Systematic Review & Meta-analysis

This scoping review investigates the connection between economic globalization and the obesity epidemic. The review meticulously analyzed 16 studies, encompassing various research designs including cross-sectional, longitudinal, and case-control studies. The majority of these studies revealed a positive correlation between aspects of economic globalization, such as market deregulation and the influx of modern food industries, and increased obesity rates. The review also highlights the unexplored avenues of globalization's impact on obesity through modern technology adoption and cultural shifts, calling for future empirical research in these areas. This study sheds light on the complex relationship between global economic trends and public health challenges like obesity.

In this systematic review and meta-analysis, we explore the influence of sugar-sweetened beverage warning labels on consumer behavior. Collating data from 23 studies, including randomized trials and nonrandomized experiments, the review provides a comprehensive analysis of different label types – ranging from text and graphic-based to nutrient and health effect-focused labels. The findings reveal that warning labels significantly reduce the odds of consumers choosing sugar-sweetened beverages, with labels that combine graphics and health effects proving most effective. This research not only underscores the power of warning labels in influencing consumer choices but also highlights the need for further studies to understand the underlying psychological mechanisms and to test these findings in diverse, real-world environments.

This systematic review examines the relationship between beef consumption and cognitive function in children and young adults. The study involved a thorough literature search across seven databases, identifying eight studies that reported on five distinct interventions in countries such as Kenya, the U.S., Guatemala, Pakistan, the Democratic Republic of the Congo, and Zambia. The findings are mixed: while one intervention with a non-feeding control arm suggested beef consumption could enhance cognitive abilities, others comparing beef to different food types showed inconsistent results. The review concludes that current evidence is limited and varied, emphasizing the need for future research with representative samples, longer follow-up periods, and comprehensive nutrient intake measurements to better understand the impact of beef consumption on cognitive development.

This systematic review explores the impact of ambient air pollution on physical activity and sedentary behavior in China. The review involved a comprehensive search in major databases like PubMed and Web of Science, adhering to strict selection criteria focusing on different study designs and subjects across mainland China. Out of the ten studies that met these criteria, six used a cross-sectional design and four a prospective cohort design. The review found that deteriorating air quality and increased PM2.5 levels were associated with reduced outdoor physical activities and increased sleeping durations among Chinese residents. However, the relationship between air pollution and sedentary behavior was found to be mixed and inconclusive. The study concludes that while there is preliminary evidence of ambient air pollution affecting physical activity behaviors in China, further research using objective measures and focusing on sensitive sub-populations is necessary to deepen the understanding of these impacts.

This systematic review and meta-analysis investigates the relationship between adiposity and outcomes following coronary artery bypass grafting (CABG). The team conducted an extensive search in databases like PubMed and Web of Science, identifying 72 studies that met their criteria focusing on adult CABG patients and outcomes like readmissions and mortality rates. The meta-analysis revealed that overweight patients had a 30% lower odds of post-CABG readmission and a 20% lower odds of mid-to-long-term mortality compared to their normal-weight counterparts. However, it found no significant difference in readmission and mortality rates between obese and normal-weight patients. These findings suggest that overweight patients might have better post-CABG outcomes than their normal-weight counterparts, indicating that preoperative weight loss may not be beneficial for overweight individuals undergoing CABG.

This systematic review and meta-analysis evaluates the effectiveness of prebiotic, probiotic, and synbiotic therapies in treating nonalcoholic fatty liver disease (NAFLD). After a thorough search in PubMed and EMBASE following PRISMA guidelines, 25 studies involving 1309 patients were analyzed. The meta-analysis revealed that microbial therapies significantly improved key health indicators, including reductions in BMI, hepatic enzymes (ALT, AST, and γ-GT), and serum cholesterol levels, although they had a less marked effect on inflammation markers. While the results show promising outcomes for the use of these therapies in managing NAFLD, the study also emphasizes the need for further research considering the limitations of current biomarkers and the personalized nature of microbial-based treatments.

This systematic review and meta-analysis examines the link between chronic noise exposure and adiposity. The team reviewed literature from PubMed, Web of Science, and the Cochrane Library, identifying eleven studies that investigated this relationship. The findings revealed a positive association between chronic noise exposure and increased waist circumference in adults, with exposure above 55-60 dBA correlated with a significant yearly increase in waist size. However, no association was found with BMI. These results suggest that while chronic noise exposure might influence specific aspects of adiposity like waist circumference, the overall evidence is still limited. The study concludes that further research is needed to explore the impact of noise in different settings, across diverse populations, and to understand the underlying mechanisms connecting noise exposure to adiposity.

This systematic review and meta-analysis assesses the glycemic impact of non-nutritive sweeteners (NNSs) through randomized controlled trials. The study, adhering to PRISMA guidelines, analyzed 29 trials involving 741 participants to determine the effect of NNSs like aspartame, saccharin, steviosides, and sucralose on blood glucose levels. The meta-analysis revealed that NNS consumption did not increase blood glucose levels, which instead showed a gradual decline post-consumption. While the glycemic impact did not vary significantly across different types of NNSs, it showed some variation based on participants' age, weight, and diabetic status. The study concludes that NNS consumption does not elevate blood glucose levels, highlighting the need for further research to understand the long-term health implications and biological mechanisms of NNSs.

This systematic review and meta-analysis evaluates the effects of isolated soluble fiber supplementation on body weight and metabolic outcomes in adults with overweight and obesity. The study meticulously reviewed 12 randomized controlled trials (RCTs) involving 609 participants, focusing on body composition (BMI, body weight, body fat percentage, waist circumference) and glucose and insulin metabolism. The results indicated that soluble fiber supplementation significantly reduced BMI, body weight, body fat, fasting glucose, and fasting insulin levels compared to placebo treatments. Despite considerable heterogeneity across studies, the findings suggest that isolated soluble fiber supplementation could be a viable strategy to improve body composition and metabolic health in overweight and obese individuals. The study calls for cautious interpretation of these results due to the variability between the studies included.

This systematic review and meta-analysis investigates the impact of nutrition labels on the dietary quality of college students, a group at heightened risk for poor nutrition. Analyzing 22 randomized controlled trials, cohort studies, and pre-post studies, the review found that exposure to nutrition labels was generally associated with improved dietary choices among college students. Specifically, of the studies examining caloric intake, the majority indicated that label posting at the point of purchase led to a decrease in calories consumed. Furthermore, nutrition labels were found to positively affect diet quality in terms of non-caloric measures in most studies. The meta-analysis of pre-post studies highlighted a significant decrease of 36 kcal with label exposure. These findings underscore the moderate yet positive influence of nutrition labels on the dietary habits of college students.

This systematic review delves into the intricate relationship between global warming and the obesity epidemic, two of the most critical challenges facing modern society. The review, which included fifty studies from databases like PubMed and Web of Science, categorizes the findings into four distinct relationships: shared drivers for both phenomena, global warming's influence on obesity, obesity's impact on global warming, and the bidirectional influence between the two. The study proposes a conceptual model where factors like the fossil fuel economy, population growth, and industrialization affect both global warming (through increased greenhouse gas emissions) and obesity (via nutrition transition and reduced physical activity). Additionally, global warming directly influences obesity through factors like food supply shocks and adaptive thermogenesis, while obesity contributes to global warming due to higher energy consumption. The review highlights the importance of policies promoting clean energy and active lifestyle urban designs to mitigate both global warming and obesity.

This systematic review and dose-response meta-analysis assesses the relationship between tomato consumption and prostate cancer (PCa) risk. The study, which gathered data from 30 studies involving 24,222 PCa cases and 260,461 participants, found that higher total tomato consumption was significantly associated with a reduced risk of PCa. The analysis specifically highlighted that cooked tomatoes and sauces were linked to a reduced risk, whereas no significant association was observed for raw tomatoes. The meta-analysis also identified a dose-response relationship, indicating that increasing tomato consumption, particularly of cooked forms, is inversely related to PCa risk. These findings suggest the potential protective effects of tomatoes against PCa, with further research needed to understand the underlying mechanisms.

This study investigates the potential rise in childhood obesity due to the pandemic-induced lifestyle changes. The study utilized a microsimulation model to project changes in a nationally representative kindergarten cohort’s BMI z-scores and obesity prevalence from April 2020 to March 2021, under various COVID-19 scenarios including prolonged school closures and reduced physical activity. The findings indicate a significant increase in mean BMI z-scores and childhood obesity prevalence across all scenarios, with the greatest impact observed in the most prolonged disruption scenario. The study also noted a slightly larger impact on boys and non-Hispanic blacks and Hispanics. These results underscore the urgent need for public health interventions to encourage active lifestyles among children to counteract the pandemic's negative effects on childhood obesity.

Our extensive research has thoroughly investigated the implications of neighborhood food environments on dietary intake and obesity across various age groups in the US. Through detailed analyses and geospatial modeling of nationally representative cross-sectional and longitudinal surveys, we consistently found no significant correlation between the proximity to different types of food outlets and the quality of diet or obesity levels. This challenges the prevailing assumptions that link neighborhood food access to diet and health outcomes. Our findings indicate that greater exposure to fast-food restaurants, convenience stores, or supermarkets does not consistently affect dietary habits or Body Mass Index (BMI) in children, adolescents, or adults. These pivotal conclusions, replicated across several studies, have been referenced in legislative briefings, underscoring the need to reconsider resource allocation strategies aimed at addressing food deserts, especially under budget constraints. This body of work emphasizes the complexity of the relationship between food environments and public health and highlights the necessity of revisiting existing assumptions in obesity prevention strategies.

Our research delves into the impact of neighborhood built environments on physical activity, revealing a significant correlation with the availability of neighborhood amenities and increased activity levels across diverse demographics. Drawing from studies in both China and the U.S., we observed key disparities in access to recreational facilities and their influence on physical activity. For instance, in rural China, residents reported lower physical activity levels and lesser satisfaction with local exercise facilities compared to urban areas, indicating a need for policy interventions to improve rural access. Our findings also showed that proximity to exercise facilities in China positively correlates with leisure-time physical activity, especially among urban, educated, and younger populations. Similarly, in the U.S., the presence of neighborhood amenities like parks and recreation centers was linked to increased physical activity among children with special health care needs, highlighting the importance of accessible recreational spaces for promoting physical activity in vulnerable groups. These studies collectively underscore the need for improving neighborhood amenities to foster an active environment.

Our research focuses on the effects of PM2.5 air pollution on physical activity behaviors in the US and China, revealing significant impacts. In Beijing, studies involving university students and retirees showed that higher PM2.5 levels led to increased sedentary behavior and reduced physical activities like walking and vigorous exercise. For instance, an increase in PM2.5 concentration was associated with a marked rise in weekly hours of sedentary behavior among freshmen students and a decrease in weekly total hours of walking and leisure-time physical activity among university retirees. These findings were consistent across different demographic groups, with some variations between genders. Additionally, air pollution was found to increase sleep duration among students, further indicating its influence on daily routines. Our collaborative cohort studies in Beijing have contributed to international discussions and the development of WHO guidelines on Physical Activity and Air Pollution. These insights underscore the pressing need for policy interventions to mitigate air pollution's adverse effects on physical activity, particularly in urban areas with high pollution levels. Prof. Ruopeng An’s involvement in drafting these WHO reports highlights our commitment to addressing environmental challenges and promoting public health.

This study, utilizing a twin design to control for genetic and shared environmental factors, investigated the impact of parenting styles on obesity, smoking, and drinking behaviors in children and adolescents. Analyzing data from 631 pairs of same-sex monozygotic twins across three age cohorts (5, 11, and 17 years) from the German Twin Family Panel, the study explored five dimensions of parenting: emotional warmth, psychological control, negative communication, monitoring, and inconsistent parenting. The findings revealed that, after controlling for genetic influences, shared environmental factors, and initial body weight status, children who experienced harsher parenting in terms of communication had a lower BMI compared to their co-twin. Interestingly, this effect of negative communication was more pronounced in the youngest cohort and among female twins. The study also distinguished between paternal and maternal parenting styles, noting differing impacts on child weight. However, no substantial concurrent or long-term effects of paternal parenting on smoking and drinking behaviors were observed. These insights underline the nuanced role of parenting, particularly family communication, in shaping children's BMI and health behaviors.

This exploratory study focused on the early phase of the COVID-19 pandemic in the United States, aiming to understand the determinants and effects of various mitigation interventions implemented by states. It involved content analysis to identify nine key types of mitigation strategies and examined the timing of their enactment across states. Employing advanced statistical methods, including a proportional hazard model, a multiple-event survival model, and a random-effect spatial error panel model, the study analyzed the data, which included zero-inflated and skewed outcomes. Contrary to initial hypotheses, findings revealed that states with higher initial COVID-19 prevalence rates were slower in enacting mitigation strategies. Specifically, three strategies—closure of nonessential businesses, bans on large gatherings, and limitations on restaurants and bars—were found to be effective in reducing cumulative cases, new cases, and death rates. The study highlights that some states may have missed the optimal timing for implementing these mitigations, emphasizing the importance of prompt action. It also cautions against the risks associated with fully lifting these interventions too soon, underscoring the need for strategic planning in managing public health crises.

This study provides an in-depth analysis of the impact of Hurricane Katrina on the mental health of US adults. Utilizing data from 70,267 respondents in the Behavioral Risk Factor Surveillance System surveys spanning from 2004 to 2006, multilevel regression analyses were employed to assess the mental health consequences post-Katrina. The findings reveal a significant increase of 0.68 poor mental health days among residents in Katrina-affected states, particularly highlighting the profound impact on Louisiana residents compared to those in Mississippi. The study also identified vulnerable groups who suffered more severely, including women, young and middle-aged adults, lower-income individuals, and those with pre-existing suboptimal physical health. This research underlines the urgent need for targeted mental health interventions in the aftermath of disasters, especially for identified high-risk groups. It also calls for further studies to explore long-term mental health effects following such catastrophic events and to better understand the dynamics of risk and protective factors in disaster-related mental health outcomes.

