Obesity is becoming increasingly common in modern society. In order to provide a more complete understanding to tackle this disease, it is essential to identify the key factors influencing the body mass index (BMI), the primary variable associated with this condition. Using a dataset on health and nutrition, we adopt the concept of variable importance within the framework of Generalized Additive Models (GAMs) to determine the most impactful variables on BMI. This approach provides a flexible and interpretable way to assess the relative contribution of different variables, offering valuable insights for clinical applications.