Long-term trends in forest attributes are typically assessed using strategic inventories such as the US Department of Agriculture (USDA) Forest Service’s Forest Inventory and Analysis (FIA) program. The implicit assumption of any trend analysis is that data are comparable over time. The 1998 Farm Bill tasked FIA with implementing nationally consistent protocols, including a spatially and temporally balanced sample design, whereas previous inventory methods varied spatially and temporally. To disentangle the artifacts of changing inventory designs from real forest change, this study assessed trends at plots that were measured both before and after implementation of the new inventory design in eight western states. Changes in live and dead tree volume, growth, and mortality were evaluated using only co-located plots and then compared with changes observed across entire inventories. The results sometimes differ in magnitude or are even contradictory, demonstrating that historical forest inventories may provide an incomplete picture of reference conditions in some western states.