Prediction intervals for early outbreak models

Abstract During epidemics an infectious disease spreads to an unexpectedly high number of cases with consequences for the health system and society at large. In this article, different models were compared for the early phase of an outbreak, focusing on the initial period characterised by approximately exponential infection growth and the subsequent second early phase where the cumulative numbers of infections begins to flatten. This paper focuses on simple models that enable valuable conclusions from limited data. In particular, Gaussian non-linear least squares, Poisson regression, and negative binomial regression with two different mean functions (exponential and Gompertz) are applied. Our analysis includes a phase-specific comparison of model fits and prediction intervals for new infections, assessing coverage accuracy and interval width to address uncertainty in case number predictions. Our results demonstrate that overdispersion in the counts must be considered, as the coverage probability can considerably deviate from the nominal level. Additionally, an improved version of well-known parametric bootstrap prediction intervals is presented for the exponential phase. To the best of our knowledge, we provide the first implementation of parametric bootstrap intervals for models with the Gompertz growth function as mean function.

Authors

Institutions

Publication Details

Journal
BMC Medical Research Methodology
Published
2026-11-17
DOI
https://doi.org/10.1186/s12874-026-03064-z
Primary Topic
Statistical Methods in Epidemiology
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Prediction intervals for early outbreak models

Andreas Leha, Tim Friede, Fabian Kück
1 links
BMC Medical Research Methodology
Statistical Methods in Epidemiology
article

Prediction intervals for early outbreak models

Andreas Leha, Tim Friede, Fabian Kück
article en

Abstract

Abstract During epidemics an infectious disease spreads to an unexpectedly high number of cases with consequences for the health system and society at large. In this article, different models were compared for the early phase of an outbreak, focusing on the initial period characterised by approximately exponential infection growth and the subsequent second early phase where the cumulative numbers of infections begins to flatten. This paper focuses on simple models that enable valuable conclusions from limited data. In particular, Gaussian non-linear least squares, Poisson regression, and negative binomial regression with two different mean functions (exponential and Gompertz) are applied. Our analysis includes a phase-specific comparison of model fits and prediction intervals for new infections, assessing coverage accuracy and interval width to address uncertainty in case number predictions. Our results demonstrate that overdispersion in the counts must be considered, as the coverage probability can considerably deviate from the nominal level. Additionally, an improved version of well-known parametric bootstrap prediction intervals is presented for the exponential phase. To the best of our knowledge, we provide the first implementation of parametric bootstrap intervals for models with the Gompertz growth function as mean function.

BMC Medical Research Methodology
Universitätsmedizin Göttingen (DE)
Georg-August-Universität Göttingen
Good health and well-being
Statistical Methods in Epidemiology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon