Type 1 diabetes is a chronic disease with high mortality rates around the world, and its treatment is not already satisfactory in all patients, in consequence, they have a poor quality of life. In this paper, an insulin treatment for type 1 diabetes patients is analyzed by means of sliding mode control strategies based on the Bergman's minimal model. A meal model is also presented as disturbances affecting the blood glucose. These equations are used with nominal parameters, accepted values for describing a patient with type 1 diabetes. The benefits of using this robust type control is that the system becomes insusceptible to parametric variation, one of the main difficulties to control this illness in real life. The results shown that instead of the parametric variation, the control is able to steer the blood glucose to the accepted range, known as normoglycemia, defined by the American Diabetes Association.