95038 research outputs found
Sort by
Connectome-wide Brain Signature During Fast Food Advertisement Exposure Predicts BMI At 2 Years
Full text linkFood advertisements target adolescents contributing to weight gain and obesity. However, whether brain connectivity during those food advertisements can predict weight gain is unknown. Here, 121 adolescents (14.1±1.0y; 50.4% female; BMI: 23.4±4.8; 71.9% White) completed both a baseline fMRI paradigm viewing advertisements (unhealthy fast-food, healthier fast-food, and non-food) and an anthropometric assessment two years later. We used connectome-based predictive modeling (CPM) to derive brain networks that were associated with BMI both at baseline and the 2-year follow-up. During exposure to unhealthy fast-food commercials, we identified a brain network comprising high-degree nodes in the hippocampus, parahippocampal, and fusiform gyrus rich with connections to prefrontal and occipital nodes that predicted lower BMI at 2-year follow-up (r =0.17; p=0.031). A similar network was derived from baseline BMI (n=168; r =0.34; p<0.001). Functional connectivity networks during exposure to the healthier fast-food (p=0.152) and non-food commercials (p=0.117) were not significant predictors of 2-year BMI. Key brain regions in our derived networks have been previously shown to encode aspects of memory formation, visual processing, and self-control. As such, the integration of these regions may reflect a mechanism of adolescents’ ability to exert self-control towards obesogenic food stimuli
Acceptability and feasibility of acceptance and commitment therapy for improving outcomes in hematopoietic stem cell transplant.
Full text linkINTRODUCTION: Allogeneic hematopoietic stem cell transplant (HCT) has the potential to cure patients with hematologic malignancies, but treatment-related morbidity and mortality is high. Transplant outcomes are optimized by patients maintaining physical activity. The aim of the current study was to examine whether a brief Acceptance and Commitment Therapy (ACT) intervention is acceptable to HCT patients and caregivers and helps patients engage in healthy behavior despite physical and emotional discomfort. METHODS: Patients ≥ 18 years of age who were undergoing allogenic HCT for any cancer or non-cancer illness and their caregivers were invited to complete six ACT sessions between transplant day - 30 and day + 90. Multiple small cohorts of n = 3 dyads were enrolled, and the protocol content was iterated after each cohort to reflect the experiences and breadth of concerns of individuals undergoing HCT. Acceptability was indexed by session completion rates and acceptability surveys. Pre-post 6-minute walk distance was collected as an index of physical function as part of standard care. RESULTS: Sixteen HCT dyads enrolled in the study; 12 continued to treatment. Most participants completed all assigned sessions. Participants perceived ACT to be helpful and 70% (5 of 7) of the patients with pre-post 6-minute walk test data showed improvement. CONCLUSION: ACT is an acceptable and potentially useful intervention for individuals undergoing HCT. Additional controlled studies are warranted
An examination of multiple illness and social related stressors during the COVID-19 pandemic on maternal and child mental health; findings from a birth cohort in rural Pakistan
Full text linkPurpose In an effort to improve preparation for future pandemics, researchers continue to examine the myriad psychosocial pathways through which the COVID-19 pandemic impacted mental health. Moving beyond extant research on factors such as financial difficulties or social isolation, we present findings on two less documented pathways: (a) COVID illness and death within one’s social network and (b) experiencing pandemic-related basic needs stressors, beyond financial difficulties, on the mental health of mothers and children in South Asia. Methods Data come from the 2021–2022 wave of the Bachpan birth cohort (6-year-old children, n = 814 mother-child dyads) in rural Pakistan. Multivariable regressions were used to estimate the association between COVID illness/death among family/friend/community members and basic needs stressors (e.g. difficulties with food, housing, medical care) with 4 outcomes: maternal anxiety (GAD-7) and depression symptoms (PHQ-9), and child emotional/behavioral problems (SDQ-TD) and anxiety (SCAS). Results Maternal mental health was independently predicted by illness within the mother’s network and basic needs stressors, even after accounting for pre-pandemic socioeconomic status, depression, and pandemic-related financial stressors. Specifically, having a family member/friend who became ill was associated with 1.29 higher PHQ-9 scores (95% CI: 0.34, 2.24), compared to those who did not know anyone affected. Increased basic needs stressors were also associated with higher PHQ-9 and GAD-7 levels and children’s behavioral/emotional problems and anxiety. Social network proximity to COVID illness/death was not associated with child mental health outcomes. Conclusion These findings provide evidence of additional pathways through which prolonged global events, like the pandemic, can have multigenerational mental health impacts
