Purpose: To examine differential associations between health literacy (HL) and end-of-life (EOL) care expenditures by rurality. Methods: This cross-sectional study included all urban and rural counties in the United States. County-level HL data were estimated using 2010 US Census and 2011 American Community Surveys data; EOL expenditures in 2010 were derived from the Dartmouth Atlas of Health Care database. Hierarchical generalized linear regressions were used to assess associations between HL and EOL care, controlling for county-level characteristics and focusing on rurality (with areas classified as urban, rural micropolitan, or rural noncore). Findings: Of 3,137 US counties, 100 (3.2%) counties where 7.6 million Americans live had low HL (LHL). Counties with LHL had significantly higher average expenditures in the last 6 months of life and during terminal hospitalization than counties with high HL (HHL) (both P <.001). There was a statistically significant interaction between HL and rurality (P <.001). EOL expenditures were significantly higher in LHL counties than HHL counties in urban areas, while no such relationship appeared in rural areas. Average estimated EOL expenditures among LHL counties decreased by rurality ($16,953, $14,939, and $12,671 for urban, rural micropolitan, and rural noncore areas, respectively), while average estimated expenditures in HHL counties were around $14,000 in each of these areas. Conclusions: HL and EOL expenditures were inversely associated with urban America but unrelated to rural areas. Counties with HHL had constant expenditures regardless of rurality. Interventions targeting HL may help reduce EOL expenditures and rural-urban disparities in EOL care.

The COVID-19 pandemic upended life for countless families and brought heightened attention to the plight of working parents. Mothers with young and school-aged children have borne the brunt of the care crisis in the United States, as schools and daycare centers across the country closed. Compared to fathers, mothers have been more likely to exit the labor force prematurely, become unemployed, and reduce their work hours during the pandemic (Heggeness 2020; Landivar et al. 2020).

The opioid epidemic has had devastating effects for the health and wellbeing of the U.S. population. However, we know little about how it has affected the structure of families where children are raised. Using the 5 percent sample of the 2000 Census, 2005-2018 American Community Survey (ACS) data and restricted Vital Statistics we assess the effect of the opioid epidemic at the local level on the rates of children living under different types of family arrangements: two married parents, two cohabiting parents, mother only, father only, or another configuration. Local fixed-effects models show that a greater intensity of the opioid epidemic, as measured by a higher opioid-overdose death rate, is associated with a lower rate of children living with two married parents, and a higher rate of children living with two cohabiting parents, with only a father, and with adults other than their parents. The opioid epidemic appears to increase the rates of children living in family structures that tend to be less stable, which has potential longterm implications for the wellbeing of future generations.

In the wake of the Affordable Care Act, some US states expanded Medicaid eligibility to low-income, working-age adults while others did not. This study investigates whether this divergence induces migration across state borders to obtain Medicaid, especially in border regions. It compares border with interior regions' in-migration in the concerned subgroup before and after the Medicaid expansion in linear probability difference-indifference and triple difference regression frameworks. Using individual-level data from the American Community Surveys over 2012-2017, this study finds a tendency towards increases in in-migration to expansion states' border regions after the expansion. The odds of having migrated increase by about 34 % in these regions after the Medicaid expansion compared to before and control regions. However, this additional migration increases the number of Medicaid-eligible working age adults by less than 1.5 % in border regions. If all additional migrants take up Medicaid, the number of Medicaid beneficiaries in these regions increases by approximately 4 %. Thus, the concerned individuals might value Medicaid, but the induced migration appears unlikely to impose meaningful fiscal externalities at the regional level.

