Sequential detergent fiber assay results used for nutritional ecology research: Evidence of bias since 2012

Abstract Forage quality surveys can provide a variety of insights into the nutritional well‐being of herbivores. Even small differences in how much of the food eaten can be digested and used for life requisites (dry matter digestibility %; DMD) can affect performance of ruminants, and thus methods for determining forage digestibility must be accurate, repeatable, and robust among laboratories and over time. In 2013, we observed levels of DMD as determined from sequential detergent fiber assays with the filter bag method and the ANKOM fiber analyzer 200/200® that were greater than expected given the plant species, season, and environmental setting. Using stockpiled and previously analyzed forage samples from Oregon, Washington, and Idaho, we evaluated whether fiber results had shifted since 2012, potential causes of the shifts, and whether results since 2012 aligned with data on ungulate performance. Beginning in 2012, our results indicated that sequential detergent fiber analysis began significantly overestimating DMD compared to the same samples analyzed using identical methodology prior to 2012. Magnitude of the difference increased as soluble fiber increased and overestimated DMD by up to 31 percentage points. We detected no changes in laboratory procedures (e.g., fiber bag design or origin of chemical solution) nor differences among laboratories using identical methodologies (e.g., Washington State University versus Dairy One) that might explain the shift, and thus were unable to pinpoint the cause. Results obtained prior to 2012 better explained ungulate performance (e.g., body fat, pregnancy, population trends) and thus we developed predictive equations (0.59 ≤ r 2 ≤ 0.88) to align post‐2012 lab results with those obtained before 2012. Shifts in assay results and inconsistencies in laboratory methods have marked potential to affect inferences from studies of ruminant nutritional ecology. We provide recommendations for researchers using sequential detergent fiber analysis to measure DMD.