Prashant Garg
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Prashant Garg has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
Thiemo Fetzer
University of Warwick, Coventry, UK
6 shared publications- Thiemo Fetzer
Department of Economics, University of Warwick, Coventry, UK
4 shared publications - Hongyu Zhou
Centre for R&D Monitoring (ECOOM), Faculty of Social Sciences University of Antwerp Antwerp Belgium
2 shared publications Eleonora Alabrese
University of Bath
1 shared publicationFrancesco Capozza
University of Barcelona
1 shared publicationHongyu Zhou
Renmin University of China
1 shared publication
Latest Publications
- Open Access
- 13 October 2026
- Published by BMJ in BMJ Health & Care Informatics
- Vol. 33 (1) , e102056
How much do language and context affect the accuracy of AI health advice?
Objectives Large language models (LLMs) are increasingly used for health advice, but evidence on how performance varies across languages, topics and sources remains limited. We assess linguistic and contextual variation in Artificial Intelligence (AI) health-claim verification.Methods We evaluated seven widely accessible LLMs using UK and European Union (EU) health-claim registers and 9088 journalist-vetted PUBHEALTH claims. The register-language analysis used authorised entries translated into 21 languages; status-labelled UK/EU extensions added non-authorised entries for diagnostic checks. PUBHEALTH claims spanned COVID-19, abortion, politics and general health across government, media and other source domains, including nature.com. Models classified claims as supported or unsupported using modal responses across repeated runs.Results Authorised-claim sensitivity was highest in English and closely related European languages and declined with syntactic distance from English. In status-labelled UK and EU register extensions, models recognised authorised claims but often overclassified non-authorised claims as supported. PUBHEALTH accuracy was lower and varied by topic and source; inaccuracy was highest for general-health claims and claims linked to nature.com.Discussion High authorised-claim sensitivity for English register claims can hide lower sensitivity in other languages and weaker performance in other public-health contexts. Training-data exposure, editorial framing and topic-specific tuning are plausible contributors to these gaps.Conclusions LLM performance in health-claim verification varies substantially by language, topic and source. English-language performance does not reliably generalise across contexts, underscoring the need for multilingual, domain-specific evaluation before public-health deployment.
- 13 October 2026
- Published by BMJ in BMJ Health & Care Informatics
- Vol. 33 (1) , e102056
Open AccessHow much do language and context affect the accuracy of AI health advice?
Objectives Large language models (LLMs) are increasingly used for health advice, but evidence on how performance varies across languages, topics and sources remains limited. We assess linguistic and contextual variation in Artificial Intelligence (AI) health-claim verification.Methods We evaluated seven widely accessible LLMs using UK and European Union (EU) health-claim registers and 9088 journalist-vetted PUBHEALTH claims. The register-language analysis used authorised entries translated into 21 languages; status-labelled UK/EU extensions added non-authorised entries for diagnostic checks. PUBHEALTH claims spanned COVID-19, abortion, politics and general health across government, media and other source domains, including nature.com. Models classified claims as supported or unsupported using modal responses across repeated runs.Results Authorised-claim sensitivity was highest in English and closely related European languages and declined with syntactic distance from English. In status-labelled UK and EU register extensions, models recognised authorised claims but often overclassified non-authorised claims as supported. PUBHEALTH accuracy was lower and varied by topic and source; inaccuracy was highest for general-health claims and claims linked to nature.com.Discussion High authorised-claim sensitivity for English register claims can hide lower sensitivity in other languages and weaker performance in other public-health contexts. Training-data exposure, editorial framing and topic-specific tuning are plausible contributors to these gaps.Conclusions LLM performance in health-claim verification varies substantially by language, topic and source. English-language performance does not reliably generalise across contexts, underscoring the need for multilingual, domain-specific evaluation before public-health deployment.
