We are delighted to announce the recipient of the 2026 Award for Innovation in Statistical Programming and Analytics (AISPAA). This award recognizes individuals and teams who have made significant contributions to the development and enhancement of tools, software, and methodologies that address problems in statistical programming and analytics. It also honors those who have made substantial contributions to the community of statistical programmers and analysts.
This year, we proudly present the award to:
Yunzhe (Bella) Qian, MS
Biostatistician II, Richard A. and Susan F. Smith Center for Outcomes Research, Beth Israel Deaconess Medical Center
Yunzhe (Bella) Qian is recognized for her leadership in developing spatialAtomizeR, the first R package to implement Bayesian atom-based regression models for spatially misaligned areal data. 
She began developing the software while completing her MS in Biostatistics at the Harvard T.H. Chan School of Public Health, where she graduated in 2025. She served as the lead developer of the package and is the primary author of the accompanying software manuscript.
Spatial misalignment occurs when variables are collected or reported over geographic areas whose boundaries do not coincide, for example, when health outcomes are measured by county while policy or environmental variables are measured by congressional district, census tract, or another administrative region. Conventional approaches often redistribute such data deterministically, without adequately accounting for uncertainty. Bayesian atom-based regression models offer a more rigorous alternative, but their computational complexity and extensive spatial bookkeeping have historically made them difficult for applied researchers to use.
spatialAtomizeR addresses these challenges by automating the construction and management of spatial atoms and providing a practical Bayesian framework for analyzing spatially misaligned data. The package extends atom-based regression methods beyond their earlier reliance on Poisson models to support multiple data types, including count, proportion, and continuous variables. It also incorporates correlation among covariates and provides rigorous uncertainty quantification for analyses involving multiple spatially referenced variables.
Implemented using NIMBLE and compiled C++ code, spatialAtomizeR combines advanced statistical methodology with accessible software tools, intuitive syntax, validation procedures, worked examples, and comprehensive documentation. By automating many of the most technically demanding steps, the package makes Bayesian atom-based regression models available to researchers without requiring specialized expertise in spatial computation.
The package has broad potential applications in epidemiology, environmental health, public policy, political science, urban planning, and other fields in which outcomes, exposures, policies, and contextual variables are frequently measured over different geographic boundaries. Through this work, Bella has helped transform spatial misalignment from a specialized methodological obstacle into a tractable analytical problem for applied researchers.
spatialAtomizeR is freely available through:
We congratulate Yunzhe (Bella) Qian on this outstanding achievement and her valuable contribution to statistical programming and analytics. Her work exemplifies the spirit of the AISPAA by combining methodological rigor, computational innovation, open-source software development, and practical impact.
Please join us in celebrating the accomplishments of the 2026 AISPAA recipient!