Economic disruption related to coronavirus disease 2019 (COVID-19) continued through the summer of 2020, affecting the lives of millions of Americans. In this visualization, the authors use recent data from the Current Population Survey to examine Americans’ cumulative risk for labor force detachment during the pandemic. The individuals in the analysis were interviewed eight times: in April, May, June, and July of 2019 and 2020. The authors document respondents’ employment experiences during the 2020 pandemic, using the 2019 data points as a baseline for comparison. Increasing detachment from the labor force varies by basic demographic characteristics (gender and parental status), but a more important divide in the COVID-19 economy is education, an already fundamental determinant of Americans’ life chances. The educational divide is especially pronounced among parents, with important repercussions for inequalities among children.

Fertility rates have been declining worldwide over the past 50 years, part of a phenomenon known as “the demographic transition.” Prior work suggests that this decline is related to population density. In the present study, we draw on life history theory to examine the relationship between population density and fertility across 174 countries over 69 years (1950 to 2019). We find a robust association between density and fertility over time, both within and between-countries. That is, increases in population density are associated with declines in fertility rates, controlling for a variety of socioeconomic, socioecological, geographic, population-based, and female empowerment variables. We also tested predictions about environmental boundary conditions. In harsher living conditions (e.g., higher homicide or pathogen rates), the effect of increased population density on fertility rates was attenuated. The density-fertility association was also moderated by religiousness and strength of social norms, where the relationship between density and fertility was attenuated in countries with high religiosity and strong social norms. We discuss why and when changes in population density may influence fertility rates and the broader implications of this work.

This paper empirically investigates the long-run effects of major health improvements on income growth in the United States. To isolate exogenous changes in health, the econometric model uses quasi-experimental variation in cardiovascular disease mortality across states over time. Based on data for the white population, the results show that there is a causal link between health and income per person, and they provide novel evidence that health dynamics shape life-cycle incomes. Life-cycle income profiles slope more strongly at the beginning and at the end of work life in 2000 than in 1960, indicating that age becomes a more prominent determinant of income dynamics over this period. The channels for this transformation include better health, higher educational attainment, and changing labor supply.

One important strategy to increase vaccination coverage is to minimize missed opportunities for vaccination. Missed opportunities for simultaneous vaccination (MOSV) occur when a child receives one or more vaccines but not all those for which they are eligible at a given visit. Household surveys that record children’s vaccination dates can be used to quantify occurrence of MOSVs and their impact on achievable vaccination coverage. We recently automated some MOSV analyses in the World Health Organization’s freely available software: Vaccination Coverage Quality Indicators (VCQI) making it straightforward to study MOSVs for any Demographic &amp; Health Survey (DHS), Multi-Indicator Cluster Survey (MICS), or Expanded Programme on Immunization (EPI) survey. This paper uses VCQI to analyze MOSVs for basic vaccine doses among children aged 12–23 months in four rounds of DHS in Colombia (1995, 2000, 2005, and 2010) and five rounds of DHS in Nigeria (1999, 2003, 2008, 2013, and 2018). Outcomes include percent of vaccination visits MOSVs occurred, percent of children who experienced MOSVs, percent of MOSVs that remained uncorrected (that is, the missed vaccine had still not been received at the time of the survey), and the distribution of time-to-correction for children who received the MOSV dose at a later visit.

Latino children are one of Massachusetts' fastest-growing segments of the population. However, evidence suggests that the social and economic context in which Latino children live does not adequately support their development and overall wellbeing. Nearly a third of Latino children in the United States live in very low-opportunity neighborhoods as defined by a scale of educational, health, environmental, and socioeconomic outcomes. Compared to non-Latino children, Latinos are more likely to grow up in households below the federal poverty threshold2 and less likely to have a mother with at least a Bachelor's degree.

Climate change has increased heat exposure in many parts of the tropics, negatively impacting outdoor worker productivity and health. Although it is known that tropical deforestation is associated with local warming, the extent to which this additional heat exposure affects people across the tropics is unknown. In this modeling study, we combine worker health guidelines with satellite, reanalysis, and population data to investigate how warming associated with recent deforestation (2003–2018) affects outdoor working conditions across low-latitude countries, and how future global climate change will magnify heat exposure for people in deforested areas. We find that the local warming from 15 years of deforestation was associated with losses in safe thermal working conditions for 2.8 million outdoor workers. We also show recent large-scale forest loss was associated with particularly large impacts on populations in locations such as the Brazilian states of Mato Grosso and Pará. Future global warming and additional forest loss will magnify these impacts.

