Sparse spatially clustered coefficient model via adaptive regularization.

Large spatial datasets with many spatial covariates have become ubiquitous in many fields in recent years. A question of interest is to identify which covariates are likely to influence a spatial response, and whether and how the effects of these covariates vary across space, including potential abrupt changes from region to region. To solve this question, a new efficient regularized spatially clustered coefficient (RSCC) regression approach is proposed, which could achieve variable selection and identify latent spatially heterogeneous covariate effects with clustered patterns simultaneously. By carefully designing the regularization term of RSCC as a chain graph guided fusion penalty plus a group lasso penalty, the RSCC model is computationally efficient for large spatial datasets while still achieving the theoretical guarantees for estimation. RSCC also adopts the idea of adaptive learning to allow for adaptive weights and adaptive graphs in its regularization terms and further improves the estimation performance. RSCC is applied to study the acceptance of COVID-19 vaccines using county-level data in the United States and discover the determinants of vaccination acceptance with varying effects across counties, revealing important within-state and across-state spatially clustered patterns of covariates effects.

A binary hidden Markov model on spatial network for amyotrophic lateral sclerosis disease spreading pattern analysis.

Amyotrophic lateral sclerosis (ALS) is a neurological disease that starts at a focal point and gradually spreads to other parts of the nervous system. One of the main clinical symptoms of ALS is muscle weakness. To study spreading patterns of muscle weakness, we analyze spatiotemporal binary muscle strength data, which indicates whether observed muscle strengths are impaired or healthy. We propose a hidden Markov model-based approach that assumes the observed disease status depends on two latent disease states. The model enables us to estimate the incidence rate of ALS disease and the probability of disease state transition. Specifically, the latter is modeled by a logistic autoregression in that the spatial network of susceptible muscles follows a Markov process. The proposed model is flexible to allow both historical muscle conditions and their spatial relationships to be included in the analysis. To estimate the model parameters, we provide an iterative algorithm to maximize sparse-penalized likelihood with bias correction, and use the Viterbi algorithm to label hidden disease states. We apply the proposed approach to analyze the ALS patients' data from EMPOWER Study.

Autologistic network model on binary data for disease progression study.

This paper focuses on analysis of spatiotemporal binary data with absorbing states. The research was motivated by a clinical study on amyotrophic lateral sclerosis (ALS), a neurological disease marked by gradual loss of muscle strength over time in multiple body regions. We propose an autologistic regression model to capture complex spatial and temporal dependencies in muscle strength among different muscles. As it is not clear how the disease spreads from one muscle to another, it may not be reasonable to define a neighborhood structure based on spatial proximity. Relaxing the requirement for prespecification of spatial neighborhoods as in existing models, our method identifies an underlying network structure empirically to describe the pattern of spreading disease. The model also allows the network autoregressive effects to vary depending on the muscles' previous status. Based on the joint distribution derived from this autologistic model, the joint transition probabilities of responses among locations can be estimated and the disease status can be predicted in the next time interval. Model parameters are estimated through maximization of penalized pseudo-likelihood. Postmodel selection inference was conducted via a bias-correction method, for which the asymptotic distributions were derived. Simulation studies were conducted to evaluate the performance of the proposed method. The method was applied to the analysis of muscle strength loss from the ALS clinical study.

Moving to an Activity-Friendly Community Can Increase Physical Activity.

