Elevated population levels of white-tailed deer (Odocoileus virginianus Zimmerman) can drastically alter forest ecosystems and negatively impact society through human interactions such as deer vehicle collisions. It is currently difficult to estimate deer populations at multiple scales ranging from stand, county, state, and regional levels. This presents a challenge as natural resource managers develop silvicultural prescriptions and forest management practices aimed at successfully regenerating tree species in the face of deer browsing. This study utilized measurements of deer browse impact from the new tree regeneration indicator developed by the United States Department of Agriculture Forest Service Forest Inventory and Analysis (FIA) program. Seedling and sapling abundance and other plot-level characteristics were analyzed across three states (Michigan, Minnesota, and Wisconsin) in the Great Lakes Region of the United States. Socio-environmental datasets (Lyme disease cases, deer vehicle collisions, and deer density estimates) were used in conjunction with FIA data to determine their predictive power in estimating deer browse impacts by county. Predictions from random forests models indicate that using Lyme disease case reports, the number of deer-vehicle collisions, deer density estimates, and forest inventory information correctly predicted deer browse impact 70–90% of the time. Deer-vehicle collisions per county ranked highly important in the random forests for predicting deer browse impacts in all three states. Lyme disease cases ranked high in importance for the Lake States combined and for Minnesota and Wisconsin, separately. Results show the effectiveness of predicting deer browse impacts using a suite of freely available forest inventory and other socio-environmental information.