Skip to content

Kate Edgar

Area of study
Fellowship level
Year awarded
Host university
Medical Statistics
London School of Hygiene & Tropical Medicine
Kate’s background is in Maths and Medical Statistics, which she studied for her MSc in 2018/19 at LSHTM. She currently lives in Harrogate, North Yorkshire.
View profile

Exploring mechanisms of action of complex interventions using trial data and causal inference methods to improve the quality and safety of healthcare for older people. (Project complete)


As the number of older people in the UK population increases, with resulting increases in acute admissions to hospital, there is concern that the National Health Service and health systems in other countries will not be able to respond with timely and high quality healthcare.

This doctoral research project has at its centre a trial of over 1055 older people with markers of frailty or prior dependence, who were referred to hospital for an acute medical event. The trial looked at whether geriatrician-led care in the patient’s home is an effective alternative to hospital admission.

This project provides a unique opportunity to extend current statistical methods for exploring causal pathways and apply them to a topical health system intervention, to improve the quality of healthcare provided to an older population who experience a sudden decline in health.


Specific objectives are,

    1. To update and broaden systematic literature reviews of studies that have compared strategies for avoiding acute admission to hospital and extract quantitative information on relationships that link aspects of the care pathway, as well as identifying challenges in standardisation, replication and scaling.
    2. To develop a better understanding of the healthcare pathways between treatment, outcomes and safety using the full range of intermediate observations in a health systems model to provide evidence to improve the quality of healthcare for this older population.
    3. To explore relationships between exposures, outcomes and safety in the proposed healthcare pathways using statistical methods.
    4. To identify patient characteristics and mechanisms of action of interventions that either directly or indirectly have a causal effect on patient outcomes and the relative effectiveness of acute admission and “Acute care at home” strategies.
    5. To embed the UK “Acute care at home” trial results in the wider evidence base, including data on life-expectancy for different patient groups, to reduce uncertainty in policy decisions.

Research articles
Edgar, K. et al. (2024) Admission avoidance hospital at home. Cochrane Database of Systematic Reviews

Edgar, K., Shepperd, S., Sharples, L. (2022) Hospital at Home, delirium and readmission to hospital: A sequential causal mediation analysis. Submitted as a poster presentation to the International Clinical Trials Methodology Conference, Harrogate

Edgar, K., Shepperd, S., Pendlebury, S., Sharples, L. (2023) Transporting a ‘Hospital at Home’ RCT treatment effect to a target population with differing measures of cognitive impairment and delirium. Submitted as a poster presentation to the International Society for Clinical Biostatistics Conference, Milan

Sign up to receive the latest news, reports and articles from THIS Institute.