Changes in mortality patterns and place of death during the COVID-19 pandemic: A descriptive analysis of mortality data across the UK

Changes in mortality patterns and place of death during the COVID-19 pandemic: A descriptive analysis of mortality data across four nations.
Palliat Med. 2021 Aug 23:2692163211040981. doi: 10.1177/02692163211040981.

• Analysis of patterns of mortality including place of death in the United Kingdom (UK) during the COVID-19 pandemic to date found that where people died changed, with an increase in deaths at home during and between pandemic waves.

Projecting the effect of easing societal restrictions on non-COVID-19 emergency demand in the UK

Projecting the effect of easing societal restrictions on non-COVID-19 emergency demand in the UK: Statistical inference using public mobility data. 

Int J Health Plann Manage. 2021 Jul 1. doi: 10.1002/hpm.3265. Epub ahead of print. PMID: 34212400.

• A study using activity data for A&E attendances and emergency admissions for all hospitals within the Bristol, North Somerset and South Gloucestershire healthcare system were regressed upon publicly available mobility data obtained from Google's Community Mobility Reports for the local area. The models were used to predict non-COVID-19 emergency demand under the UK Government's plan to sequentially lift all restrictions. 
• The study concludes that non-COVID-19 emergency demand associates with the level of societal restrictions, with rates of public mobility representing a key determinant.

Modelling tool to support decision-making in the NHS Health Check programme

Modelling tool to support decision-making in the NHS Health Check programme: workshops, systematic review and co-production with users.
Health Technology Assessment 2021;25(35)

• This report discusses how delivery of the NHS Health Checks programme could be improved, in particular by the use of a web-based tool - workHORSE (working Health Outcomes Research Simulation Environment). A series of workshops with commissioners resulted in a useful ‘real-world’ tool for local commissioners that can calculate the current and potential future benefits of different programmes. The study examined different delivery programmes and their benefit in health and value and impact and reducing inequalities.

Use of a Telemedicine Risk Assessment Tool to Predict the Risk of Hospitalization of 496 Outpatients With COVID-19: Retrospective Analysis.

Use of a Telemedicine Risk Assessment Tool to Predict the Risk of Hospitalization of 496 Outpatients With COVID-19: Retrospective Analysis.
JMIR Public Health and Surveillance. 2021 Apr;7(4):e25075. DOI: 10.2196/25075.

• A COVID-19 telemedicine home monitoring program serving health care workers and the community in Atlanta, Georgia USA, prospectively applied to an outpatient population (n=496) with COVID-19 identified populations with low, intermediate, and high risk of hospitalization.

Predicting future state for adaptive clinical pathway management

Predicting future state for adaptive clinical pathway management. 

J Biomed Inform. 2021 May;117:103750. doi: 10.1016/j.jbi.2021.103750. Epub 2021 Mar 25. 

• This paper introduces weighted state transition logic, a logic to model state changes based on actions planned in clinical pathways. Weighted state transition logic extends linear logic by taking weights – numerical values indicating the quality of an action or an entire clinical pathway – into account. It allows us to predict the future states of a patient and it enables adaptive clinical pathway management based on these predictions.

NHSScotland COPD Support Service: remote and self-management of high-risk patients with COPD using a web app and machine learning predictive modelling

NHSScotland COPD Support Service: remote and self-management of high-risk patients with COPD using a web app and machine learning predictive modelling
Scottish Health Technology Group, Healthcare Improvement Scotland February 2021

• The NHSScotland COPD Support Service comprises a web app which is in use and machine learning predictive modelling which is still in development. The purpose of the modelling is to give real time predictions of how likely a patient is to have an exacerbation in the next 72 hours, their risk of readmission to hospital in the next 3 months and their 12-month mortality risk. As well as giving a measure of risk, the machine learning predictive modelling categorises the user of the web app as either high or low priority for these three outcomes, alerting a clinician to a negative change in condition and allowing them to act proactively.

Data modelling tool can forecast vulnerability of local populations to COVID-19

Data modelling tool can forecast vulnerability of local populations to COVID-19
NIHR 11 February 2021

• Researchers have analysed 6,789 small areas in England and assessed the association between COVID-19 mortality in each area and five vulnerability measures relating to ethnicity, poverty, and prevalence of long-term health conditions, living in care homes and living in overcrowded housing. They developed a Small Area Vulnerability Index (SAVI) modelling tool, which forecasts the vulnerability of the local population to the virus.

Impact of vaccination by priority group on UK deaths, hospital admissions and intensive care admissions from COVID-19

Impact of vaccination by priority group on UK deaths, hospital admissions and intensive care admissions from COVID-19.
Anaesthesia. 2021 Feb 11. doi: 10.1111/anae.15442.

• Researchers developed a model of how the vaccination program in the UK is likely to impact on COVID‐19‐related deaths, hospital admissions and ICU admissions among adults. The model shows vaccination will have a much slower impact on hospital and ICU admissions than on deaths. The model suggests substantial reductions in hospital and ICU admissions will not occur until late March and into April 2021.

Estimating the COVID-19 epidemic trajectory and hospital capacity requirements in South West England: a mathematical modelling framework.

Estimating the COVID-19 epidemic trajectory and hospital capacity requirements in South West England: a mathematical modelling framework.
BMJ Open. 2021 Jan 7;11(1):e041536. doi: 10.1136/bmjopen-2020-041536.

• A regional model of COVID-19 dynamics has been developed for use in estimating the number of infections, deaths and required acute and intensive care (IC) beds using the South West England as an example.