Purpose: In recent years, breast cancer detection and treatment have advanced. As a result, increased attention to breast cancer survivorship should have improved their health-related quality of life (HRQoL). Our aim was to examine the trends in the HRQoL of female breast cancer survivors between 2008 and 2016, to determine whether or not the increased focus on survivorship has translated into improved HRQoL. Furthermore, stratified analyses were conducted by race/ethnicity and age group and these were compared to a similar group of women without a breast cancer history. Methods: Repeated cross-sectional analyses using the Medical Expenditure Panel Survey between 2008 and 2016 were conducted. Pooled ordinary least squares (OLS) regression was used to examine the trends in physical component scores (PCS-12) and mental component scores (MCS-12) among breast cancer survivors and a similar population of women without a breast cancer history. Analyses stratified by race/ethnicity and age group were also conducted. Results: Among breast cancer survivors, after adjusting for confounders, there was no change in PCS-12 scores over time, but the MCS-12 scores increased by 0.27 points (95% CI 0.09–0.45). Those without a history of breast cancer had mean PCS-12 scores that were 0.13 points greater each year (95% CI 0.02–0.24) while their mean MCS-12 scores were 0.10 (95% CI 0.00–0.21) points greater each year. After stratifying by race/ethnicity, Hispanic breast cancer survivors had a small increase in PCS-12 (β: 0.65; 95% CI 0.01–1.29), and MCS-12 scores (β: 0.70; 95% CI 0.06–1.33) over time. Similar small effects were found when stratified by age group, both among breast cancer survivors and those without a history of breast cancer. The younger age group (< 50 years) reported poorer MCS-12 than the older population (age 50 years and above). Conclusion: Our study generated findings showing the trends in the HRQoL of breast cancer survivors and compared these to a similar population of women without a history of breast cancer. This paper highlights the importance of focusing on the mental health of young breast cancer survivors to improve their prospects at a good quality of life post-breast cancer diagnosis and treatment.

Recent work suggests that the area of the United States in which a child is raised has a substantial effect on their income in adulthood. We measure differences in time use between areas which are better or worse at producing incomes in adulthood. The main differences are that, in areas which produce higher incomes, people spend more time at work, and adults spend more time with children. The data does not support some theories of what makes communities effective at producing human capital: People do not spend more time on educational activities, or on community events and institutions, in areas which increase incomes by more.

Background: This study compared drug overdose mortality rates in Puerto Rican-heritage and Non-Hispanic (NH) White individuals in the United States (US), examining time trends and recent variation by age, sex, state of residence, and drugs involved in overdose. Methods: Death certificate data from the National Center for Health Statistics, as well as American Community Survey population estimates, were used to calculate age-specific and age-adjusted drug overdose mortality rates for Puerto Rican-heritage and NH White residents of the 50 United States or District of Columbia (DC). Rates for 2018 were compared between Puerto Rican-heritage and NH White individuals, overall and by sex, age, state, and specific drug involved in overdose. Joinpoint Regression was used to examine trends in drug overdose mortality rates from 2009 to 2018. Results: From 2009 to 2018, the age-adjusted drug overdose mortality rate in stateside Puerto Ricans doubled among women (from 6.0 to 12.5 per 100,000) and nearly tripled among men (from 15.3 to 45.2 per 100,000). In 2018, the age-adjusted drug overdose mortality rate was significantly higher in Puerto Rican-heritage than NH White individuals (28.7 vs. 26.2 per 100,000, respectively). The 2018 drug overdose mortality rate was highest among Puerto Rican-heritage men ages 45–54 (104.1 per 100,000). Conclusion: Findings emphasize the necessity of policies, programs, and interventions to mitigate risk of fatal overdose in stateside Puerto Rican communities.