BACKGROUND: Creating activity-friendly communities (AFCs) is an important strategy to increase physical activity (PA). While cross-sectional links between community environments and PA are well documented, their causal relationships remain insufficiently explored. METHODS: Using the accelerometer and survey data collected from adults who moved to an AFC (cases) and similar non-AFC-residing adults who did not move (comparisons), this pre-post, case-comparison study examines if moving to an AFC increases PA. Data came from 115 participants (cases = 37, comparisons = 78) from Austin, Texas, who completed 2 waves of 1-weeklong data collection. Difference-in-difference analyses and fixed-effect models were used to test the significance of the pre-post differences in moderate-to-vigorous PA (MVPA) between cases and comparisons, for the full sample and the subsample of 37 pairs matched in key covariates using the Propensity Score Matching method. RESULTS: Average treatment effect generated based on Propensity Score Matching and difference-in-difference showed that moving to this AFC led to an average of 10.88 additional minutes of daily MVPA (76.16 weekly minutes, P = .015). Fixed-effect models echoed the result with an increase of 10.39 minutes of daily MVPA after moving to the AFC. We also found that case participants who were less active at baseline and had higher income increased their MVPA more than their counterparts. CONCLUSIONS: This study showed that, among our study sample, moving to an AFC increased residents' PA significantly when compared to their premove level and the comparison group. This causal evidence suggests the potential of AFCs as sustainable interventions for PA promotion.

Age-volume associations in cerebellar lobules by sex and reproductive stage.

The cerebellum has established associations with motor function and a well-recognized role in cognition. In advanced age, cognitive and motor impairments contribute to reduced quality of life and are more common. Regional cerebellar volume is associated with performance across these domains and sex hormones may influence this volume. Examining sex differences in regional cerebellar volume in conjunction with age, and in the context of reproductive stage stands to improve our understanding of cerebellar aging and pathology. Data from 508 healthy adults (ages 18-88; 47% female) from the Cambridge Centre for Ageing and Neuroscience database were used here. CERES was used to assess lobular volume in T1-weighted images. We examined sex differences in adjusted regional cerebellar volume while controlling for age. A subgroup of participants (n = 370, 50% female) was used to assess group differences in female reproductive stages as compared to age-matched males. Sex differences in adjusted volume were seen across most anterior and posterior cerebellar lobules. Most of these lobules had significant linear relationships with age in males and females. While there were no interactions between sex and reproductive stage groups, exploratory analyses in females alone revealed multiple regional differences by reproductive stage. We found sex differences in volume across much of the cerebellum, linear associations with age, and did not find an interaction for sex and reproductive stage on regional cerebellar volume. Longitudinal investigation into hormonal influences on cerebellar structure and function is warranted as hormonal changes with menopause may impact cerebellar volume over time.

Nursing Home Design and COVID-19: Implications for Guidelines and Regulation.

OBJECTIVES: Nursing homes (NHs) are important health care and residential environments for the growing number of frail older adults. The COVID-19 pandemic highlighted the vulnerability of NHs as they became COVID-19 hotspots. This study examines the associations of NH design with COVID-19 cases, deaths, and transmissibility and provides relevant design recommendations. DESIGN: A cross-sectional, nationwide study was conducted after combining multiple national data sets about NHs. SETTING AND PARTICIPANTS: A total of 7785 NHs were included in the study, which represent 50.8% of all Medicare and/or Medicaid NH providers in the United States. METHODS: Zero-inflated negative binomial models were used to predict the total number of COVID-19 resident cases and deaths, separately. The basic reproduction number (R0) was calculated for each NH to reflect the transmissibility of COVID-19 among residents within the facility, and a linear regression model was estimated to predict log(R0 - 1). Predictors of these models included community factors and NHs' resident characteristics, management and rating factors, and physical environmental features. RESULTS: Increased percentage of private rooms, larger living area per bed, and presence of a ventilator-dependent unit are significantly associated with reductions in COVID-19 cases, deaths, and transmissibility among residents. After setting the number of actual residents as the exposure variable and controlling for staff cases and other variables, increased number of certified beds in the NH is associated with reduced resident cases and deaths. It also correlates with reduced transmissibility among residents when other risk factors, including staff cases, are controlled. CONCLUSIONS AND IMPLICATIONS: Architectural design attributes have significant impacts on COVID-19 transmissions in NHs. Considering the vulnerability of NH residents in congregated living environments, NHs will continue to be high-risk settings for infection outbreaks. To improve safety and resilience of NHs against future health disasters, facility guidelines and regulations should consider the need to increase private rooms and living areas.