This comprehensive study investigates the association between trade openness and obesity prevalence across 175 countries from 1975 to 2016. Using a robust dataset from the World Health Organization and the World Bank, the study employs two-way fixed-effects regressions to explore how the openness index (a measure of export and import relative to gross domestic product) correlates with national obesity rates. The findings reveal a significant positive correlation, indicating that a 10% increase in the openness index is associated with a 0.80% increase in the obesity rate. Notably, this relationship is most pronounced in Asian countries, followed by North America and Africa. When stratifying by income levels, the study highlights that the impact of trade openness on obesity is most acute in lower-middle-income countries, followed by upper-middle-income and low-income countries. Interestingly, no significant relationship is observed in high-income countries. These insights underscore the complex interplay between global economic factors and health outcomes, particularly in developing nations. Policymakers are urged to consider these findings in their strategies to mitigate the adverse health impacts of trade liberalization and the accompanying nutrition transition.

This study delves into the relationship between night-time eating and body weight status among US adults, utilizing comprehensive data from the 2007-2016 National Health and Nutrition Examination Survey (NHANES). The analysis encompassed data from 23,003 adult participants across five NHANES waves, assessing the impact of night-time consumption of energy, sugar, fat, and saturated fat on key indicators such as body mass index (BMI), waist circumference (WC), and prevalence of obesity and abdominal obesity. Intriguingly, the study found that approximately 36.5% of participants were obese and 56.7% had abdominal obesity. On average, a small proportion of daily energy and nutrient intake occurred during late-night hours (midnight to 5:59 AM and 10:00 PM to 11:59 PM). Contrary to common assumptions, the research revealed that night-time energy intake was not significantly associated with BMI, WC, obesity, or abdominal obesity. This finding persisted even when applying different time windows to define night-time eating. The study's results suggest that merely restricting food consumption at night, without reducing overall daily caloric intake, may not effectively prevent obesity.

This study explores the relationship between ambient air pollution and television usage among residents of Shanghai, China, offering a unique perspective on behavioral adaptations to environmental conditions. Data were collected from 300 households between January 2014 and December 2016, capturing daily average television use which was then correlated with air quality index (AQI) and weather data. The study employed an autoregressive moving-average model to analyze these relationships. Findings revealed a non-linear association between AQI and television viewing habits. Interestingly, television use decreased on days with fair (50<AQI≤100) and light air pollution (100<AQI≤150) compared to days with good air quality (0≤AQI≤50). However, on days with moderate-to-severe pollution levels (AQI>150), this trend did not hold. Additionally, television use was found to vary with other factors, decreasing with higher daytime temperatures, while increasing on weekends, holidays, and rainy days. The study suggests that in the face of mild air pollution, residents may choose to engage in other indoor activities or extend their sleep, rather than watching television.

This comprehensive study delves into the impact of weekend hospital admissions on in-hospital mortality among U.S. adults, spanning a decade from 2003 to 2013. Analyzing over 50 million hospital discharge records from the National Inpatient Sample, the study employed Cox regression models to evaluate the hazard ratios of in-hospital mortality linked to weekend admissions. Key findings highlight a 5% increased mortality risk for patients admitted during weekends compared to weekdays. The risk varied across demographic groups: young adults exhibited a lower incremental risk (2.7%) compared to middle-aged (5.3%) and older adults (5.2%). Notably, patients hospitalized for malignant neoplasms, diabetes mellitus, and heart diseases during weekends faced the highest incremental mortality risks, at 12.1%, 11.7%, and 8.2%, respectively. The study also uncovered that the additional mortality risk associated with weekend admissions was more pronounced in patients with higher initial mortality risk assessments. Encouragingly, the study revealed a declining trend in the weekend effect on mortality, decreasing from 6.9% in 2003 to 2.5% in 2013. This insightful research underscores the importance of continuous monitoring and improvement of healthcare services, particularly during weekends, to mitigate associated mortality risks.

This study explores the nutritional profiles of yogurt, frozen yogurt, and ice cream, including the impact of adding toppings, based on data from the 1999-2014 National Health and Nutrition Examination Survey. Focusing on the daily energy and nutrient intake of U.S. adults, the analysis involves 6453 individuals who reported consumption of these dairy products. Findings reveal that frozen yogurt and ice cream, especially when consumed with toppings, are significantly less healthy compared to plain yogurt. Specifically, frozen yogurt with and without toppings increases daily energy intake by 214.6 and 97.9kj, respectively, compared to yogurt. For ice cream, these increments are even higher, at 427.2 and 343.5kj, respectively. Furthermore, consuming frozen yogurt leads to a decreased intake of many vitamins and minerals, yet increases sugar, fat, cholesterol, fiber, and iron intake. Adding toppings to frozen yogurt exacerbates the intake of total and saturated fats. Ice cream consumption similarly results in lower micronutrient intake but raises consumption of sugar, fats, cholesterol, fiber, vitamins A and E, and iron, with added toppings further increasing sugar intake. This study underscores the lesser health benefits of frozen yogurt and ice cream compared to yogurt, particularly when toppings are included, highlighting the need for careful dietary choices among consumers.

