Neonatal pathological jaundice (NPJ) is easy to cause bilirubin encephalopathy, which has high mortality and sequelae rate. Therefore, accurate risk evaluation can help clinicians take appropriate measures to timely intervene in neonatal jaundice level and avoid complications. In this article, five indexes are extracted as the factor set for the risk evaluation of NPJ, and the diagnostic criteria are determined. Then, five index sets are described by interval type-2 fuzzy sets, and the corresponding membership functions and membership function figures are provided. The feasibility of interval type-2 fuzzy comprehensive evaluation in risk evaluation of NPJ is demonstrated through example, and the fuzzy rule bases of prevention and treatment for NPJ are constructed according to the results of risk evaluation. Finally, this article demonstrates that the proposed risk evaluation and treatment process of NPJ is actually a dynamic closed-loop control process, which is consistent with the clinical treatment process. This article provides a new solution for the aided diagnosis and decision-making treatment of NPJ, which is of great significance in reducing neonatal mortality and alleviating the pressure of medical staff.