Climate change and poverty are intertwined global challenges that disproportionately impact Least Developed Countries (LDCs). However, how global institutions discursively construct the climate-poverty nexus to legitimize their policy recommendations remains underexplored. Drawing on Critical Policy Discourse Analysis (CPDA), this study investigates how the World Bank Group frames the relationship between climate change and poverty in its Country Climate and Development Reports (CCDRs) for LDCs, as well as the discursive legitimation strategies embedded in these constructions. Findings identify two dominant, complementary discursive frames: the vulnerability frame and the causality frame. The vulnerability frame constructs poor and marginalized groups as passive victims of climate impacts, leveraging on attributive relational and passive material processes, and deploys moral evaluation as a legitimization strategy to position adaptation policies as a non-negotiable moral imperative. In contrast, the causality frame positions climate change as an active, causal agent driving poverty dynamics, utilizing active material processes and extended causal chains, and employs scientific rationalization to legitimize mitigation policies as rational, long-term investments aligned with LDCs’ development priorities. These two frames collectively shape a hybrid policy agenda that integrates ethical imperatives with technocratic efficiency, reflecting the World Bank’s attempt to legitimize its institutional influence on LDC climate-development trajectories. This research contributes to the scholarship on discourse in global climate governance by equipping stakeholders to engage with international policy advice critically and fostering more context-sensitive strategies for LDCs.
Although large language models (LLMs) have undergone substantial development, their applicability to epidemiological research has not been sufficiently examined. This study aims to develop and evaluate an LLM-based framework for hypothesis generation and testing, demonstrating its application in childhood asthma in the National Health and Nutrition Examination Survey (NHANES). Pilot study was conducted to explore factors associated with childhood asthma in the 2001–2020 NHANES cycles. A modular agent system was developed, including Database Query, Statistic, Paper Search, and Paper Download tools, along with two LLM models (Key Generator and Hypothesis Tester). Multivariable logistic regression was used to test for the association between each variable and current asthma, generating a tentative affirmative claim. The Key Generator module produced keywords for literature search, the Paper Search and Paper Download tools queried PubMed and retrieved relevant studies, and the Hypothesis Tester module synthesized evidence and determined the support for claims for each variable. Keywords and conclusions were reviewed by researchers and validated using multiple LLMs (ChatGPT, DeepSeek, and Gemini) to ensure consistency and robustness. 25,839 children with (n = 2928) and without (n = 22,911) current asthma, and 10,359 variables were included in the multivariable analysis, which yielded 100 variables associated with asthma. Of these, 21 were directly related to asthma (supporting published studies), 43 were indirectly related to asthma (based on background knowledge, though not explicitly discussed in the available publications), and 34 were unrelated to asthma. Two variables were excluded due to a lack of discriminative keywords. This study demonstrates the effectiveness of LLM-based models for generating and testing hypotheses about childhood asthma.
Depressive symptoms are prevalent and demonstrate distinct developmental trajectories throughout adolescence. Although previous research has suggested central symptoms as possible intervention targets, few studies have explored the effects of targeting these symptoms on global network states. Utilizing the Ising model and the NodeIdentifyR algorithm, this study aimed to identify effective intervention targets and examine their associations with central symptoms across different stages of adolescence. A total of 46,842 participants completed the Center for Epidemiologic Studies Depression Scale and provided demographic information. Participants were categorized into early (n = 15,299), middle (n = 15,596), and late (n = 15,547) adolescence. The Ising model identified “feeling sad” and “feeling depressed” as the symptoms with the highest expected influence in early and middle-to-late adolescence, respectively. The expected influence value of “feeling depressed” increased from early to late adolescence. Simulated interventions projected that decreasing the thresholds of “feeling bad” (early adolescence) and “feeling depressed” (middle and late adolescence) would yield the greatest reduction in network activation, identifying them as effective treatment targets. Worsening “feeling sad” and “feeling depressed” (early adolescence) and “feeling blue” (middle and late adolescence) was projected to result in the greatest increase in network activation, making them the effective prevention targets. The most central symptoms were not necessarily congruent with the effective intervention targets identified by simulations. These findings may help practitioners optimize treatment and prevention efforts for adolescents with depressive symptoms across distinct developmental stages.
This study provides a physicochemical characterisation of commercial Brazilian sparkling wines, aiming to describe the typicity of products obtained using the Charmat and Traditional methods. A total of 261 wines were analysed, including 119 produced by the Charmat method and 142 by the Traditional method. The results show distinct compositional patterns across the analysed samples. Wines produced by the Traditional method, predominantly based on blends of Chardonnay and Pinot Noir, showed higher levels of lactic acid, volatile acidity, alcohol, and pressure, together with lower residual sugar contents. In contrast, Charmat sparkling wines displayed greater varietal diversity, including the widespread use of Glera, and higher levels of residual sugar, malic, and citric acids. A relatively high proportion of sparkling wines were identified as “Long Charmat”, with maturation periods of six months or more on lees in tanks, while a subset of Traditional method wines showed ageing times shorter than 12 months. In both production methods, Riesling Italico (Welschriesling) ranked among the four most frequently used grape varieties. Overall, the results highlight consistent compositional tendencies within a broad set of commercial wines. This study establishes a reference compositional dataset for Brazilian sparkling wines, contributing to the understanding of this expanding wine category by characterizing production practices and grape variety usage and identifying “Long Charmat” as a distinctive feature in the Brazilian context.
The development of high-efficiency copper indium gallium diselenide (CIGS) solar cells is currently driven by a dual strategy of internal structural refinement and integration into multi-junction tandem architectures. This study aims to systematically analyze the key design and optimization strategies required to overcome the 33.7% Shockley–Queisser limit of single-junction devices. The results demonstrate that bandgap engineering, particularly through double-graded “notch” profiles, significantly enhances charge carrier collection and improves overall device performance, while alkali metal post-deposition treatments effectively reduce interface recombination losses. Furthermore, integrating CIGS with perovskite top cells in two-terminal (2T) and four-terminal (4T) configurations is a promising pathway to achieving efficiencies exceeding 30%. By combining advanced vacuum-based fabrication techniques, such as the three-stage co-evaporation process, with precise optical management, CIGS technology is positioned as a versatile candidate for both high-performance terrestrial and radiation-tolerant space applications.