We conducted a comprehensive evaluation of South Africa's HealthyFood program, a pioneering healthy food subsidy initiative funded by the country's largest private health insurance company. This nationwide program offered price rebates for purchasing healthy food items at a network of supermarkets. Employing rigorous quasi-experimental designs, including matched case-control difference-in-differences and instrumental variable methods, we aimed to assess the program's impact on dietary behaviors. Our findings revealed significant dietary improvements among participants, characterized by increased consumption of fruits, vegetables, and whole grains, and a decrease in the intake of foods high in sugar and salt, fried foods, processed meats, and fast food. To address potential self-selection bias in program participation, we employed an instrumental variable method and analyzed purchase data, which confirmed the efficacy of a 25% price rebate in promoting healthier food choices and consumption. The research highlights the effectiveness of financial incentives, such as substantial food subsidies, in enhancing diet quality across large populations.

Our study presents a cost-effectiveness analysis of extending the Healthy Incentives Pilot (HIP) to all participants in the Supplemental Nutrition Assistance Program (SNAP) across the U.S. Originally a trial offering a 30% rebate on fruits and vegetables to 7,500 SNAP households, our decision model evaluates the potential impact of this program’s nationwide expansion. The findings estimate a life-time per capita cost of $1,323 to the Federal government and an average gain of 0.082 quality-adjusted life years (QALYs) per SNAP participant. This results in an incremental cost-effectiveness ratio (ICER) of $16,172 per QALY gained. Sensitivity analysis indicates a high likelihood that the ICER would be below standard cost-effectiveness thresholds. Although the expansion of HIP could promote healthier eating habits among SNAP households, its impact on significant dietary behavior change, weight management, and overall health outcomes is limited, suggesting that more comprehensive strategies are necessary for profound public health improvements.

This study explores the potential impact of nationwide sugar-sweetened beverage (SSB) warning labels and restaurant menu labeling regulations on dietary habits, weight status, and health care expenditures in U.S. adults. Using a stochastic microsimulation model based on data from the National Health and Nutrition Examination Survey and the Medical Expenditure Panel Survey, the study projects significant reductions in daily energy intake, body weight, BMI, and health care costs from these policy interventions. The model estimates that SSB warning labels could reduce daily energy intake by 19.13 kcal, body weight by 0.92 kg, BMI by 0.32, and health care expenses by $26.97 per capita over 10 years. Similarly, menu labeling regulations are projected to decrease daily energy intake by 33.09 kcal, body weight by 1.57 kg, BMI by 0.55, and health care expenditures by $45.47 per capita. These reductions correspond to annual medical cost savings of $0.69 billion for SSB warning labels and $1.16 billion for menu labeling. The findings suggest that implementing SSB warning labels and menu labeling regulations could be effective strategies for controlling obesity and related health care costs.

In our Missouri project on school diabetes care, we conducted a detailed survey and interviews with school nurses, uncovering significant barriers impacting diabetes care quality. Key challenges included nurses' workloads, inadequate training for teachers and parents, and funding shortages, especially in disadvantaged areas. These barriers directly affected care aspects like training, policy adherence, resource availability, and physical activity for diabetic students. Our analysis revealed a direct inverse correlation between these barriers and care quality. Interviews with nurses highlighted complexities in their roles and responsibilities, exacerbated during the COVID-19 pandemic. The findings emphasize the need for multi-level policy interventions to improve diabetes care in schools, ensuring better health outcomes for students.

We conducted a cost-benefit analysis of a nationwide expansion of a school-based water access intervention, initially trialed in New York, to promote plain water consumption during lunchtime. The decision model compared the implementation of water dispensers in school cafeterias across the country with a no-action scenario. The analysis showed an estimated incremental cost of $18 per student for the intervention, against a benefit of $192 per student, resulting in a net benefit of $174 per student. Subgroup analysis indicated higher benefits for boys. The national rollout of this intervention could prevent approximately 0.57 million cases of childhood overweight, leading to lifetime cost savings of $13.1 billion. The model projected that for every dollar spent on the intervention, there would be a $14.5 return in savings, highlighting a favorable benefit-cost portfolio. This research underscores the potential economic and health benefits of extending the water access intervention to schools nationwide.

