OBJECTIVES: A study of recently published health economic models found that face/internal validation, while recommended by ISPOR guidelines, was reported in less than half and third party validation (performed by individuals outside of the model building process with no stake in model results) was utilized in less than ¼ of published models.  This study follows up on those primary results to investigate whether the rates of face, internal, and third-party model validation varies by type of publication, type of model, and geographic region. METHODS:   The published models (n=136) from the primary study were categorized by type of publication (HEOR or non-HEOR), by type of model (cost-effectiveness analysis, cost-utility analysis, budget impact analysis, or other), and by geographic region (North America, Europe, or rest of the world).  The rate of face and internal validation for each category was aggregated and compared.  The analysis was also supplemented with additional published modeling studies for categories with limited numbers of studies from the primary analysis.  The primary study’s methodology for determining model validation was followed for the supplemental models. RESULTS:   The percentage of models that were either face/ internally validated varied with respect to the region, publication type, or type of model. Among the three different categories analyzed, each failed to conduct model validation at greater than a 50% rate. More specifically, the universally lowest method of validation by authors in all categories surveyed was through the use of a third party. CONCLUSIONS: Whether models were published in different types of journals, from a variety of countries, or varied by type of model, the rate of validation was similarly low throughout the published scientific literature. While the rate varies among the model characteristics analyzed, the results suggest that, regardless of model characteristic, ISPOR validation guidelines are not widely followed.