- Open Access
- 1 September 2026
- Published by Springer Nature in Research Square
Health shocks reallocate the direction of medical research
We examine whether research systems reallocate scientific effort as health needs change. We assemble a global disease–location panel for 204 countries and territories (1990–2021) by linking disease-specific publication output to disease burden in the same place and year. Using large language models, we extract diseases from article text, map them into a disease classification, and classify research funders by type. We estimate how publication output co-moves with disease burden within countries and diseases over time, and use event-study designs around the timing of outbreak alerts, including models with alerted-cell pre-outbreak trends. We find that responsiveness to endemic burden has increased over time but remains highly uneven across locations; outbreak alerts are followed by rapid, statistically significant research surges concentrated after 2000; and funder portfolios differ sharply in the diseases they support, with philanthropic funders tilted toward several diseases concentrated in lower-income settings and corporate funders tilted toward major non-communicable diseases. JEL Classification: I10, I18, O31, O32, O33, C55
- 1 September 2026
- Published by Springer Nature in Research Square
Open AccessHealth shocks reallocate the direction of medical research
We examine whether research systems reallocate scientific effort as health needs change. We assemble a global disease–location panel for 204 countries and territories (1990–2021) by linking disease-specific publication output to disease burden in the same place and year. Using large language models, we extract diseases from article text, map them into a disease classification, and classify research funders by type. We estimate how publication output co-moves with disease burden within countries and diseases over time, and use event-study designs around the timing of outbreak alerts, including models with alerted-cell pre-outbreak trends. We find that responsiveness to endemic burden has increased over time but remains highly uneven across locations; outbreak alerts are followed by rapid, statistically significant research surges concentrated after 2000; and funder portfolios differ sharply in the diseases they support, with philanthropic funders tilted toward several diseases concentrated in lower-income settings and corporate funders tilted toward major non-communicable diseases. JEL Classification: I10, I18, O31, O32, O33, C55
- Open Access
- 14 August 2026
- Published by Springer Nature in Nature Human Behaviour
- p. 1-17
Uneven patterns of cross-border media coverage following natural disasters
Many natural disasters central to climate-policy debates are hydro-meteorological hazards, yet it remains unclear how media attention is distributed across disaster types. Using 135 million news articles from 466 sources in 123 countries since 2016, we estimated cross-border reporting changes after disasters. Attention rises but is highly uneven (95% confidence intervals are shown in parentheses after the estimates): short-run increases are largest following earthquakes ( β = 0.0785 (0.0758, 0.0812)), dry-mass movements ( β = 0.0531 (0.0349, 0.0713)) and volcanic eruptions ( β = 0.0425 (0.0359, 0.0490)), and smaller for floods ( β = 0.0069 (0.0062, 0.0076)) and droughts ( β = 0.0001 (−0.0036, 0.0039)). Conditional on severity and duration, hydro-meteorological disasters receive less attention than geophysical disasters ( θ = −0.0065 (−0.0115, −0.0015)). Coverage is higher for 100+ versus 0–9 deaths ( β = 0.0360 (0.0231, 0.0490)), with stronger fatality gradients for country pairs with tighter social ties ( β = 0.1498 (0.0822, 0.2174)) and ancestral relatedness ( β = 0.2965 (0.1887, 0.4043)).
- 14 August 2026
- Published by Springer Nature in Nature Human Behaviour
- p. 1-17
Open AccessUneven patterns of cross-border media coverage following natural disasters
Many natural disasters central to climate-policy debates are hydro-meteorological hazards, yet it remains unclear how media attention is distributed across disaster types. Using 135 million news articles from 466 sources in 123 countries since 2016, we estimated cross-border reporting changes after disasters. Attention rises but is highly uneven (95% confidence intervals are shown in parentheses after the estimates): short-run increases are largest following earthquakes ( β = 0.0785 (0.0758, 0.0812)), dry-mass movements ( β = 0.0531 (0.0349, 0.0713)) and volcanic eruptions ( β = 0.0425 (0.0359, 0.0490)), and smaller for floods ( β = 0.0069 (0.0062, 0.0076)) and droughts ( β = 0.0001 (−0.0036, 0.0039)). Conditional on severity and duration, hydro-meteorological disasters receive less attention than geophysical disasters ( θ = −0.0065 (−0.0115, −0.0015)). Coverage is higher for 100+ versus 0–9 deaths ( β = 0.0360 (0.0231, 0.0490)), with stronger fatality gradients for country pairs with tighter social ties ( β = 0.1498 (0.0822, 0.2174)) and ancestral relatedness ( β = 0.2965 (0.1887, 0.4043)).
- Open Access
- 21 April 2026
- Published by Springer Nature in Research Square
Politicized Scientists: Credibility Cost of Political Expression on Twitter
As social media becomes prominent within academia, we examine its reputational costs for academics. Analyzing Twitter posts from 98,000 scientists (2016–22), we uncover substantial political expression. Online experiments with 6,000 U.S. respondents and 135 journalists, rating synthetic academic profiles with different political affiliations, reveal that politically neutral scientists are seen as the most credible. Strikingly, political expressions result in monotonic penalties: Stronger posts reduce perceived credibility of scientists and their research and audience engagement more, particularly among oppositely aligned respondents. Two surveys with scientists highlight their awareness to penalties, their perceived benefits, and a consensus on limiting political expression outside their expertise.
- 21 April 2026
- Published by Springer Nature in Research Square
Open AccessPoliticized Scientists: Credibility Cost of Political Expression on Twitter
As social media becomes prominent within academia, we examine its reputational costs for academics. Analyzing Twitter posts from 98,000 scientists (2016–22), we uncover substantial political expression. Online experiments with 6,000 U.S. respondents and 135 journalists, rating synthetic academic profiles with different political affiliations, reveal that politically neutral scientists are seen as the most credible. Strikingly, political expressions result in monotonic penalties: Stronger posts reduce perceived credibility of scientists and their research and audience engagement more, particularly among oppositely aligned respondents. Two surveys with scientists highlight their awareness to penalties, their perceived benefits, and a consensus on limiting political expression outside their expertise.
