive, and psychological clocks provides an integrative contribution to healthy and advanced aging. At the molecular level, genome, proteome, and lipidome instability are widely recognized as the primary causal factors in aging. We narrow attention to the roles of protein aging linked to prevalent amino acids chirality, enzymatic and spontaneous (non-enzymatic) post-translational modifications (PTMs SP), and non-equilibrium phase transitions. The homochirality of protein synthesis, resulting in the steady-state non-equilibrium condition of protein structure, makes them prone to multiple types of enzymatic and spontaneous PTMs, including racemization and isomerization. Spontaneous racemization leads to the loss of the balanced prevalent chirality. Advanced biological aging related to irreversible PTMs SP has been associated with the nontrivial interplay between somatic (molecular aging) and mental (psychological aging) health conditions. Through stress response systems (SRS), the environmental and psychological stressors contribute to the age-associated “collapse” of protein homochirality. The role of prevalent protein chirality and entropy of protein folding in biological aging is mainly overlooked. In a more generalized context, the time-dependent shift from enzymatic to the non-enzymatic transformation of biochirality might represent an important and yet underappreciated hallmark of aging. We provide the experimental arguments in support of the racemization theory of aging.

Measurement Invariance in the Assessment of Mood Between American and Mexican Community Studies

The Health and Retirement Study (HRS), a principal source for American public health research, has numerous global sister studies. Harmonization efforts seeking to establish measurement equivalence amongst these various datasets, is a critical prerequisite to cross-cultural research. Given well-known cultural variability in depressive symptom endorsement, the purpose of this study was to assess measurement invariance in a brief mood measure used in the HRS and the Mexican Health and Aging Study (MHAS). Total sample size using both groups was 15,319 participants (10,931 HRS; 4,388 MHAS) who were 65 and older from Waves 6 to 13 in the HRS and Waves 1 to 4 in the MHAS. MPlus Version 8.4 was used to conduct CFA analyses of measurement invariance. A contemporary approach with categorical data calls for examining threshold invariance first while establishing configural invariance, before examining invariance tests of thresholds, loadings, and intercepts in a second step. Results were that measurement invariance was not supported in this series of two steps with four out of six indices showing model fit in the first model and none of the indices showing model fit in the second model. These findings implied that there were differences in ways of responding to the brief mood measure between HRS and MHAS participants at the conceptual level. Thus, comparisons based on these measures may result in misleading findings and should be interpreted very conservatively. This study adds to the growing body of literature guiding harmonization efforts from the Program on Global Aging, Health and Policy.

Effects of Loneliness and Social Isolation on Cognitive Health: Latest Perspectives and Future Directions

Loneliness and social isolation as antecedents of cognitive decline have received substantial attention in recent research. This symposium addresses this year’s conference theme of aging in the “new normal”. The COVID-19 pandemic has highlighted the negative impacts of loneliness and social isolation on older adults’ health and wellbeing. This symposium includes studies that shed light on the relationships between loneliness, social isolation and cognitive health using a multidisciplinary approach, and provide recommendations and future directions for advancing this research area. The first presentation examines cardiovascular biomarkers as potential mechanisms that mediate the longitudinal relationship between loneliness and cognitive decline with the HRS dataset. The second presentation examines several social isolation indicators and their effects on cognitive decline in a Canadian longitudinal study. Using the US ADRC longitudinal study of aging, the third study shows the effect of loneliness on cognitive health in older adults pre- and post-onset of the COVID-19 pandemic. The symposium concludes with a literature review of the different measures employed to operationalize the constructs of loneliness, isolation, which resulted in heterogeneous study findings on their influences on the risk of developing dementia. This review calls for consistent measures to produce comparable evidence on the health consequences of loneliness and isolation. In all, this symposium reports and reviews the latest evidence on the association between social isolation, loneliness and cognitive health amidst the ongoing COVID-19 pandemic. It also echoes the conference theme of transforming disruption to opportunities in aging health service and research.