Physical Activity Impacts of an Activity-Friendly Community: A Natural Experiment Study Protocol.

Background: Stakeholders from multiple sectors are increasingly aware of the critical need for identifying sustainable interventions that promote healthy lifestyle behaviors. Activity-friendly communities (AFCs) have been known to provide opportunities for engaging in physical activity (PA) across the life course, which is a key to healthy living and healthy aging. Purpose: Our purpose is to describe the study protocol developed for a research project that examines: (a) the short- and long-term changes in total levels and spatial and temporal patterns of PA after individuals move from non-AFCs to an AFC; and (b) what built and natural environmental factors lead to changes in PA resulting from such a move, either directly or indirectly (e.g., by affecting psychosocial factors related to PA). Methods: This protocol is for a longitudinal, case-comparison study utilizing a unique natural experiment opportunity in Austin, Texas, USA. Case participants were those adults who moved from non-AFCs to an AFC. Matching comparison participants were residents from similar non-AFCs who did not move during the study period. Recruitment venues included local businesses, social and print media, community events, and individual referrals. Objectively measured moderate-to-vigorous PA and associated spatial and temporal patterns served as the key outcomes of interest. Independent (e.g., physical environments), confounding (e.g., demographic factors), and mediating variables (e.g., psychosocial factors) were captured using a combination of objective (e.g., GIS, GPS, Tanita scale) and subjective measures (e.g., survey, travel diary). Statistical analyses will be conducted using multiple methods, including difference-in-differences models, repeated-measures linear mixed models, hierarchical marked space-time Poisson point pattern analysis, and hierarchical linear mixed models. Conclusion: Natural experiment studies help investigate causal relationships between health and place. However, multiple challenges associated with participant recruitment, extensive and extended data collection activities, and unpredictable intervention schedules have discouraged many researchers from implementing such studies in community-based populations. This detailed study protocol will inform the execution of future studies to explore how AFCs impact population health across the life course.

COVID-19: Short term prediction model using daily incidence data.

BACKGROUND: Prediction of the dynamics of new SARS-CoV-2 infections during the current COVID-19 pandemic is critical for public health planning of efficient health care allocation and monitoring the effects of policy interventions. We describe a new approach that forecasts the number of incident cases in the near future given past occurrences using only a small number of assumptions. METHODS: Our approach to forecasting future COVID-19 cases involves 1) modeling the observed incidence cases using a Poisson distribution for the daily incidence number, and a gamma distribution for the series interval; 2) estimating the effective reproduction number assuming its value stays constant during a short time interval; and 3) drawing future incidence cases from their posterior distributions, assuming that the current transmission rate will stay the same, or change by a certain degree. RESULTS: We apply our method to predicting the number of new COVID-19 cases in a single state in the U.S. and for a subset of counties within the state to demonstrate the utility of this method at varying scales of prediction. Our method produces reasonably accurate results when the effective reproduction number is distributed similarly in the future as in the past. Large deviations from the predicted results can imply that a change in policy or some other factors have occurred that have dramatically altered the disease transmission over time. CONCLUSION: We presented a modelling approach that we believe can be easily adopted by others, and immediately useful for local or state planning.

Improving the performance of predictive process modeling for large datasets.

Advances in Geographical Information Systems (GIS) and Global Positioning Systems (GPS) enable accurate geocoding of locations where scientific data are collected. This has encouraged collection of large spatial datasets in many fields and has generated considerable interest in statistical modeling for location-referenced spatial data. The setting where the number of locations yielding observations is too large to fit the desired hierarchical spatial random effects models using Markov chain Monte Carlo methods is considered. This problem is exacerbated in spatial-temporal and multivariate settings where many observations occur at each location. The recently proposed predictive process, motivated by kriging ideas, aims to maintain the richness of desired hierarchical spatial modeling specifications in the presence of large datasets. A shortcoming of the original formulation of the predictive process is that it induces a positive bias in the non-spatial error term of the models. A modified predictive process is proposed to address this problem. The predictive process approach is knot-based leading to questions regarding knot design. An algorithm is designed to achieve approximately optimal spatial placement of knots. Detailed illustrations of the modified predictive process using multivariate spatial regression with both a simulated and a real dataset are offered.

