Dose-response modeling of human data typically employs standard analytical methods that include candidate mathematical functions from which the analyst must choose in order to estimate risk or establish toxicity thresholds.  Smoothing regression splines offer a much simpler approach to dose-response modeling and have a number of desirable properties that include flexibility, requiring no pre-specification of a functional form, and achieving optimal model fits under a wide range of circumstances. drsmooth, a freely available R package that models dose-response relationships using smoothing regression splines was initially developed to assess continuous data but has recently been enhanced to allow for assessment of health outcome data that are dichotomous. The objective of this presentation is to compare the exposure-response relationships for several continuous measures (e.g., thyroid stimulating hormone, TSH) that may be associated with adverse health outcomes and a dichotomous endpoint (diabetes) as a function of serum dioxin concentration using conventional regression models and smoothing spline regression in a human population exposed to dioxin. The smoothing spline regression modeling approach, in some cases, provided risk estimates similar to benchmark dose (BMD) values using BMD software, but allows for greater flexibility to assess the data with and without grouping by exposure. Smoothing spline regression can assess nonlinearity and the potential for thresholds in a given dataset. In addition, the drsmooth package contains bilinear (e.g. hockey stick) models for assessing thresholds. A live demonstration of the R tool will be provided, demonstrating the use of this new tool in quantitative risk assessment with its current applications in the study of human dioxin exposure as a case study.