Identification and Attribution of Weekly Periodic Biases in Epidemiological Time Series Data

Gallagher, Kit; Creswell, Richard; Gavaghan, David; Lambert, Ben

doi:10.1101/2023.06.13.23290903

Cited by 2 publications

(1 citation statement)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The modulation accounts for the weekly periodicity evident in the data and is characterized by two parameters f w and

. This significant periodicity probably arises from processes involved in the reporting of COVID-19 cases and deaths [ 21 ]. Specifically, cases C t are modelled by

1

where

1

where

.…”

Section: Fixed Step Solvers Applied To a Susceptible–infected–recover...mentioning

confidence: 99%

Understanding the impact of numerical solvers on inference for differential equation models

Creswell,

Shepherd,

Lambert

et al. 2024

J. R. Soc. Interface.

Self Cite

View full text Add to dashboard Cite

Most ordinary differential equation (ODE) models used to describe biological or physical systems must be solved approximately using numerical methods. Perniciously, even those solvers that seem sufficiently accurate for the forward problem , i.e. for obtaining an accurate simulation, might not be sufficiently accurate for the inverse problem , i.e. for inferring the model parameters from data. We show that for both fixed step and adaptive step ODE solvers, solving the forward problem with insufficient accuracy can distort likelihood surfaces, which might become jagged, causing inference algorithms to get stuck in local ‘phantom’ optima. We demonstrate that biases in inference arising from numerical approximation of ODEs are potentially most severe in systems involving low noise and rapid nonlinear dynamics. We reanalyse an ODE change point model previously fit to the COVID-19 outbreak in Germany and show the effect of the step size on simulation and inference results. We then fit a more complicated rainfall run-off model to hydrological data and illustrate the importance of tuning solver tolerances to avoid distorted likelihood surfaces. Our results indicate that, when performing inference for ODE model parameters, adaptive step size solver tolerances must be set cautiously and likelihood surfaces should be inspected for characteristic signs of numerical issues.

show abstract

“…The modulation accounts for the weekly periodicity evident in the data and is characterized by two parameters f w and

. This significant periodicity probably arises from processes involved in the reporting of COVID-19 cases and deaths [ 21 ]. Specifically, cases C t are modelled by

1

where

1

where

.…”

Section: Fixed Step Solvers Applied To a Susceptible–infected–recover...mentioning

confidence: 99%

Understanding the impact of numerical solvers on inference for differential equation models

Creswell,

Shepherd,

Lambert

et al. 2024

J. R. Soc. Interface.

Self Cite

View full text Add to dashboard Cite

show abstract

Rapid human movement and dengue transmission in Bangladesh: a spatial and temporal analysis based on different policy measures of COVID-19 pandemic and Eid festival

Islam,

2024

Infect Dis Poverty

View full text Add to dashboard Cite

Background Rapid human movement plays a crucial role in the spatial dissemination of the dengue virus. Nevertheless, robust quantification of this relationship using both spatial and temporal models remains necessary. This study aims to explore the spatial and temporal patterns of dengue transmission under various human movement contexts. Methods We obtained district-wise aggregated dengue incidence data from the Management Information System, Directorate General of Health Services of Bangladesh. The stringency index (SI), along with eight individual policy measures (from the Oxford Coronavirus Government Response Tracker database) and six mobility indices (as measured by Google's Community Mobility Reports) were obtained as human movement indicators. A multi-step correlative modelling approach, including various spatial and temporal models, was utilized to explore the associations of dengue incidence with the SI, fourteen human movement indices and the Eid festival. Results The global Moran’s I indicated significant spatial autocorrelation in dengue incidence during the pre-pandemic (Moran’s I: 0.14, P < 0.05) and post-pandemic periods (Moran’s I: 0.42, P < 0.01), while the pandemic period (2020–2022) showed weaker, non-significant spatial clustering (Moran’s I: 0.07, P > 0.05). Following the pandemic, we identified the emergence of new dengue hotspots. We found a strong negative relationship between monthly dengue incidence and the SI (rspearman: − 0.62, P < 0.01). Through the selection of an optimal Seasonal autoregressive integrated moving average model, we observed that the closure of public transport (β = − 1.66, P < 0.10) and restrictions on internal movement (β = − 2.13, P < 0.10) were associated with the reduction of dengue incidence. Additionally, observed cases were substantially lower than predicted cases during the period from 2020 to 2022. By utilising additional time-series models, we were able to identify in 2023 a rise in dengue incidence associated with the Eid festival intervention, even after adjusting for important climate variables. Conclusions Overall, rapid human movement was found to be associated with increased dengue transmission in Bangladesh. Consequently, the implemention of effective mosquito control interventions prior to large festival periods is necessary for preventing the spread of the disease nationwide. We emphasize the necessity for developing advanced surveillance and monitoring networks to track real-time human movement patterns and dengue incidence. Graphical Abstract

show abstract

Identification and Attribution of Weekly Periodic Biases in Epidemiological Time Series Data

Cited by 2 publications

References 7 publications

Understanding the impact of numerical solvers on inference for differential equation models

Understanding the impact of numerical solvers on inference for differential equation models

Rapid human movement and dengue transmission in Bangladesh: a spatial and temporal analysis based on different policy measures of COVID-19 pandemic and Eid festival

Contact Info

Product

Resources

About