Our study investigated the effect of state laws governing physical education (PE) on PE class attendance among U.S. high school students from 2003 to 2017. Analyzing data from over half a million students in the national Youth Risk Behavior Survey, merged with the National Cancer Institute's policy data, we found that stronger state PE laws varied in their impact on weekly PE attendance. Increases in scores for laws on class time, staffing, joint use agreements, fitness assessment, and curriculum positively affected attendance, with increases ranging from 0.13 to 0.30 days per week. Conversely, laws focusing on physical activity intensity, PE proficiency, and recess time were associated with decreases in attendance, ranging from 0.09 to 0.25 days. Most notably, the influence of these policies was more significant among female students. This research underscores the nuanced role of state legislation in shaping PE class attendance and highlights the need for thoughtful policy design to effectively promote physical education in schools.

Our study evaluated the impact of state laws governing competitive foods and beverages sold in schools on childhood overweight and obesity among children with and without special healthcare needs (SHCN), aged 10-17 years. We merged individual-level data from the 2007-2008, 2011-2012, and 2016 National Survey of Children's Health (totaling 108,009 participants) with state laws regulating competitive foods/beverages in schools. The analysis, using state random-effect logistic regressions and adjusting for child/family characteristics and state soda sales tax, found no significant association between these state laws and childhood overweight/obesity among children, regardless of their SHCN status. Additionally, state soda sales taxes did not impact childhood overweight/obesity among children with SHCN. This study highlights the absence of a protective effect from state laws on competitive school foods/beverages against obesity risk among children, particularly those with SHCN, and suggests the need for further research to replicate these findings and assess the role of school district policies on adiposity in this vulnerable child population.

Our study conducted a nuanced risk-benefit analysis to establish the optimal duration for moderate-intensity outdoor physical activity (MPA) in the context of varying levels of fine particulate matter (PM2.5) air pollution. The study discovered an inverse nonlinear relationship, demonstrating that as background PM2.5 concentration increases, the advisable duration for outdoor MPA decreases. For instance, when PM2.5 levels reached 186 µg/m3, the optimal MPA duration fell to 2.5 hours per week, aligning with the minimum level recommended in current physical activity guidelines. Further elevation of PM2.5 levels to 235 µg/m3 reduced the optimal MPA time to just 1 hour per week. This relationship was more pronounced in areas closer to pollution sources and among adults aged 60 years and older, with the optimal MPA duration dropping to 2.5 hours per week at a lower PM2.5 concentration of 45 µg/m3. The study concluded that for the average global urban background PM2.5 concentration of 22 µg/m3, the health benefits of outdoor MPA significantly outweigh the risks associated with air pollution.

This longitudinal study investigates the impact of participation in food assistance programs like the Supplemental Nutrition Assistance Program (SNAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) on food security (FI) patterns among women in Illinois. Engaging a cohort of 297 women, the study uniquely tracks FI status over a 15-month period through repeated phone surveys using the USDA Food Security Survey Module. The FI patterns observed were categorized as stable, transitioning, or unstable, based on changes over time. Among the participants, a significant majority were in households with children, and a substantial proportion utilized SNAP or WIC. Intriguingly, while SNAP usage did not show a statistically significant relationship with FI patterns, WIC users notably transitioned from being food insecure to secure. The study, though challenged by participant attrition, offers valuable insights into the dynamics of FI and the potential effectiveness of food assistance programs in promoting food security over time.

This study, conducted in 2016, assesses the network dynamics among 89 agencies in four rural counties of Michigan, all working towards promoting healthy eating and lifestyles among populations with limited resources. The agencies, diverse in their focus and target audiences, include K-12 schools, early childhood centers, emergency food providers, health-related agencies, social resource centers, low-income/subsidized housing complexes, continuing education organizations, and others. The network analysis focused on four key structures: communication, funding, cooperation, and collaboration. Results indicate that while there was moderate cooperation, the agencies were generally loosely connected in terms of communication, funding, and collaboration, as shown by low network density. The study also found that these networks were decentralized, with no single or few agencies dominating, as evidenced by low centralization scores. Additionally, the study revealed homophily within the networks, where agencies of similar types tended to correlate significantly. Furthermore, connections in one network often implied connections in the others, suggesting inter-network linkage. The findings highlight the importance of robust partnerships in community health initiatives, especially for populations with limited resources.

This study assessed the interagency networks in Illinois communities that advocate for healthy eating and lifestyle habits among populations with limited resources. Surveys were conducted with 159 agencies, categorized into eight types including K-12 schools, early childhood centers, and health-related agencies. Four key network structures – communication, funding, cooperation, and collaboration – were analyzed to understand how these agencies interact within their communities. Results indicated that these networks are loosely connected, evidenced by low network density and reciprocity, and lack a centralized structure around influential agencies, as shown by low betweenness centrality. The study also found signs of homophily, where agencies of similar types are more likely to interact. Crucially, agencies involved in one network were often connected in other networks as well. These findings highlight the need for stronger partnerships and collaboration among community agencies to promote healthier lifestyles effectively, especially in resource-limited areas. Network analysis emerged as a vital tool for evaluating and enhancing community partnerships, essential for public health initiatives.