Low-Carbon City Policies and Employment in China: Impact Effects and Spatial Spillovers
Full text linkThis study examines the impact of low-carbon city policies on urban employment using panel data from 2006 to 2021. The findings reveal that these policies significantly enhance urban employment by promoting green technological innovation, which drives economic growth and creates new job opportunities. Low-carbon policies also exhibit spatial spillover effects, benefiting neighboring cities within a 200 km radius. However, these effects vary non-linearly with distance. The key mechanisms include green technology adoption, industrial structure optimization, and the promotion of green consumption habits. These mechanisms accelerate industrial upgrading, foster the growth of tertiary and green industries, and expand job opportunities in emerging markets. Heterogeneity analysis shows more substantial employment effects in non-resource-based cities, provincial capitals, cities with high government innovation preferences, tertiary sector dominance, and higher urbanization rates. This highlights the need for policies tailored to specific urban characteristics. In conclusion, low-carbon policies integrate climate action with employment growth. Policymakers should invest in green technologies, support industrial transformation, and enhance public environmental awareness to maximize employment benefits, fostering sustainable urban development
Testing factors that enhance private participation in payments for ecosystem service programs targeting flood mitigation
Full text linkTo address the severe consequences of low participation, more studies are needed that empirically evaluate how different factors affect enrollment in payments for ecosystem services (PES) programs. In this paper, we provide empirical insight into how different land-leasing, purchase, and management arrangements might affect potential participants’ willingness to enroll in these programs. We administer a choice experiment in the coastal plain of the State of North Carolina (USA), to explore how a hypothetical, flood mitigation-focused PES program could optimize participation across a variety of natural infrastructure practices and across multiple aspects of program design. We find evidence that higher proportions of total household income from farming have a significant relationship with the likelihood of participation in PES programs that seek to mitigate flooding. Other factors that we examined – including income loss from past flood events, respondents’ previous participation in PES programs, and different combinatorial arrangements of payment structuring and timing – had no discernable relationship to initial decisions about program participation. Furthermore, we find evidence for the importance of previous participation in PES programs as a key factor in decisions regarding the extent or level of program participation. In light of this, we propose a risk-based reframing of PES program participation decisions. From this risk avoidance perspective, we theorize that potential participants who are more dependent on farming revenue for their income may be proportionally more risk averse to any actions that could complicate farming or endanger future farming revenue
An In-Depth Analysis of Youth Sport Coaches’ Perceptions of Sport Culture, Safety, and Injury Prevention
Full text linkCoaches are key stakeholders in the world of youth-organized sports. As key stakeholders, they are influential to the “culture of sport” and how it pertains to safety and injury in sport. However, little data exists to inform injury prevention intervention in youth sport. There is a critical need to understand the perceptions and norms that contribute to youth sport culture and coaches’ perceptions of this culture. The purpose of this study was to develop an understanding of youth sport coaches’ attitudes, perceptions, and norms regarding sport culture and injury. Thirty coaches were divided into five recorded focus groups. Following transcription, researchers utilized a consensual qualitative research framework. The research team individually coded data into themes and categories, meeting to attain a consensus on findings. An auditor controlled for bias. Participants highlighted three key areas related to the purpose of the study: (1) influences and expectations, (2) sport safety, and (3) education and development. Findings suggest that youth sport coaches view themselves as an integral component of sports culture, safety, and injury prevention. As such, coaches must build relationships with the community to provide a foundation for optimal sport and health outcomes
Preparing for Inclusion of Agricultural Workers in Climate-Related and Other Public Health Emergencies: A Quantitative Content Analysis of County Emergency Preparedness Plans, Eastern North Carolina, 2022-2023
