The Impact of Healthcare Pressures on the COVID-19 Hospitalisation Fatality Risk in England.

BACKGROUND: As the impact of the SARS-CoV-2 pandemic extends into 2023 and beyond, the treatment and outcomes of infected patients continues to evolve. Unlike earlier in the pandemic there are now further infectious disease pressures placed on hospitals, which influence patient care and triage decisions. METHODS: The manuscript uses individual patient records linked with associated hospital management information of system pressure characteristics to attribute COVID-19 hospitalisation fatality risks (HFR) to patients and hospitals, using generalised additive mixed effects models. RESULTS: Between 01 September 2022 and 09 October 2023, the COVID-19 hospitalisation fatality risk in England was estimated as 12.71% (95% confidence interval (CI) 12.53%, 12.88%). Staff absences had  an adjusted odds ratio of 1.038 (95% CI 1.017, 1.060) associated with the HFR when accounting for patient and hospital characteristics. INTERPRETATION: This observational research presents evidence that a range of local hospital effects can have a meaningful impact on the risk of death from COVID-19 once hospitalised and should be accounted for when reporting estimates. We show that both the patient case mix and hospital pressures impact estimates of patient outcomes.

A comparative study of influenza surveillance systems and administrative data in England during the 2022-2023 season.

Accurate and representative surveillance is essential for understanding the impact of influenza on healthcare systems. During the 2022-2023 influenza season, the Northern Hemisphere experienced its most significant epidemic wave since the onset of the COVID-19 pandemic in 2020. Concurrently, new surveillance systems, developed in response to the pandemic, became available within health services. In this study, we analysed per capita admission rates from National Health Service hospital Trusts across four surveillance systems in England during the winter of 2022-2023. We examined differences in reporting timeliness, data completeness, and regional coverage, modelling key epidemic metrics including the maximum admission rates, cumulative seasonal admissions, and growth rates by fitting generalised additive models at national and regional levels. From modelling the admission rates per capita, we find that different surveillance systems yield varying estimates of key epidemiological metrics, both spatially and temporally. While national data from these systems generally align on the maximum admission rate and growth trends, discrepancies emerge at the subnational level, particularly in the cumulative admission rate estimates, with notable issues observed in London and the East of England. The rapid growth and decay phases of the epidemic contributed to higher uncertainty in these estimates, especially in regions with variable data quality. The study highlights that the choice of surveillance system can significantly influence the interpretation of influenza trends, especially at the subnational level, where regional disparities may mask true epidemic dynamics. Comparing multiple data sources enhances our understanding of the impact of seasonal influenza epidemics and highlights the limitations of relying on a single system.

Assessing the Performance of Machine Learning Methods Trained on Public Health Observational Data: A Case Study From COVID-19.

From early in the coronavirus disease 2019 (COVID-19) pandemic, there was interest in using machine learning methods to predict COVID-19 infection status based on vocal audio signals, for example, cough recordings. However, early studies had limitations in terms of data collection and of how the performances of the proposed predictive models were assessed. This article describes how these limitations have been overcome in a study carried out by the Turing-RSS Health Data Laboratory and the UK Health Security Agency. As part of the study, the UK Health Security Agency collected a dataset of acoustic recordings, SARS-CoV-2 infection status and extensive study participant meta-data. This allowed us to rigorously assess state-of-the-art machine learning techniques to predict SARS-CoV-2 infection status based on vocal audio signals. The lessons learned from this project should inform future studies on statistical evaluation methods to assess the performance of machine learning techniques for public health tasks.

A large-scale and PCR-referenced vocal audio dataset for COVID-19.