- Open Access
- 13 January 2026
- Published by Springer Nature in Research Square
Medical research responds better to disease burden and health shocks, yet global disparities persist
Medical research remains concentrated in high-income settings, raising concerns about alignment with global health needs. Yet systematic evidence on how research responds to both disease burden and acute health shocks across geography and time remains limited. We analyze global research, disease burden and outbreak event history across 204 countries and territories and 15 major disease groups from 1990 to 2021. Research responsiveness to domestic disease burden has more than doubled since 1990, but major gaps persist: maternal–neonatal, nutritional and many infectious diseases remain under-studied, and lower-income countries contribute disproportionately as research settings rather than authors. Funding priorities help explain these patterns. Removing philanthropic-funded research reduces estimated responsiveness growth in lower-income countries by nearly 40%, whereas corporate funding remains focused on profitable chronic diseases. Using an event-study design based on WHO outbreak alerts, we show that health shocks trigger rapid and durable increases in research attention. These responses are substantially stronger in the 2010s than in the 1990s, with response magnitudes shaped in part by digital connectivity, population structure and research capacity. These findings highlight the need for sustained monitoring and intentional funding strategies to better align research with shifting global health risks.
- 13 January 2026
- Published by Springer Nature in Research Square
Open AccessMedical research responds better to disease burden and health shocks, yet global disparities persist
Medical research remains concentrated in high-income settings, raising concerns about alignment with global health needs. Yet systematic evidence on how research responds to both disease burden and acute health shocks across geography and time remains limited. We analyze global research, disease burden and outbreak event history across 204 countries and territories and 15 major disease groups from 1990 to 2021. Research responsiveness to domestic disease burden has more than doubled since 1990, but major gaps persist: maternal–neonatal, nutritional and many infectious diseases remain under-studied, and lower-income countries contribute disproportionately as research settings rather than authors. Funding priorities help explain these patterns. Removing philanthropic-funded research reduces estimated responsiveness growth in lower-income countries by nearly 40%, whereas corporate funding remains focused on profitable chronic diseases. Using an event-study design based on WHO outbreak alerts, we show that health shocks trigger rapid and durable increases in research attention. These responses are substantially stronger in the 2010s than in the 1990s, with response magnitudes shaped in part by digital connectivity, population structure and research capacity. These findings highlight the need for sustained monitoring and intentional funding strategies to better align research with shifting global health risks.
- Open Access
(How) Do Health Shocks Reallocate Research Direction?
We examine whether research systems reallocate scientific effort as health needs change. We assemble a global disease–location panel for 204 countries and territories (1990–2021) by linking diseasespecific publication output to disease burden in the same place and year. Using large language models, we extract diseases from article text, map them into a standardized disease classification, and classify research funders by type. Empirically, we estimate how publication output co-moves with disease burden within countries and diseases over time, and we use event-study difference-in-differences designs that exploit plausibly exogenous variation from the timing of outbreak alerts. We find that responsiveness to endemic burden has increased over time but remains highly uneven across locations; outbreak alerts trigger rapid, statistically significant research surges that have strengthened in recent years; and funding composition is strongly associated with adjustment dynamics, with philanthropic and government-supported research contributing disproportionately to responsiveness growth in lower-income settings.
Open Access(How) Do Health Shocks Reallocate Research Direction?
We examine whether research systems reallocate scientific effort as health needs change. We assemble a global disease–location panel for 204 countries and territories (1990–2021) by linking diseasespecific publication output to disease burden in the same place and year. Using large language models, we extract diseases from article text, map them into a standardized disease classification, and classify research funders by type. Empirically, we estimate how publication output co-moves with disease burden within countries and diseases over time, and we use event-study difference-in-differences designs that exploit plausibly exogenous variation from the timing of outbreak alerts. We find that responsiveness to endemic burden has increased over time but remains highly uneven across locations; outbreak alerts trigger rapid, statistically significant research surges that have strengthened in recent years; and funding composition is strongly associated with adjustment dynamics, with philanthropic and government-supported research contributing disproportionately to responsiveness growth in lower-income settings.
- Open Access
- 9 September 2025
- Published by Springer Nature in Research Square
AI health advice accuracy varies across languages and contexts.
