Related Experiment Video
Updated: Sep 16, 2026

A Method of Trigonometric Modelling of Seasonal Variation Demonstrated with Multiple Sclerosis Relapse Data
Published on: December 9, 2015
Quantifying ASF Outbreak Dynamics in the Philippines (2019-2023) Using Joinpoint Regression and the Estimated
Samuel Joseph M Castro1, Roderick T Salvador2, Romeo S Gundran2
1Faculty of Veterinary Medicine, Chiang Mai University, Chiang Mai 50100, Thailand.
Abstract:
African swine fever (ASF) remains a major constraint on swine production in the Philippines, but interpretation of its national trajectory is complicated by changes in surveillance effort and regional heterogeneity. We analyzed domestic-pig surveillance records from August 2019 through December 2023. After exact-row deduplication and validity screening, 50,852 laboratory records were retained, including 8929 ASF-positive records and 324,741 tested samples. Monthly positive-record counts, annual and surveillance-stream-specific sample positivity, geographic and regional trajectories, year-standardized seasonality indices, Bayesian Information Criterion-selected joinpoint regression, and sliding-window Estimated Dissemination Ratios (EDRs) were evaluated. Positive laboratory records peaked at 733 in September 2020. Annual sample positivity reached 20.6% in 2020 and declined to 3.9% in 2023, despite tested-sample volume increasing from 41,035 to 102,333. Regional peaks were staggered from 2020 through 2023, demonstrating asynchronous detected activity. September and October had the highest mean seasonality indices, although their 95% confidence intervals included 1. The selected joinpoint model identified breakpoints in June, September, and December 2020 and August 2022. A first-order autoregressive correlation structure (AR(1)) sensitivity model supported the 2020 changes but not the August 2022 change. The primary 14-day EDR identified 25 sustained expansion and 28 sustained contraction episodes; 7-day estimates were more volatile, whereas 14- and 21-day estimates were more concordant. These findings indicate a transition from intense nationally detected activity in 2020 to lower sample positivity but persistent, recurrent, and regionally heterogeneous activity through 2023. Integrating record counts, testing denominators, regional patterns, joinpoint regression, and EDR provides a reproducible surveillance framework, but does not estimate population prevalence, causal policy effects, or future transmission.
Related Concept Videos
Steps in Outbreak Investigation
Investigation of Disease Outbreaks
Statistical Methods for Analyzing Epidemiological Data
Prevalence and Incidence
Prevalence indicates the proportion of individuals in a population who have a specific disease or health condition at a...
Distributions to Estimate Population Parameter
Design Example: Analyzing Capacity Contours for Flood Risk Assessment