Daphne Duval

Daphne Duval

Auto-generated profile, based on publicly available information

Daphne Duval

Research, Evidence and Knowledge Division, UK Health Security Agency, London E14 4PU, United Kingdom

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 June 2022 in BMJ
Verified
Journal Article
Open access
PDF file
DDDaphne Duval
Daphne Duval
JPJennifer C Palmer
Jennifer C Palmer
ITIsobel Tudge
Isobel Tudge
NPNicola Pearce-Smith
Nicola Pearce-Smith
EOEmer O’cOnnell
Emer O’cOnnell
RCRachel Clark
Rachel Clark
Objectives To evaluate the potential for long distance airborne transmission of SARS-CoV-2 in indoor community settings and to investigate factors that might influence transmission. Design Rapid systematic review and narrative synthesis. Data sources Medline, Embase, medRxiv, Arxiv, and WHO COVID-19 Research Database for studies published from 27 July 2020 to 19 January 2022; existing relevant rapid systematic review for studies published from 1 January 2020 to 27 July 2020; and citation analysis in Web of Science and Cocites. Eligibility criteria for study selection Observational studies reporting on transmission events in indoor community (non-healthcare) settings in which long distance airborne transmission of SARS-CoV-2 was the most likely route. Studies such as those of household transmission where the main transmission route was likely to be close contact or fomite transmission were excluded. Data extraction and synthesis Data extraction was done by one reviewer and independently checked by a second reviewer. Primary outcomes were SARS-CoV-2 infections through long distance airborne transmission (>2 m) and any modifying factors. Methodological quality of included studies was rated using the quality criteria checklist, and certainty of primary outcomes was determined using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) framework. Narrative synthesis was themed by setting. Results 22 reports relating to 18 studies were identified (methodological quality was high in three, medium in five, and low in 10); all the studies were outbreak investigations. Long distance airborne transmission was likely to have occurred for some or all transmission events in 16 studies and was unclear in two studies (GRADE: very low certainty). In the 16 studies, one or more factors plausibly increased the likelihood of long distance airborne transmission, particularly insufficient air replacement (very low certainty), directional air flow (very low certainty), and activities associated with increased emission of aerosols, such as singing or speaking loudly (very low certainty). In 13 studies, the primary cases were reported as being asymptomatic, presymptomatic, or around symptom onset at the time of transmission. Although some of the included studies were well conducted outbreak investigations, they remain at risk of bias owing to study design and do not always provide the level of detail needed to fully assess transmission routes. Conclusion This rapid systematic review found evidence suggesting that long distance airborne transmission of SARS-CoV-2 might occur in indoor settings such as restaurants, workplaces, and venues for choirs, and identified factors such as insufficient air replacement that probably contributed to transmission. These results strengthen the need for mitigation measures in indoor settings, particularly the use of adequate ventilation. Systematic review registration PROSPERO CRD42021236762.

27 March 2023 in Systematic Reviews
Verified
Journal Article
Open access
PDF file
DDDaphne Duval
Daphne Duval
NPNicola Pearce-Smith
Nicola Pearce-Smith
JPJennifer C Palmer
Jennifer C Palmer
JSJason Kwasi Sarfo-Annin
Jason Kwasi Sarfo-Annin
PRPaul Rudd
Paul Rudd
RCRachel Clark
Rachel Clark
In this letter, we briefly describe how we selected and implemented the quality criteria checklist (QCC) as a critical appraisal tool in rapid systematic reviews conducted to inform public health advice, guidance and policy during the COVID-19 pandemic. As these rapid reviews usually included a range of study designs, it was key to identify a single tool that would allow for reliable critical appraisal across most experimental and observational study designs and applicable to a range of topics. After carefully considering a number of existing tools, the QCC was selected as it had good interrater agreement between three reviewers (Fleiss kappa coefficient 0.639) and was found to be easy and fast to apply once familiar with the tool. The QCC consists of 10 questions, with sub-questions to specify how it should be applied to a specific study design. Four of these questions are considered as critical (on selection bias, group comparability, intervention/exposure assessment and outcome assessment) and the rating of a study (high, moderate or low methodological quality) depends on the responses to these four critical questions. Our results suggest that the QCC is an appropriate critical appraisal tool to assess experimental and observational studies within COVID-19 rapid reviews. This study was done at pace during the COVID-19 pandemic; further reliability analyses should be conducted, and more research is needed to validate the QCC across a range of public health topics.

29 February 2024 in Journal of Public Health
Verified
Journal Article
Open access
PDF file
DDD Duval
D Duval
BEB Evans
B Evans
ASA Sanders
A Sanders
JHJ Hill
J Hill
ASAmeze Simbo
Ameze Simbo
IOI Oliver
I Oliver
Background: Non-pharmaceutical interventions (NPIs) were crucial in the response to the COVID-19 pandemic, although uncertainties about their effectiveness remain. This work aimed to better understand the evidence generated during the pandemic on the effectiveness of NPIs implemented in the UK. Methods: We conducted a rapid mapping review (search date: 1 March 2023) to identify primary studies reporting on the effectiveness of NPIs to reduce COVID-19 transmission. Included studies were displayed in an interactive evidence gap map. Results: After removal of duplicates, 11 752 records were screened. Of these, 151 were included, including 100 modelling studies but only 2 randomized controlled trials and 10 longitudinal observational studies. Most studies reported on NPIs to identify and isolate those who are or may become infectious, and on NPIs to reduce the number of contacts. There was an evidence gap for hand and respiratory hygiene, ventilation and cleaning. Conclusions: Our findings show that despite the large number of studies published, there is still a lack of robust evaluations of the NPIs implemented in the UK. There is a need to build evaluation into the design and implementation of public health interventions and policies from the start of any future pandemic or other public health emergency.