Using basic health statements authorized by UK and EU registers and ~9,100 journalist-vetted public-health assertions on topics such as abortion, COVID-19 and politics from sources ranging from peer-reviewed journals and government advisories to social media and news across the political spectrum, we benchmark seven leading large language models in 21 languages. We find that, despite high accuracy on English-centric textbook claims, performance falls in multiple non-European languages and fluctuates by topic and source. This highlights the urgency of comprehensive multilingual, domain-aware validation before deploying AI in global health communication.
- 9 September 2025
- Published by Springer Nature in Research Square
Open AccessAI health advice accuracy varies across languages and contexts.
Using basic health statements authorized by UK and EU registers and ~9,100 journalist-vetted public-health assertions on topics such as abortion, COVID-19 and politics from sources ranging from peer-reviewed journals and government advisories to social media and news across the political spectrum, we benchmark seven leading large language models in 21 languages. We find that, despite high accuracy on English-centric textbook claims, performance falls in multiple non-European languages and fluctuates by topic and source. This highlights the urgency of comprehensive multilingual, domain-aware validation before deploying AI in global health communication.
- Open Access
- 3 June 2025
- Published by Springer Nature in Nature Human Behaviour
- Vol. 9 (9) , 1-18
Political expression of academics on Twitter
Academics play a vital role in the generation and dissemination of knowledge, ideas and narratives. Social media provide new, more direct ways of science communication. Yet, since not all academics engage with social media, the sample that does so may have an outsized influence on shaping public perceptions of academia through the set topics they engage with and their style and tone of communication. We describe patterns in academics’ expression online using an international dataset covering nearly 100,000 scholars linking their Twitter content to academic records. We document large and systematic variation in politically salient academic expression concerning climate action, cultural and economic concepts. We show that US academics often diverge from the US Twitter population at large in topic focus and style, although academics are not necessarily more extreme in their beliefs. Future work should examine potential impacts on public trust and the reasons why academics express themselves politically on social media.
- 3 June 2025
- Published by Springer Nature in Nature Human Behaviour
- Vol. 9 (9) , 1-18
Open AccessPolitical expression of academics on Twitter
Academics play a vital role in the generation and dissemination of knowledge, ideas and narratives. Social media provide new, more direct ways of science communication. Yet, since not all academics engage with social media, the sample that does so may have an outsized influence on shaping public perceptions of academia through the set topics they engage with and their style and tone of communication. We describe patterns in academics’ expression online using an international dataset covering nearly 100,000 scholars linking their Twitter content to academic records. We document large and systematic variation in politically salient academic expression concerning climate action, cultural and economic concepts. We show that US academics often diverge from the US Twitter population at large in topic focus and style, although academics are not necessarily more extreme in their beliefs. Future work should examine potential impacts on public trust and the reasons why academics express themselves politically on social media.
- 3 May 2025
- Published by Elsevier in Economics Letters
- Vol. 252
Open AccessLocal decline and populism
- Open Access
- 26 March 2025
- Published by Springer Nature in Research Square
Network Determinants of Cross-Border Media Coverage of Natural Disasters
Climate change is increasing the frequency and severity of natural disasters worldwide. Media coverage of these events may be vital to generate empathy and mobilize global populations to address the common threat posed by climate change. Using a dataset of 466 news sources from 123 countries, covering 135 million news articles since 2016, we apply an event study framework to measure cross-border media activity following natural disasters. Our results shows that while media attention rises after disasters, it is heavily skewed towards certain events, notably earthquakes, accidents, and wildfires. In contrast, climatologically salient events such as floods, droughts, or extreme temperatures receive less coverage. This cross-border disaster reporting is strongly related to the number of deaths associated with the event, especially when the affected populations share strong social ties or genetic similarities with those in the reporting country. Achieving more balanced media coverage across different types of natural disasters may be essential to counteract skewed perceptions. Further, fostering closer social connections between countries may enhance empathy and mobilize the resources necessary to confront the global threat of climate change.
- 26 March 2025
- Published by Springer Nature in Research Square
Open AccessNetwork Determinants of Cross-Border Media Coverage of Natural Disasters
Climate change is increasing the frequency and severity of natural disasters worldwide. Media coverage of these events may be vital to generate empathy and mobilize global populations to address the common threat posed by climate change. Using a dataset of 466 news sources from 123 countries, covering 135 million news articles since 2016, we apply an event study framework to measure cross-border media activity following natural disasters. Our results shows that while media attention rises after disasters, it is heavily skewed towards certain events, notably earthquakes, accidents, and wildfires. In contrast, climatologically salient events such as floods, droughts, or extreme temperatures receive less coverage. This cross-border disaster reporting is strongly related to the number of deaths associated with the event, especially when the affected populations share strong social ties or genetic similarities with those in the reporting country. Achieving more balanced media coverage across different types of natural disasters may be essential to counteract skewed perceptions. Further, fostering closer social connections between countries may enhance empathy and mobilize the resources necessary to confront the global threat of climate change.