Predicting geographical variation in health-related quality of life

Multiscale Impact of Environmental and Socio-Economic Factors on Low Physical Fitness among Chinese Adolescents and Regionalized Coping Strategies

As low physical fitness in adolescents increases their risk of all-cause mortality in future adulthood as well as regional public health budgets, many scholars have studied the factors influencing physical fitness in adolescents. However, the spatial non-stationarity and scale between physical fitness and influencing factors in adolescents are often neglected. To rectify this situation, this study constructed a multi-scale geographically weighted regression model based on data from the China National Student Fitness Survey and the China Statistical Yearbook in 2018 to investigate the spatial patterns of factors influencing low physical fitness among adolescents. The results showed that the influencing factors for measuring the physical fitness of Chinese adolescents had significant spatial heterogeneity and multi-scale effects. The local R² values were relatively low in the western region of China. Consideration should be given to increasing the lifestyle and ethnic and cultural characteristics of local residents when selecting influencing factors in the future. The physical fitness of men was mainly influenced by socio-economic factors, while that of women was influenced by natural environmental factors. According to the different spatial distribution patterns of MGWR, this study suggests that each region should develop regionalized strategies to cope with the low physical fitness of adolescents, including taking advantage of the natural environment to develop physical fitness promotion projects, accelerating the upgrading of industrial structures in the north-eastern and western regions, and the need to remain cautious of rapid urbanization in the east.

Evaluating the Impact of Social and Built Environments on Health-Related Quality of Life among Cancer Survivors

BACKGROUND

With almost 17 million U.S. cancer survivors, understanding multilevel factors impacting health-related quality of life (HRQOL) is critical to improving survivorship outcomes. Few studies have evaluated neighborhood impact on HRQOL among cancer survivors.

METHODS

We combined sociodemographic, clinical, and behavioral data from three registry-based studies in California. Using a three-level mixed linear regression model (participants nested within block groups and study/regions), we examined associations of both independent neighborhood attributes and neighborhood archetypes, which capture interactions inherent among neighborhood attributes, with two HRQOL outcomes, physical (PCS) and mental (MCS) composite scores.

RESULTS

For the 2,477 survivors, 46% were 70+ years, 52% were non-Hispanic White, and 53% had localized disease. In models minimally adjusted for age, stage, and cancer recurrence, HRQOL was associated with neighborhood socioeconomic status (nSES), racial/ethnic composition, population density, street connectivity, restaurant environment index, traffic density, urbanicity, crowding, rental properties, and non-single family units. In fully adjusted models, higher nSES remained associated with better PCS, and restaurant environment index, specifically more unhealthy restaurants, associated with worse MCS. In multivariable-adjusted models of neighborhood archetype, compared with upper middle-class suburb, Hispanic small town and inner city had lower PCS, and high status had higher MCS.

CONCLUSIONS

Among survivors, higher nSES was associated with better HRQOL; more unhealthy restaurants were associated with worse HQROL. As some neighborhood archetypes were associated with HRQOL, they provide an approach to capture how neighborhood attributes interact to impact HRQOL.

IMPACT

Elucidating the pathways through which neighborhood attributes influence HRQOL is important in improving survivorship outcomes.

Food, Big Data, and Decision-making: a Scoping Review-the 3-D Commission

Food is an important determinant of health, featuring prominently in the Sustainable Development Goals. The term "big data" is seldom used in relation to food, partly because food data are scattered across different sectors. The increasing availability of food-related data presents an opportunity to glean new insights on food and food systems. These insights may enhance the quality of products and services and improve decision-making on optimizing food availability, all to the end of producing better health. Yet, knowledge gaps remain about the unique opportunities and challenges linked to big data on food and their use in decision-making. This scoping review explored the available literature linking food with big data and decision-making, using the following research question: What is the current literature on data about food, and how are these data used in decision-making? We searched PubMed until 29 February 2020 and Embase, Web of Sciences, and the Cochrane Database of Systematic Reviews until 8 March 2020. We included studies written in English and conducted narrative analyses to identify relevant themes from included studies. Sixteen studies fulfilled our eligibility criteria, including big data analyses, modelling studies, and reviews. These studies described the added value of using big data and how evidence from big data had or can be used for decision-making, as well as challenges and opportunities for such use. The majority of the included studies examined the link between food and big data, while hypothesizing of how these insights could inform decision-making, including policies, interventions, programs, and financing. There were only two examples wherein big data on food informed decision-making directly. The review highlights several false dichotomies in how the subject is approached in the literature and the importance of context, both between and within countries, in shaping the availability and types of data that can be used as meaningful evidence to inform decision-making. This review shows the paucity of research around the intersection of food, big data, and decision-making, as well as the potential in using big data on food systems to the end of informing decisions to improve the health of populations. Future research and decision-making around health systems can benefit from examining the full spectrum of perspectives on the subject. Future research and decision-making around health systems can also employ the steadfast embrace of technology, which will potentially reduce disparities in big data availability, to the end of improving the health of populations.

Geographically weighted machine learning model for untangling spatial heterogeneity of type 2 diabetes mellitus (T2D) prevalence in the USA

Type 2 diabetes mellitus (T2D) prevalence in the United States varies substantially across spatial and temporal scales, attributable to variations of socioeconomic and lifestyle risk factors. Understanding these variations in risk factors contributions to T2D would be of great benefit to intervention and treatment approaches to reduce or prevent T2D. Geographically-weighted random forest (GW-RF), a tree-based non-parametric machine learning model, may help explore and visualize the relationships between T2D and risk factors at the county-level. GW-RF outputs are compared to global (RF and OLS) and local (GW-OLS) models between the years of 2013-2017 using low education, poverty, obesity, physical inactivity, access to exercise, and food environment as inputs. Our results indicate that a non-parametric GW-RF model shows a high potential for explaining spatial heterogeneity of, and predicting, T2D prevalence over traditional local and global models when inputting six major risk factors. Some of these predictions, however, are marginal. These findings of spatial heterogeneity using GW-RF demonstrate the need to consider local factors in prevention approaches. Spatial analysis of T2D and associated risk factor prevalence offers useful information for targeting the geographic area for prevention and disease interventions.