A Bayesian nonparametric method for detecting rapid changes in disease transmission
Whether an outbreak of infectious disease is likely to grow or dissipate is determined through the time-varying reproduction number, 𝑅<sub>𝑡</sub>. Real-time or retrospective identification of changes in 𝑅<sub>𝑡</sub> following the imposition or relaxation of interventions ca...
| Main Authors: | , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Elsevier
2022
|
| _version_ | 1797110526113742848 |
|---|---|
| author | Creswell, R Robinson, M Gavaghan, D Parag, KV Lei, CL Lambert, B |
| author_facet | Creswell, R Robinson, M Gavaghan, D Parag, KV Lei, CL Lambert, B |
| author_sort | Creswell, R |
| collection | OXFORD |
| description | Whether an outbreak of infectious disease is likely to grow or dissipate is determined through the time-varying reproduction number, 𝑅<sub>𝑡</sub>. Real-time or retrospective identification of changes in 𝑅<sub>𝑡</sub> following the imposition or relaxation of interventions can thus contribute important evidence about disease transmission dynamics which can inform policymaking. Here, we present a method for estimating shifts in 𝑅<sub>𝑡</sub> within a renewal model framework. Our method, which we call EpiCluster, is a Bayesian nonparametric model based on the Pitman– Yor process. We assume that 𝑅𝑡 is piecewise-constant, and the incidence data and priors determine when or whether 𝑅<sub>𝑡</sub> should change and how many times it should do so throughout the series. We also introduce a prior which induces sparsity over the number of changepoints. Being Bayesian, our approach yields a measure of uncertainty in 𝑅<sub>𝑡</sub> and its changepoints. EpiCluster is fast, straightforward to use, and we demonstrate that it provides automated detection of rapid changes in transmission, either in real-time or retrospectively, for synthetic data series where the 𝑅<sub>𝑡</sub> profile is known. We illustrate the practical utility of our method by fitting it to case data of outbreaks of COVID-19 in Australia and Hong Kong, where it finds changepoints coinciding with the imposition of non-pharmaceutical interventions. Bayesian nonparametric methods, such as ours, allow the volume and complexity of the data to dictate the number of parameters required to approximate the process and should find wide application in epidemiology. This manuscript was submitted as part of a theme issue on "Modelling COVID-19 and Preparedness for Future Pandemics". |
| first_indexed | 2024-03-07T07:56:05Z |
| format | Journal article |
| id | oxford-uuid:798eaaed-0bee-4bf5-818e-76f1feaccadf |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:56:05Z |
| publishDate | 2022 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | oxford-uuid:798eaaed-0bee-4bf5-818e-76f1feaccadf2023-08-15T14:56:03ZA Bayesian nonparametric method for detecting rapid changes in disease transmissionJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:798eaaed-0bee-4bf5-818e-76f1feaccadfEnglishSymplectic ElementsElsevier 2022Creswell, RRobinson, MGavaghan, DParag, KVLei, CLLambert, BWhether an outbreak of infectious disease is likely to grow or dissipate is determined through the time-varying reproduction number, 𝑅<sub>𝑡</sub>. Real-time or retrospective identification of changes in 𝑅<sub>𝑡</sub> following the imposition or relaxation of interventions can thus contribute important evidence about disease transmission dynamics which can inform policymaking. Here, we present a method for estimating shifts in 𝑅<sub>𝑡</sub> within a renewal model framework. Our method, which we call EpiCluster, is a Bayesian nonparametric model based on the Pitman– Yor process. We assume that 𝑅𝑡 is piecewise-constant, and the incidence data and priors determine when or whether 𝑅<sub>𝑡</sub> should change and how many times it should do so throughout the series. We also introduce a prior which induces sparsity over the number of changepoints. Being Bayesian, our approach yields a measure of uncertainty in 𝑅<sub>𝑡</sub> and its changepoints. EpiCluster is fast, straightforward to use, and we demonstrate that it provides automated detection of rapid changes in transmission, either in real-time or retrospectively, for synthetic data series where the 𝑅<sub>𝑡</sub> profile is known. We illustrate the practical utility of our method by fitting it to case data of outbreaks of COVID-19 in Australia and Hong Kong, where it finds changepoints coinciding with the imposition of non-pharmaceutical interventions. Bayesian nonparametric methods, such as ours, allow the volume and complexity of the data to dictate the number of parameters required to approximate the process and should find wide application in epidemiology. This manuscript was submitted as part of a theme issue on "Modelling COVID-19 and Preparedness for Future Pandemics". |
| spellingShingle | Creswell, R Robinson, M Gavaghan, D Parag, KV Lei, CL Lambert, B A Bayesian nonparametric method for detecting rapid changes in disease transmission |
| title | A Bayesian nonparametric method for detecting rapid changes in disease transmission |
| title_full | A Bayesian nonparametric method for detecting rapid changes in disease transmission |
| title_fullStr | A Bayesian nonparametric method for detecting rapid changes in disease transmission |
| title_full_unstemmed | A Bayesian nonparametric method for detecting rapid changes in disease transmission |
| title_short | A Bayesian nonparametric method for detecting rapid changes in disease transmission |
| title_sort | bayesian nonparametric method for detecting rapid changes in disease transmission |
| work_keys_str_mv | AT creswellr abayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT robinsonm abayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT gavaghand abayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT paragkv abayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT leicl abayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT lambertb abayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT creswellr bayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT robinsonm bayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT gavaghand bayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT paragkv bayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT leicl bayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission AT lambertb bayesiannonparametricmethodfordetectingrapidchangesindiseasetransmission |