Refining epidemiological forecasts with simple scoring rules.

Philos Trans A Math Phys Eng Sci

Department of Electrical Engineering and Electronics, University of Liverpool, Brownlow Hill, Liverpool L69 3GJ, UK.

Published: October 2022

Estimates from infectious disease models have constituted a significant part of the scientific evidence used to inform the response to the COVID-19 pandemic in the UK. These estimates can vary strikingly in their bias and variability. Epidemiological forecasts should be consistent with the observations that eventually materialize. We use simple scoring rules to refine the forecasts of a novel statistical model for multisource COVID-19 surveillance data by tuning its smoothness hyperparameter. This article is part of the theme issue 'Technical challenges of modelling real-life epidemics and examples of overcoming these'.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9376716PMC
http://dx.doi.org/10.1098/rsta.2021.0305DOI Listing

Publication Analysis

Top Keywords

epidemiological forecasts
8
simple scoring
8
scoring rules
8
refining epidemiological
4
forecasts simple
4
rules estimates
4
estimates infectious
4
infectious disease
4
disease models
4
models constituted
4

Similar Publications

Background: Despite an established association between heart failure (HF) and lung cancer (LC), there is limited evidence available regarding mortality patterns among the older (≥65 years) population in the United States.

Methods: The mortality data, spanning 1999 to 2019, was surveyed using the Centers for Disease Control and Prevention's Wide-Ranging Online Data for Epidemiologic Research (CDC WONDER) database with HF and LC identified as underlying or contributing causes of death. Crude and age-adjusted mortality rates (AAMR) were calculated per 100,000 individuals.

View Article and Find Full Text PDF

Background This research examines mortality patterns and the place of death in individuals with chronic rheumatic heart disease (RHD) in the United States, aiming to identify demographic predictors for home or hospice death. Additionally, the study aims to uncover trends in mortality due to RHD and provide a predictive forecast. Methods The study utilized data from the Centers for Disease Control and Prevention (CDC)-Wide-Ranging Online Data for Epidemiologic Research (WONDER) database, which spans 22 years (1999-2020), and was categorized based on place of death, including home or hospice care, inpatient, outpatient, or emergency room deaths, and nursing home facility deaths.

View Article and Find Full Text PDF

Background: Despite government efforts, tuberculosis (TB) remains a major public health threat in Brazil. In 2023, TB incidence was 39.8 cases per 100,000 population, far above the WHO's target of 6.

View Article and Find Full Text PDF

Background: According to World Health Organisation data, on the top ten causes of death in the world in 2019, ischemic heart diseases ranked first, followed by stroke and chronic respiratory diseases. This study aimed to make cross-section of the current mortality rates of indicators of circulatory diseases, ischemic heart diseases and cerebrovascular diseases, access the trends of indicators in Serbia more than two decades and to find the correlation of these observed indicators between males and females.

Methods: Medical indicators were taken from the publicly available Health for all databases that deals with long-term evaluation and monitoring of indicators obtained from national authorities.

View Article and Find Full Text PDF

Objective: There are limited epidemiological data on myocarditis in children aged 0-14 years. This study aims to investigate the trends in incidence, mortality, disability-adjusted life years (DALYs), and corresponding estimated annual percentage change (EAPC) of myocarditis in children aged 0-14 years from 1990 to 2021.

Methods: We utilized the 2021 Global Burden of Disease, Injuries, and Risk Factors Study (GBD) analytical tools to examine the incidence, mortality, and DALYs of myocarditis in children aged 0-14 years, considering factors such as age, sex, region, sociodemographic index (SDI), and data from 204 countries or regions.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!