A Multi-Level, Mobile-Enabled Intervention to Promote Physical Activity in Older Adults in the Primary Care Setting (iCanFit 2.0): Protocol for a Cluster Randomized Controlled Trial.

BACKGROUND: Most older adults do not adhere to the US Centers for Disease Control physical activity guidelines; their physical inactivity contributes to overweight and multiple chronic conditions. An urgent need exists for effective physical activity-promotion programs for the large number of older adults in the United States. OBJECTIVE: This study presents the development of the intervention and trial protocol of iCanFit 2.0, a multi-level, mobile-enabled, physical activity-promotion program developed for overweight older adults in primary care settings. METHODS: The iCanFit 2.0 program was developed based on our prior mHealth intervention programs, qualitative interviews with older patients in a primary care clinic, and iterative discussions with key stakeholders. We will test the efficacy of iCanFit 2.0 through a cluster randomized controlled trial in six pairs of primary care clinics. RESULTS: The proposed protocol received a high score in a National Institutes of Health review, but was not funded due to limited funding sources. We are seeking other funding sources to conduct the project. CONCLUSIONS: The iCanFit 2.0 program is one of the first multi-level, mobile-enabled, physical activity-promotion programs for older adults in a primary care setting. The development process has actively involved older patients and other key stakeholders. The patients, primary care providers, health coaches, and family and friends were engaged in the program using a low-cost, off-the-shelf mobile tool. Such low-cost, multi-level programs can potentially address the high prevalence of physical inactivity in older adults.

Work and Chronic Disease: Comparison of Cardiometabolic Risk Markers Between Truck Drivers and the General US Population.

OBJECTIVE: US long-haul truck drivers experience a wide array of excess cardiometabolic disease (CMD) risks unique to their occupation. How these risks translate to, and potentially induce, elevations in the clinical CMD risk profile of this population is unknown. METHODS: A non-experimental, descriptive, cross-sectional design was employed to collect anthropometric and biometric data from 115 long-haul truckers to generate for the first time a comprehensive CMD risk marker profile, which was then compared with the general US population. The relationships between CMD risk markers and CMD outcomes were examined for both populations. RESULTS: The long-haul trucker sample presented elevated CMD risk markers, generally scoring significantly worse than the general population. Associations between CMD risk markers and disease states varied between both populations. CONCLUSIONS: US long-haul truck drivers' distinctive CMD risk profile indicates occupationally-linked CMD pathogenesis.

Gaussian predictive process models for large spatial data sets.

With scientific data available at geocoded locations, investigators are increasingly turning to spatial process models for carrying out statistical inference. Over the last decade, hierarchical models implemented through Markov chain Monte Carlo methods have become especially popular for spatial modelling, given their flexibility and power to fit models that would be infeasible with classical methods as well as their avoidance of possibly inappropriate asymptotics. However, fitting hierarchical spatial models often involves expensive matrix decompositions whose computational complexity increases in cubic order with the number of spatial locations, rendering such models infeasible for large spatial data sets. This computational burden is exacerbated in multivariate settings with several spatially dependent response variables. It is also aggravated when data are collected at frequent time points and spatiotemporal process models are used. With regard to this challenge, our contribution is to work with what we call predictive process models for spatial and spatiotemporal data. Every spatial (or spatiotemporal) process induces a predictive process model (in fact, arbitrarily many of them). The latter models project process realizations of the former to a lower dimensional subspace, thereby reducing the computational burden. Hence, we achieve the flexibility to accommodate non-stationary, non-Gaussian, possibly multivariate, possibly spatiotemporal processes in the context of large data sets. We discuss attractive theoretical properties of these predictive processes. We also provide a computational template encompassing these diverse settings. Finally, we illustrate the approach with simulated and real data sets.