This study assesses the efficacy and cost-effectiveness of a novel approach in nursing homes: training nonnursing staff to assist in nutritional care delivery. Conducted across five community nursing homes and involving 122 long-stay residents requiring caloric supplementation, the study utilized an 8-hour training curriculum for nonnursing personnel. These trained staff members were then tasked with delivering between-meal supplements or snacks to the intervention group, in contrast to the control group which received usual care. The results were significant; the intervention group exhibited a substantial increase in between-meal caloric intake, consuming an average of 163.33 additional calories per day compared to the control group. Notably, the intervention added only $1.27 per person per day to the usual care costs, translating to a cost-effectiveness ratio of 134 additional calories per dollar spent. The increased costs were primarily attributed to the augmented frequency and quantity of snack items provided, along with the staff time allocated for assistance. This study demonstrates the cost-effective potential of leveraging nonnursing staff for nutritional care, significantly boosting residents' caloric intake between meals and enhancing overall care quality in nursing homes.

In our comprehensive scoping review, we delve into the application of artificial intelligence (AI) in enhancing physical activity interventions. The review examines a range of AI methodologies - from machine learning to deep learning and reinforcement learning - in their role of advancing physical activity outcomes. The study analyzes 24 peer-reviewed articles, highlighting the effectiveness of AI in areas like pattern recognition, outcome prediction, and intervention improvement. It also identifies key future directions for AI in this field, such as personalized interventions and multimodal data integration. This pivotal work underscores the growing importance of AI in shaping effective physical activity strategies and promoting public health advancements.

This scoping review provides a comprehensive overview of the applications of Artificial Intelligence (AI), specifically machine learning (ML) and deep learning (DL), in obesity research. The review, which analyzed 46 studies from PubMed and Web of Science, showcases how AI models have been effectively used to measure, predict, and treat obesity. It was found that AI models, in most cases, outperformed traditional statistical approaches in predicting obesity-related outcomes, indicating their potential in revealing complex patterns and relationships in obesity data. The review also observes a growing trend towards the use of advanced DL models for sophisticated tasks like computer vision and natural language processing. Furthermore, the review serves as an introductory guide to popular ML and DL models, highlighting their specific applications in the included studies. The study concludes by discussing future trends in AI applications in obesity research, such as multimodal models and synthetic data generation, underscoring AI's evolving role in advancing obesity research methodologies.

In this research, we developed machine learning (ML) models to enhance the accuracy of self-reported anthropometric data, a critical issue in monitoring population obesity risk. Utilizing data from 50,274 adults in the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2020, the study addressed the substantial discrepancies often found between self-reported and objectively measured height, weight, and body mass index (BMI). The team applied nine ML models to predict these measures accurately. The most effective models significantly reduced the differences, decreasing the discrepancy in average height by 22.08%, weight by 2.02%, BMI by 11.14%, and obesity prevalence by 99.52%. Remarkably, the predicted obesity prevalence (36.05%) closely matched the objectively measured figure (36.03%), underscoring the potential of these models to provide reliable obesity prevalence estimates from survey data. This study demonstrates the promising application of ML in improving data accuracy for public health surveillance.

In our study, we utilized deep neural network models to detect and quantify the nutritional content of common edible nuts from photographs. Our dataset comprised 1,380 images, each showcasing a variety of 11 popular nut types. Using transfer learning, the models were trained for precise multi-label classification and object detection. The model demonstrated a mean average precision of 0.7596 in nut localization and achieved a high accuracy of 97.9% in identifying nut types and quantities. Significantly, it accurately estimated the aggregate nutrient profiles, including total energy, protein, carbohydrates, fats, and essential vitamins and minerals, with an error margin between 0.8-2.6%. This advancement marks a significant step in incorporating AI into diet-tracking apps, offering potential benefits for nutritional tracking and promoting healthier dietary choices. (Article under peer review)

In two related studies, deep neural network models were employed to analyze public sentiment on Twitter regarding soda taxes and menu labeling regulations, key health policies aimed at combating obesity. The soda tax study, analyzing around 370,000 tweets from 2015 to 2022, observed a peak in public attention in 2016 with a subsequent shift towards neutral sentiment, while negative sentiments steadily increased over time. The menu labeling study, covering 2008 to 2022, also found that public discourse peaked around major policy announcements, with a trend towards neutral and news-related tweets, and a recent rise in negative sentiment. Both studies achieved high accuracy in sentiment classification and identified predictors like the author's followers and tweeting frequency. These findings highlight the dynamic public attitudes toward health policies and underscore the potential of social media analysis in informing policy design and refinement.