Full text linkOBJECTIVE: Agricultural workers such as migrant and seasonal farmworkers are a population uniquely vulnerable to the negative effects of climate-related disasters. The primary aim of this study was to assess how counties in eastern North Carolina (NC) incorporated best practices related to agricultural workers in emergency preparedness planning. METHODS: The authors conducted a quantitative content analysis of 47 emergency preparedness plans from 41 eastern NC counties between November 2022 and March 2023. The research team developed a codebook based on recommended best practices from the Centers for Disease Control and Prevention, Farmworker Advocacy Network, Federal Emergency Management Agency, and National Center for Farmworker Health. Best practices included having materials in Spanish language, mapping vulnerable populations, and garnering input from agricultural workers. RESULTS: Of the 47 plans evaluated, most lacked inclusion of agricultural workers in emergency preparedness planning. Furthermore, plans demonstrated few relevant best practices that address agricultural workers' specific challenges in preparation for and recovery from climate-related disasters. CONCLUSIONS: Public emergency preparedness plans in eastern NC rarely include mention of recommended best practices related to agricultural workers. Local emergency preparedness officials should consider collaboration with advocacy groups, community health workers, and federal emergency management agencies to build disaster resilience
Rethinking machine unlearning for large language models
Full text linkWe explore machine unlearning in the domain of large language models (LLMs), referred to as LLM unlearning. This initiative aims to eliminate undesirable data influence (for example, sensitive or illegal information) and the associated model capabilities, while maintaining the integrity of essential knowledge generation and not affecting causally unrelated information. We envision LLM unlearning becoming a pivotal element in the life-cycle management of LLMs, potentially standing as an essential foundation for developing generative artificial intelligence that is not only safe, secure and trustworthy but also resource-efficient without the need for full retraining. We navigate the unlearning landscape in LLMs from conceptual formulation, methodologies, metrics and applications. In particular, we highlight the often-overlooked aspects of existing LLM unlearning research, for example, unlearning scope, data–model interaction and multifaceted efficacy assessment. We also draw connections between LLM unlearning and related areas such as model editing, influence functions, model explanation, adversarial training and reinforcement learning. Furthermore, we outline an effective assessment framework for LLM unlearning and explore its applications in copyright and privacy safeguards and sociotechnical harm reduction
2024 Robinson Honors Fellowship
No full text2024 Robinson Fellowship on Rome's transition from a republic to an empire in Pompeii, Herculaneum, Capri, Rome, Florence, and Milan
Gains vs losses in pay-for-performance: Stated preference evidence from a U.S. survey
Full text linkBACKGROUND: Pay-for-performance (P4P) incentives can be paid as a bonus (gain) or a penalty (loss). Diminishing marginal utility of wealth suggests that, starting from the same initial wealth, individuals dislike losses more than they like equivalent gains. OBJECTIVE: This study reports the minimum financial gain or loss required to motivate primary care providers and clinical staff to try to increase their human papillomavirus (HPV) vaccination rates. DATA: In 2022, we conducted a national U.S. survey through WebMD's Medscape Network of clinical staff working in primary care clinics that provided HPV vaccination to children ages 9 through 12 years (N = 2,527; response rate = 57%). METHODS: We randomized respondents to one of two hypothetical HPV vaccine incentive designs: a bonus for reaching an unspecified target HPV vaccination rate and a penalty for failing to reach the unspecified target. The primary outcome is the self-reported smallest incentive amount (U.S. dollars) that would motivate participants to try and increase their HPV vaccination rates. We tested for differences across P4P designs using unadjusted responses and linear regressions adjusting for clinic and respondent characteristics. We also tested for heterogeneous responses by experience with incentizves, training, and rurality. RESULTS: The mean amount required to motivate effort was 1,185 in the loss P4P design (unadjusted difference = 970 [p < 0.001], adjusted difference = 967 [p < 0.001]). There were no heterogeneous effects by rurality or experience with incentives. Physicians reported the highest differences (in dollars) between gain and loss P4P designs. CONCLUSIONS: Stated preference data from primary care clinical staff suggests that effective P4P incentives could be half as large if designed as losses rather than gains