Compare methods

	Besag-York-Mollie Model	Small-Area Health Estimation
Field	Spatial Epidemiology	Social Epidemiology
Family	Regression model	Regression model
Year of origin	1991	1979
Originator	Julian Besag, Jeremy York & Annie Mollie (BYM2 by Riebler, Sorbye, Simpson & Rue)	Robert E. Fay & Roger A. Herriot; J. N. K. Rao & Isabel Molina
Type	Hierarchical Bayesian Poisson model with structured and unstructured spatial random effects	Model-based estimator for reliable indicators in data-sparse areas
Seminal source	Besag, J., York, J., & Mollie, A. (1991). Bayesian image restoration, with two applications in spatial statistics. Annals of the Institute of Statistical Mathematics, 43(1), 1-20.	Fay, R. E., & Herriot, R. A. (1979). Estimates of Income for Small Places: An Application of James-Stein Procedures to Census Data. Journal of the American Statistical Association, 74(366), 269-277.
Aliases	BYM Model, Convolution Prior Model, CAR Convolution Model, BYM2 Reparameterization	Small Area Estimation for Health, Fay-Herriot Health Estimation, Model-Based Small-Area Prevalence, Local Health Indicator Estimation
Related	4	3
Summary	The Besag-York-Mollie (BYM) model is the workhorse hierarchical Bayesian model for small-area disease mapping. Proposed by Julian Besag, Jeremy York, and Annie Mollie (1991), it models area-level disease counts with a Poisson likelihood whose log relative risk is the sum of two random effects: a spatially structured component, given an intrinsic conditional autoregressive (ICAR) prior that borrows strength from neighboring areas, and an unstructured component capturing area-specific heterogeneity that is not spatially patterned. This convolution of structured and unstructured effects lets the model smooth noisy small-area rates toward local and global means while distinguishing genuine spatial trend from independent overdispersion. Because the original parameterization makes the two variance components hard to interpret and depends on the graph, Riebler, Sorbye, Simpson, and Rue (2016) introduced the scaled BYM2 reparameterization, which mixes a scaled spatial effect and an unstructured effect through a single interpretable mixing parameter and a total-variance parameter, improving prior specification and identifiability.	Small-area estimation produces reliable health indicators for places where the survey sample is too thin to support a trustworthy direct estimate. A national health survey may interview only a handful of people in a given county or census tract, so a county-level prevalence computed straight from the data swings wildly from area to area. The model-based solution, pioneered by Robert Fay and Roger Herriot in 1979 for estimating income in small places, is to borrow strength: combine each area's noisy direct estimate with a regression prediction built from auxiliary variables that are known for every area, weighting the two by their relative reliability. Rao and Molina's comprehensive treatment codified this area-level mixed model and its variants as the foundation of small area estimation. Applied to public health, the approach underpins local prevalence maps for chronic disease and health behaviors, such as the CDC PLACES project, that decision-makers use to target resources at neighborhood and county scale.