BackgroundThe TreeScan method is an emerging tool for active safety signal surveillance and has been increasingly applied in post-marketing monitoring of pharmaceuticals and vaccines.ObjectiveTo evaluate methodological developments and application patterns of the TreeScan method.MethodsA scoping review was conducted by searching Embase, Medline, Cochrane Library, China National Knowledge Infrastructure, Wanfang, VIP, and SinoMed from inception to 16 May 2025. Two researchers independently screened studies and extracted data. Descriptive analyses were performed on study characteristics, methodologies, and application domains. Included studies were categorized as methodological or applied research.ResultsForty-four articles were included, comprising 13 methodological studies (29.5%) and 31 applied studies (70.5%). Applied studies included drug safety surveillance (n = 10), vaccine safety surveillance (n = 16), and other areas such as drug repurposing and epidemiology (n = 5). Most studies originated from the United States (n = 28) and South Korea (n = 8). The Bernoulli model (43.2%), Poisson model (18.2%), and tree-temporal scan statistic (29.5%) were the most frequently used approaches. Positive controls we1re used in 63.6% of studies, while 36.4% employed within-group controls. Vaccine safety surveillance represented the most common application area, whereas methodological innovations focused on improving statistical performance, controlling confounding, and extending TreeScan to new data structures.ConclusionTreeScan research is increasingly application-oriented, particularly in vaccine safety surveillance Recent methodological advances have improved its performance in handling confounding, hierarchical outcomes, and complex data structures. Future research should should prioritize validating newer TreeScan variants across diverse real-world databases and expanding applications beyond safety surveillance.