Plant Polyphenols for Aging Health: Implication from Their Autophagy Modulating Properties in Age-Associated Diseases

Polyphenols are a family of naturally occurring organic compounds, majorly present in fruits, vegetables, and cereals, characterised by multiple phenol units, including flavonoids, tannic acid, and ellagitannin. Some well-known polyphenols include resveratrol, quercetin, curcumin, epigallocatechin gallate, catechin, hesperetin, cyanidin, procyanidin, caffeic acid, and genistein. They can modulate different pathways inside the host, thereby inducing various health benefits. Autophagy is a conserved process that maintains cellular homeostasis by clearing the damaged cellular components and balancing cellular survival and overall health. Polyphenols could maintain autophagic equilibrium, thereby providing various health benefits in mediating neuroprotection and exhibiting anticancer and antidiabetic properties. They could limit brain damage by dismantling misfolded proteins and dysfunctional mitochondria, thereby activating autophagy and eliciting neuroprotection. An anticarcinogenic mechanism is stimulated by modulating canonical and non-canonical signalling pathways. Polyphenols could also decrease insulin resistance and inhibit loss of pancreatic islet β-cell mass and function from inducing antidiabetic activity. Polyphenols are usually included in the diet and may not cause significant side effects that could be effectively used to prevent and treat major diseases and ailments.

The Long-Term Public Health Impact of Social Distancing on Brain Health: Topical Review

Social distancing has been a critical public health measure for the COVID-19 pandemic, yet a long history of research strongly suggests that loneliness and social isolation play a major role in several cognitive health issues. What is the true severity and extent of risks involved and what are potential approaches to balance these competing risks? This review aimed to summarize the neurological context of social isolation and loneliness in population health and the long-term effects of social distancing as it relates to neurocognitive aging, health, and Alzheimer’s disease and related dementias. The full scope of the underlying causal mechanisms of social isolation and loneliness in humans remains unclear partly because its study is not amenable to randomized controlled trials; however, there are many detailed experimental and observational studies that may provide a hypothesis-generating theoretical framework to better understand the pathophysiology and underlying neurobiology. To address these challenges and inform future studies, we conducted a topical review of extant literature investigating associations of social isolation and loneliness with relevant biological, cognitive, and psychosocial outcomes, and provide recommendations on how to approach the need to fill key knowledge gaps in this important area of research.

The Sharing Economy in China’s Aging Industry: Applications, Challenges, and Recommendations

Background All aging societies face the challenge of allocating limited resources for the highest value of use. The sharing economy provides one method to address the imbalance between the demand and supply of health services to the older adult population. With a substantial aging population, China’s practices in the sharing aging industry may set examples for other “getting old before getting rich” countries. Objective There is a gap in both the data and research on China’s aging industry sharing economy. This paper addresses these data and research lacunae by constructing a framework for the application of a sharing model in China’s aging industry, by assessing the current state of the aging industry sharing economy, by setting out the challenges to the sharing aging health care and service economy, and by making recommendations for the development of the aging industry sharing economy. Methods This paper constructs a sharing economy framework in the aging industry covering four aspects (people, facilities, capital, and information) to test the current state and future prospects of China’s aging industry sharing economy. Results In people sharing, we analyzed the sharing of emotional companionship, doctors, nurses, nursing attendants, and domestic helpers. We discussed facility sharing models from the point of land and housing, medical devices, and other items such as pensioner meals and shared medicine bins. We acknowledge that crowdfunding platforms have developed fast in China, but many older adult users faced problems in their operation. Information sharing is a developing field, which can optimize users’ experiences and should help older adults filter out misinformation, but China currently does not have adequate sharing information platforms for older adults. Conclusions We identified four major challenges in China’s aging industry sharing economy: poor adaptability to technology for older adults, mediocre quality of shared services, one-size-fits-all and the concept of the useless elderly, and shortage of qualified practitioners. We make recommendations for specific measures by governments, communities, and enterprises to improve the sharing economy in the aging industry.