Mapping Population Vulnerability and Community Support during COVID-19

Mapping Population Vulnerability and Community Support during COVID-19 a case study from Wales
International Journal of Population Data Science Vol. 5 No. 4 (2020): IJPDS Special Issue: Population Data Science for COVID-19

• This paper describes the development of an accessible interactive map of the citizen-led community response to need during the COVID-19 pandemic in Wales, UK that combines information gathered from multiple data providers to reflect different interpretations of need and support.

Artificial intelligence driven assessment of routinely collected healthcare data is an effective screening test for COVID-19 in patients presenting to hospital

Artificial intelligence driven assessment of routinely collected healthcare data is an effective screening test for COVID-19 in patients presenting to hospital [Preprint]
medRxiv 8 July 2020

• As an alternative to RT-PCR testing, the authors from Oxford University describe the development of two early-detection models to identify COVID-19 using routinely collected data typically available within one hour.
• On 115,394 emergency presentations and 72,310 admissions to a large UK teaching hospital, their emergency department (ED) model achieved 77.4% sensitivity and 95.7% specificity (AUROC 0.939) for COVID- 19 amongst all patients attending hospital, and admissions model achieved 77.4% sensitivity and 94.8% specificity (AUROC 0.940) for the subset admitted to hospital.
• Both models achieve high negative predictive values (>99%) across a range of prevalences (<5%), facilitating rapid exclusion during triage to guide infection control.

National COVID-19 Chest Imaging Database (NCCID)

National COVID-19 Chest Imaging Database (NCCID)
NHSX

• The National COVID-19 Chest Imaging Database (NCCID) is a centralised UK database containing X-Ray, CT and MRI images from hospital patients across the country. This is to support a better understanding of the COVID-19 virus and develop technology which will enable the best care for patients hospitalised with a severe infection. It is a joint initiative established by NHSX, the British Society of Thoracic Imaging, Royal Surrey NHS Foundation Trust and Faculty. It is being made available to all those wanting to investigate the disease and develop solutions that can support the COVID-19 patient care pathway.
• A database of chest scans that could lay the groundwork for AI-powered assessments of coronavirus patients is being built by Faculty, which is also involved in the development of a data store to assess and predict demand on the NHS.

Forecasting spatial, socioeconomic and demographic variation in COVID-19 health care demand in England and Wales.
BMC Med 18, 203 (29 June 2020). https://doi.org/10.1186/s12916-020-01646-2

• By combining multiple sources, the researchers have produced geospatial risk maps on an online dashboard that dynamically illustrate how the pre-crisis health system capacity matches local variations in hospitalization risk related to age, social deprivation, population density and ethnicity, also adjusting for the overall infection rate and hospital capacity.

Abstract

Adapting hospital capacity to meet changing demands during the COVID-19 pandemic

Adapting hospital capacity to meet changing demands during the COVID-19 pandemic
MRC Centre for Global Infectious Disease Analysis, Imperial College London 15 June 2020

• In this report, we aim to calculate hospital capacity for emergency treatment of COVID-19 and other patients during the pandemic surge in April and May 2020; to evaluate the increase in capacity achieved via five interventions (cancellation of elective surgery, field hospitals, use of private hospitals, and deployment of former and newly qualified medical staff); and to determine how to re-introduce elective surgery considering continued demand from COVID-19 patients.

COVID-19 Rapid Summary: Modelling the Pandemic

COVID-19 Rapid Summary: Modelling the Pandemic
Lords Select Committee 8 June 2020

• The House of Lords Science and Technology Committee are producing rapid summaries of oral evidence sessions, aimed at providing the public with the latest information on the science of COVID-19.
• The second summary covers the Committee meeting of Tuesday 2 June, where we spoke to experts about the epidemiological models that have contributed to the pandemic response in the UK, and about approaches for future stages of the pandemic.

Planning Hospital Needs for Ventilators and Respiratory Therapists in the COVID-19 Crisis

Planning Hospital Needs for Ventilators and Respiratory Therapists in the COVID-19 Crisis
RAND May 2020

• Description of a model that can be used to calculate the number of ventilators and RTs needed to achieve a target wait time — the average delay for a ventilator experienced by a new patient.

Covid-19 Mortality Risk Indicator.

Covid-19 Mortality Risk Indicator
Health Ttech Newsletter 27 May 2020

• Scientists from University College London, UCL Hospitals NHS Trust, the University of Cambridge and Health Data Research U.K. have developed an online tool to help calculate and understand the risks of COVID-19. Using over 3.8 million health records, researchers developed the tool to estimate the excess risk of mortality which may be associated with COVID-19.
• The aims and initial learnings are hoped to support the implementation and development of policy responses to the COVID-19 pandemic.

Reporting of artificial intelligence prediction models

Reporting of artificial intelligence prediction models
The Lancet; London Vol. 393(10181), 20 April 2019 [NHS OpenAthens]

• The number of prediction model studies is increasing rapidly, with hundreds of different models being developed for some of the same targeted populations and outcomes.
• "Therefore, the clinical community must not get mesmerised by the artificial intelligence and machine learning revolution, and artificial intelligence and machine learning prediction models must be appropriately developed, evaluated, and—if needed—tailored to different situations before they are used in daily medical practice."

Abstract

Dynamic models to predict health outcomes: current status and methodological challenges

Dynamic models to predict health outcomes: current status and methodological challenges
Diagnostic and Prognostic Research December 2018, v2(23)

• A systematic review identified eleven papers that discussed seven dynamic clinical prediction modelling methods. 
• There were three categories. The first category uses frequentist methods to update models in discrete steps, the second uses Bayesian methods for continuous updating and the third, based on varying coefficients, explicitly describes the relationship between predictors and outcome variable as a function of calendar time.

Abstract