4 November 2025 in Cochrane Evidence Synthesis and Methods
Verified
Journal Article
Open access
ZSZalaya Simmons
Zalaya Simmons
BEBeti Evans
Beti Evans
THTamsyn Harris
Tamsyn Harris
HWHarry Woolnough
Harry Woolnough
LDLauren Dunn
Lauren Dunn
DDDaphne Duval
Daphne Duval
Introduction: Data extraction is a critical but resource‐intensive step of the evidence review process. Whilst there is evidence that artificial intelligence (AI) and large language models (LLMs) can improve the efficiency of data extraction from randomized controlled trials, their potential for other study designs is unclear. In this context, this study aimed to evaluate the performance of a bespoke LLM model pipeline (Retrieval‐Augmented Generation pipeline utilizing LLaMa 3‐70B) to automate data extraction from a range of study designs by assessing the accuracy and reliability of the extractions measured as error types and acceptability. Methods: Accuracy was assessed by retrospectively comparing the LLM extractions against human extractions from a review previously conducted by the authors. A total of 173 data fields from 24 articles (including experimental, observational, qualitative, and modeling studies) were assessed, of which three were used for prompt engineering. Reliability was assessed by calculating the mean maximum agreement rate (the highest proportion of identical returns from 10 consecutive extractions) for 116 data fields from 16 of the 24 studies. An evaluation framework was developed to assess the accuracy and reliability of LLM outputs measured as error types and acceptability (acceptability was assessed on whether it would be usable in real‐world settings if the model acted as one reviewer and a human as a second reviewer). Results: Of the 173 data fields evaluated for accuracy, 68% were rated by human reviewers as acceptable (consistent with what is deemed to be acceptable data extraction from a human reviewer). However, acceptability ratings varied depending on the data field extracted (33% to 100%), with at least 90% acceptability for “objective,” “setting,” and “study design,” but 54% or less for data fields such as “outcome” and “time period.” For reliability, the mean maximum agreement rate was 0.71 (SD: 0.28), with variation across different data fields. Conclusion: This evaluation demonstrates the potential for LLMs, when paired with human quality assurance, to support data extraction in evidence reviews that include a range of study designs. However, further improvements in performance and validation are required before the model can be introduced into review workflows.

19 May 2025 in IFLA Journal
Verified
Journal Article
Closed access
ZSZalaya Simmons
Zalaya Simmons
CBCharlotte Bruce
Charlotte Bruce
STSamuel Thomas
Samuel Thomas
PLPatricia Lacey
Patricia Lacey
WMWendy Marsh
Wendy Marsh
DDDaphne Duval
Daphne Duval
The UK Health Security Agency’s Knowledge and Library Services established an artificial intelligence working group in 2022 to explore the potential applications of artificial intelligence of relevance to its function. This article describes the working group's initial approach to testing and evaluating artificial intelligence and machine-learning-assisted tools for information retrieval and evidence review processes. These include duplicate-reference removal, citation searching, title and abstract screening, full text screening, data extraction and critical appraisal. These exploratory tests have demonstrated varying degrees of potential for implementation, while also contributing to broader discussions on ethical considerations, copyright and licensing issues, the transparency of artificial intelligence methodology, and evidence integrity. Reflecting the exploratory nature of this work, the testing presented in this article focused on identifying potential opportunities and challenges rather than formal validation. This overview outlines the approach taken and insights gained from navigating the evolving artificial intelligence landscape and its potential implications for knowledge, library and evidence services within a public health organisation.

14 May 2025 in medRxiv
Verified
Preprint
Open access
PDF file
MMatthew Bosworth
Matthew Bosworth
ZSZalaya Simmons
Zalaya Simmons
RCRhiannon Cordiner
Rhiannon Cordiner
TKTabitha Kavoi
Tabitha Kavoi
BEBeti Evans
Beti Evans
DDDaphne Duval
Daphne Duval
Objective To assess and summarise the evidence on the health equity impacts of climate change in the UK. Methods We conducted a rapid systematic review (search: February 2024) of primary studies reporting on the health impacts of climate change exposure pathways (hazards; health risks; solutions and responses) for populations at-risk of experiencing health inequalities, identified using the Core20PLUS framework (people experiencing the greatest deprivation; people from protected characteristics groups related to ethnicity, gender reassignment, sexual orientation, and religion or belief; people from inclusion health groups). Results Of the 20,014 records identified, 24 were included. Of these, 20 reported on people experiencing the greatest deprivation, with some evidence that increase in ambient temperature, extreme cold, and extreme heat had similar impacts on mortality in the most and least deprived areas. No conclusions could be drawn for this population group for other outcomes or exposures due to the limited number of studies identified and their methodological quality. In people experiencing homelessness, 2 studies reported that increase in ambient temperature was associated with adverse health impacts but the disproportionate impact on this population group was not assessed. No conclusions could be drawn for ethnic minority groups or people with drug and/or alcohol dependence due to the limited evidence identified. No studies were identified for the remaining population groups. Conclusions We identified important evidence gaps and methodological limitations of the existing evidence on the health equity impacts of climate change the UK that should be used to inform future research in this area.