We evaluate the medical diagnostic capabilities of ChatGPT using simulated patients. The study, conducted with standardized scripts covering nine common diseases, assessed ChatGPT 3.5's ability to diagnose and recommend treatments based on these scenarios. The findings revealed that in its first suggestions, ChatGPT accurately diagnosed 66.7% of cases and recommended appropriate medication in 70.4% of cases. The accuracy increased when considering any of ChatGPT's three recommendations per case. However, the study also noted ChatGPT's recommendation of potentially harmful treatments in 33.3% of trials. Interestingly, ChatGPT was more accurate with non-communicable diseases than with infectious diseases. The study underscores ChatGPT's potential as a supplementary tool in healthcare, particularly in resource-limited settings, but also highlights the need for caution due to the risks of inaccuracies. This research opens new avenues for utilizing AI in medical diagnostics, emphasizing the balance between accessibility and safety.

(Article under peer-review)

In this innovative study, deep neural networks were applied to analyze short audio recordings for predicting obesity status, exploring the potential of voice characteristics as a non-invasive biomarker for obesity detection. Conducted with 696 participants, the study classified individuals into obesity and non-obesity groups based on self-reported body mass index (BMI). Participants' voice recordings were converted into spectrograms and analyzed using an adapted YOLOv8 model. The model achieved a global accuracy of 0.70 and a macro-F1 score of 0.65, showing greater efficacy in identifying non-obesity cases. Despite these promising results, the study acknowledges its limitations, including reliance on self-reported BMI and a small, homogenous sample size, which may affect the generalizability of the findings. The variability in recording quality also suggests the need for further research with improved methodologies. This study represents a significant step towards utilizing vocal characteristics in medical diagnostics for obesity, highlighting the potential of voice-based biomarkers while recognizing the challenges and scope for future advancements in the field.

(Article under peer-review)

We developed artificial intelligence (AI) models to identify and correct news headlines that exaggerate obesity-related research findings, a critical issue that can mislead public perception and erode trust in scientific communication. The research involved collecting 523 exaggerated headlines from digital media, identifying common exaggeration patterns such as inferring causality from observational studies or generalizing findings inaccurately. The team created a BERT model fine-tuned to distinguish between exaggerated and non-exaggerated headlines, and developed generative language models (BART, PEGASUS, and T5) to automatically generate accurate headlines based on scientific abstracts. The performance of these models was high, with the BERT model achieving 92.5% accuracy and the generative models surpassing baseline ROUGE scores, demonstrating their potential to enhance the accuracy and integrity of media reporting on scientific research.

Funded by OpenAI, our team is pioneering the use of advanced AI models to revolutionize the analysis of qualitative data, specifically over 500 transcripts from interviews and focus groups. This approach, leveraging sophisticated prompt engineering and LangChain techniques, is designed to enhance the summarization and theme identification processes, offering a more efficient alternative to traditional, labor-intensive qualitative analysis methods. The AI models are tasked with dissecting and interpreting the extensive data, focusing on extracting nuanced insights and themes quickly and comprehensively. We are validating the effectiveness of this AI-driven analysis by comparing its outputs with established findings in published literature. This comparison aims to assess the accuracy and depth of AI-generated insights, potentially uncovering novel perspectives or reinforcing existing understanding. Our study aims to demonstrate the potential of AI in qualitative research, significantly reducing the time and resources needed for data analysis and opening up new possibilities for extensive use of qualitative data in various research fields.

(Article under development)

In this study, we evaluated the responses of 14 popular chatbots to questions about body image, a crucial mental health concern, particularly among adolescents and young adults. The study focused on both companion and therapeutic chatbots, assessing their reactions to ten body image-related questions based on validated instruments. The findings revealed a modest overall quality in the chatbots’ responses, with scores averaging five out of nine and a wide range of individual scores from one to eight. Notably, companion chatbots tended to focus on comforting users, while therapeutic ones aimed more at identifying causes and suggesting remedies. Some therapeutic chatbots could recognize potential mental health crises. However, the study highlights substantial variability in the content and quality of responses among chatbots, raising concerns about the potential for misleading or biased advice. This underscores the need for continued technical and supervisory enhancements to ensure the safe and effective use of chatbots in sensitive areas like body image counseling.

In this study, a novel method is introduced to differentiate between machine-generated and human-written text, addressing the growing concerns about academic integrity and plagiarism in the context of advanced language models like ChatGPT. The study employed a dataset of student-written essays and comparable essays generated by ChatGPT, focusing on similarity scores within and between these sets. The methodology, which combines analysis of prompts and essays, demonstrated high accuracy in identifying machine-generated text, with impressive AUC, false positive, and false negative rates. This approach significantly outperforms traditional plagiarism detection tools that rely solely on text analysis. The research highlights the potential of this method in maintaining academic integrity but also acknowledges the need for further studies to assess its applicability in varied educational settings and with different model parameters.