Child undernutrition remains high in India with far-reaching consequences for child health and development. Anthropometry reflects undernutrition. We examined the state-level trends in underweight, stunting, and wasting prevalence and inequality by living standards using four rounds of the National Family Health Surveys in 26 states in India, conducted in 1992–1993, 1998–1999, 2005–2006, and 2015–2016. The average annual reduction (AAR) for underweight ranged from 0.04 percentage points (pp) (95% CI − 0.12, 0.20) in Haryana to 1.05&nbsp;pp (95% CI 0.88, 1.22) in West Bengal for underweight; 0.35&nbsp;pp (95% CI 0.11, 0.59) in Manipur to 1.47 (95% CI 1.19, 1.75) in Himachal Pradesh for stunting; and − 0.65&nbsp;pp (95% CI − 0.77, − 0.52) in Haryana to 0.36&nbsp;pp (95% CI 0.22, 0.51) in Bihar &amp; Jharkhand for wasting. We find that change in the pp difference between children with the poorest and richest household living standards varied by states: statistically significant decline (increase) was observed in 5 (3) states for underweight, 5 (4) states for stunting, and 2 (1) states for wasting. Prevalence of poor anthropometric outcomes as well as disparities by states and living standards remain a problem in India.

Real-world data is dirty, which causes serious problems in (supervised) machine learning (ML). The widely used practice in such scenario is to first repair the labeled source (a.k.a. train) data using rule-, statistical- or ML-based methods and then use the “repaired” source to train an ML model. During production, unlabeled target (a.k.a. test) data will also be repaired, and is then fed in the trained ML model for prediction. However, this process often causes a performance degradation when the source and target datasets are dirty with different noise patterns, which is common in practice. In this paper, we propose an adaptive data augmentation approach, for handling missing data in supervised ML. The approach extracts noise patterns from target data, and adapts the source data with the extracted target noise patterns while still preserving supervision signals in the source. Then, it patches the ML model by retraining it on the adapted data, in order to better serve the target. To effectively support adaptive data augmentation, we propose a novel generative adversarial network (GAN) based framework, called DAGAN, which works in an unsupervised fashion. DAGAN consists of two connected GAN networks. The first GAN learns the noise pattern from the target, for target mask generation. The second GAN uses the learned target mask to augment the source data, for source data adaptation. The augmented source data is used to retrain the ML model. Extensive experiments show that our method significantly improves the ML model performance and is more robust than the state-of-the-art missing data imputation solutions for handling datasets with different missing value patterns.

Space heating is the single greatest source of building-related greenhouse gas emissions in the industrialized world, giving urgency to the development of strategies for carbon-free heating. Recent advances have shown that the direct capture, storage, and deployment of solar energy, without conversion to electricity, has considerable potential to address space-heating needs even in cold and cloudy climates. However, the solar energy available for direct heating at climatic and metropolitan scales is both unquantified and widely assumed to be negligible, impeding further investigation, development, and policy responses. To estimate the magnitude and distribution of solar resources concurrent with space-heating needs, we spatially integrate datasets characterizing solar radiation, outdoor temperature, and heating energy use across U.S. climates. Results show that the median resource incident upon collectors of residential scale (10m2) and distribution is much greater than previously realized, equaling 7MWh per household annually; by comparison, the median household heating need is currently 10.3MWh. Unexpectedly, cloud-diffused solar radiation accounts for over one-quarter of this resource in all but semi-arid climates. Metropolitan residential resources exceed 5TWh in areas including Detroit and Boston (cold continental), WashingtonD.C. (humid subtropical), Seattle and San Francisco (Mediterranean), and Denver (semi-arid), and national resources exceed 750TWh annually, compared to approximately 1200TWh of annual heating need. Current technology is able to capture and retain over half of a direct solar heating resource, revealing that the untapped U.S. solar heating potential is comparable to one-third of the national residential space-heating need and implying that analogous resources exist in analogous climates worldwide.

