Exploring Temporal Patterns of American Foulbrood Disease in Türkiye Through the Seasonal-Trend Decomposition (STL) Method

Abstract

American Foulbrood (AFB), a highly dangerous and fatal disease of honey bees, is accountable for significant economic losses in honey harvesting. This study analyzes temporal changes in reported numbers of AFB outbreaks and cases using a new analytical tool, STL decomposition. Time-series analyses were carried out using data supplied over an 18-year period beginning in January 2005 to uncover trends and the existence of seasonal fluctuations to enable the industry strategies to track and manage this significant bee disease. STL, a seasonal-trend decomposition method developed using locally weighted regression, was used for visual statistical analysis to model the monthly number for AFB outbreaks and cases. A logistic regression model was employed to investigate the importance of the seasonality. A total of 18483 AFB cases were recorded in Turkiye from January 1, 2005 to December 31, 2022, with a mean of 86 cases per month. STL decomposition demonstrated that the outbreak trend cycle was more flexible, whereas the case trend cycle abruptly rose and fell between 2018 and 2019. The STL decomposition revealed that the highest seasonal peaks of AFB incidence occurred in the spring and summer. This disease was more likely to occur throughout these two seasons (OR = 1.66, 95% CI: 1.16-2.37 and OR = 1.93, 95% CI: 1.36-2.75, respectively, with autumn regarded as reference). These findings provide critical insights into seasonal risk periods, enabling more effective timing for surveillance and control efforts, and can support the development of data-driven action plans in combating the disease.