Forecasting influenza hospital admissions within English sub-regions using hierarchical generalised additive models
Abstract 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 forec...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-12-01
|
| Series: | Communications Medicine |
| Online Access: | https://doi.org/10.1038/s43856-023-00424-4 |
| _version_ | 1827326398506729472 |
|---|---|
| author | Jonathon Mellor Rachel Christie Christopher E. Overton Robert S. Paton Rhianna Leslie Maria Tang Sarah Deeny Thomas Ward |
| author_facet | Jonathon Mellor Rachel Christie Christopher E. Overton Robert S. Paton Rhianna Leslie Maria Tang Sarah Deeny Thomas Ward |
| author_sort | Jonathon Mellor |
| collection | DOAJ |
| description | Abstract 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. |
| first_indexed | 2024-03-07T14:43:55Z |
| format | Article |
| id | doaj.art-a598b2dabf744195ab09cd260b975e9d |
| institution | Directory Open Access Journal |
| issn | 2730-664X |
| language | English |
| last_indexed | 2024-03-07T14:43:55Z |
| publishDate | 2023-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Medicine |
| spelling | doaj.art-a598b2dabf744195ab09cd260b975e9d2024-03-05T20:05:42ZengNature PortfolioCommunications Medicine2730-664X2023-12-013111210.1038/s43856-023-00424-4Forecasting influenza hospital admissions within English sub-regions using hierarchical generalised additive modelsJonathon Mellor0Rachel Christie1Christopher E. Overton2Robert S. Paton3Rhianna Leslie4Maria Tang5Sarah Deeny6Thomas Ward7UK Health Security Agency, Data Analytics and Surveillance, 10 South ColonnadeUK Health Security Agency, Data Analytics and Surveillance, 10 South ColonnadeUK Health Security Agency, Data Analytics and Surveillance, 10 South ColonnadeUK Health Security Agency, Data Analytics and Surveillance, 10 South ColonnadeUK Health Security Agency, Data Analytics and Surveillance, 10 South ColonnadeUK Health Security Agency, Data Analytics and Surveillance, 10 South ColonnadeUK Health Security Agency, Data Analytics and Surveillance, 10 South ColonnadeUK Health Security Agency, Data Analytics and Surveillance, 10 South ColonnadeAbstract 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.https://doi.org/10.1038/s43856-023-00424-4 |
| spellingShingle | Jonathon Mellor Rachel Christie Christopher E. Overton Robert S. Paton Rhianna Leslie Maria Tang Sarah Deeny Thomas Ward Forecasting influenza hospital admissions within English sub-regions using hierarchical generalised additive models Communications Medicine |
| title | Forecasting influenza hospital admissions within English sub-regions using hierarchical generalised additive models |
| title_full | Forecasting influenza hospital admissions within English sub-regions using hierarchical generalised additive models |
| title_fullStr | Forecasting influenza hospital admissions within English sub-regions using hierarchical generalised additive models |
| title_full_unstemmed | Forecasting influenza hospital admissions within English sub-regions using hierarchical generalised additive models |
| title_short | Forecasting influenza hospital admissions within English sub-regions using hierarchical generalised additive models |
| title_sort | forecasting influenza hospital admissions within english sub regions using hierarchical generalised additive models |
| url | https://doi.org/10.1038/s43856-023-00424-4 |
| work_keys_str_mv | AT jonathonmellor forecastinginfluenzahospitaladmissionswithinenglishsubregionsusinghierarchicalgeneralisedadditivemodels AT rachelchristie forecastinginfluenzahospitaladmissionswithinenglishsubregionsusinghierarchicalgeneralisedadditivemodels AT christophereoverton forecastinginfluenzahospitaladmissionswithinenglishsubregionsusinghierarchicalgeneralisedadditivemodels AT robertspaton forecastinginfluenzahospitaladmissionswithinenglishsubregionsusinghierarchicalgeneralisedadditivemodels AT rhiannaleslie forecastinginfluenzahospitaladmissionswithinenglishsubregionsusinghierarchicalgeneralisedadditivemodels AT mariatang forecastinginfluenzahospitaladmissionswithinenglishsubregionsusinghierarchicalgeneralisedadditivemodels AT sarahdeeny forecastinginfluenzahospitaladmissionswithinenglishsubregionsusinghierarchicalgeneralisedadditivemodels AT thomasward forecastinginfluenzahospitaladmissionswithinenglishsubregionsusinghierarchicalgeneralisedadditivemodels |