This paper develops a method to dynamically model urban passenger mode trade-offs at fine-grained spatial and temporal scales using data from OpenTripPlanner (OTP) and the City of Chicago's Transportation Network Providers (TNP) dataset. This approach can be used to calculate dynamic modal cost-distance trade-offs for specific times, routes, and geographic areas of interest, providing a framework for creating aggregate mode choice profiles for individual cities and neighbourhoods that can be used to assess structural differences in transportation investment and mobility, as well as to test various assumptions about travel behaviour, observe temporal changes in modal trade-offs, and model the system-wide implications of changes to the transportation system to modal trade-offs. Using this dynamic mode choice framework, this paper explores the features underlying observed structural heterogeneity in the ratio of cost to distance (i.e., speed or potential mobility) for observed flows across the city for each mode. It finds that Census tracts with larger proportions of Black and Hispanic population tend to have significantly larger cost-distance ratios (i.e., slower speeds/lower potential mobility) for non-auto modes, while Census tracts with higher proportions of “creative class” employment and features of walkable built environments have significantly lower cost-distance ratios (i.e., faster speeds/higher potential mobility).

Telling the story of America’s babies is more important than ever. The State of Babies Yearbook: 2021 shows that, even before the COVID-19 pandemic, the littlest among us did not have the supports they need to thrive. It also shows that in America, racial and economic inequities start even before birth. An unacceptable number of infants and toddlers—2 in 5—lived in families whose income was inadequate to make ends meet. A concerning proportion, especially among babies of color, did not have preventive medical care, adequate household food security, or safe and stable housing. These shortcomings, visible in previous years, presaged the havoc wrought by the pandemic on families’ stability, and created conditions that could undermine babies’ development. The pandemic’s impacts were predictable and some avoidable. Yet, for decades, our nation has stood on the sidelines while families juggled meeting economic and child caring needs amidst threadbare systems of support and while families of color faced systemic barriers to economic security. Our nation’s lack of strong, permanent policies that recognize families’ dual roles of participating in the economy and nurturing their children weakened families’ ability to withstand the additional hardships imposed by the pandemic.

Studies on socio-economic impacts of climate and energy policies tend to focus on income and expenditure ef-fects. For analyses that go beyond monetary dimensions, time-diary data have proven to be useful. Here we investigate how work time relates to leisure activity structures and associated energy use for different types of employees. To this end, an analysis of time-use data is undertaken for Finland and France. Novel elements are (1) a differentiation between part-time and full-time employees, (2) the use of distinct energy intensities of different activities by household type instead of average energy intensities, and (3) allowing for non-linear relationships between work time and the allocation of other activities. Our results suggest that the effects of work time on energy use are rather homogeneous in Finland, whereas we find more differences between employee types in France. In both countries, adjustment of leisure-activity duration is sometimes strong initially but flattening for longer work hours. This relates to another finding, namely that the composition of leisure activities differs be-tween people with distinct work hours. Our study suggests that analysis of disaggregated time-use data can add relevant insights for evaluation, and possibly design, of energy, climate and labour-market policies.

This report presents and reviews the available sources of data on public transit riders and ridership. We intend it to be a resource for those who manage or simply wish to understand U.S. transit. In conducting this review, we consider the advantages and disadvantages of publicly available data on transit from a variety of public and private sources. We consider as well the relatively scarcer and less available sources of data on other providers of shared mobility, like ride-hail services, that compete with and complement public transit, as well as pieces we see as missing from the transit analytics pie. We conclude by discussing how data gaps both align with existing inequities and enable them to continue, unmeasured, and how the COVID-19 pandemic has made closing these gaps all the more important.