The UK COVID-19 Vocal Audio Dataset is designed for the training and evaluation of machine learning models that classify SARS-CoV-2 infection status or associated respiratory symptoms using vocal audio. The UK Health Security Agency recruited voluntary participants through the national Test and Trace programme and the REACT-1 survey in England from March 2021 to March 2022, during dominant transmission of the Alpha and Delta SARS-CoV-2 variants and some Omicron variant sublineages. Audio recordings of volitional coughs, exhalations, and speech were collected in the 'Speak up and help beat coronavirus' digital survey alongside demographic, symptom and self-reported respiratory condition data. Digital survey submissions were linked to SARS-CoV-2 test results. The UK COVID-19 Vocal Audio Dataset represents the largest collection of SARS-CoV-2 PCR-referenced audio recordings to date. PCR results were linked to 70,565 of 72,999 participants and 24,105 of 25,706 positive cases. Respiratory symptoms were reported by 45.6% of participants. This dataset has additional potential uses for bioacoustics research, with 11.3% participants self-reporting asthma, and 27.2% with linked influenza PCR test results.

Forecasting influenza hospital admissions within English sub-regions using hierarchical generalised additive models.

BACKGROUND: Seasonal influenza places a substantial burden annually on healthcare services. Policies during the COVID-19 pandemic limited the transmission of seasonal influenza, making the timing and magnitude of a potential resurgence difficult to ascertain and its impact important to forecast. METHODS: We have developed a hierarchical generalised additive model (GAM) for the short-term forecasting of hospital admissions with a positive test for the influenza virus sub-regionally across England. The model incorporates a multi-level structure of spatio-temporal splines, weekly cycles in admissions, and spatial correlation. Using multiple performance metrics including interval score, coverage, bias, and median absolute error, the predictive performance is evaluated for the 2022-2023 seasonal wave. Performance is measured against autoregressive integrated moving average (ARIMA) and Prophet time series models. RESULTS: Across the epidemic phases the hierarchical GAM shows improved performance, at all geographic scales relative to the ARIMA and Prophet models. Temporally, the hierarchical GAM has overall an improved performance at 7 and 14 day time horizons. The performance of the GAM is most sensitive to the flexibility of the smoothing function that measures the national epidemic trend. CONCLUSIONS: This study introduces an approach to short-term forecasting of hospital admissions for the influenza virus using hierarchical, spatial, and temporal components. The methodology was designed for the real time forecasting of epidemics. This modelling framework was used across the 2022-2023 winter for healthcare operational planning by the UK Health Security Agency and the National Health Service in England.

Seasonal influenza causes a burden for hospitals and therefore it is useful to be able to accurately predict how many patients might be admitted with the disease. We attempted to predict influenza admissions up to 14 days in the future by creating a computational model that incorporates how the disease is reported and how it spreads. We evaluated our optimised model on data acquired during the winter of 2022-2023 data in England and compared it with previously developed models. Our model was better at modelling how influenza spreads and predicting future hospital admissions than the models we compared it to. Improving how influenza admissions are forecast can enable hospitals to prepare better for increased admissions, enabling improved treatment and reduced death for all patients in hospital over winter.

Understanding the leading indicators of hospital admissions from COVID-19 across successive waves in the UK.

Following the end of universal testing in the UK, hospital admissions are a key measure of COVID-19 pandemic pressure. Understanding leading indicators of admissions at the National Health Service (NHS) Trust, regional and national geographies help health services plan for ongoing pressures. We explored the spatio-temporal relationships of leading indicators of hospitalisations across SARS-CoV-2 waves in England. This analysis includes an evaluation of internet search volumes from Google Trends, NHS triage calls and online queries, the NHS COVID-19 app, lateral flow devices (LFDs), and the ZOE app. Data sources were analysed for their feasibility as leading indicators using Granger causality, cross-correlation, and dynamic time warping at fine spatial scales. Google Trends and NHS triages consistently temporally led admissions in most locations, with lead times ranging from 5 to 20 days, whereas an inconsistent relationship was found for the ZOE app, NHS COVID-19 app, and LFD testing, which diminished with spatial resolution, showing cross-correlation of leads between -7 and 7 days. The results indicate that novel surveillance sources can be used effectively to understand the expected healthcare burden within hospital administrative areas though the temporal and spatial heterogeneity of these relationships is a key determinant of their operational public health utility.