It is not uncommon to think that drugs can have a negative impact on users and lead to people committing crimes. By studying various factors between drugs and crime, further explore the causal relationship between drugs and crime. Explore how these relationships develop and analyze the relationship between drug use and crime through latent and significant variables. The latent variable model and the explicit variable model are used to represent the causal relationship between drug use and post drug crime. The latent variable model includes pre-existing bias structures as exogenous factors, while the explicit variable model includes drug trafficking or trading volume as influencing variables. Derive the causal relationship between drugs and crime by combining explicit and implicit variable models. For the collected drug crime data, this article uses Python visualization packages bnlearn and pandas for causal inference. Firstly, establish a directed acyclic graph using existing data in the application of bnlearn and pandas libraries; secondly, draw a causal diagram based on experience. Finally, based on the causal diagram, the causal relationships between various factors are derived. The research results indicate that this method enables us to obtain causal relationships from data, providing a beneficial attempt and exploration. Through close collaboration with experts in causal reasoning, we have explored the theoretical roots and causal relationships of drugs and crime, providing valuable insights for drug governance and drug crime research.