Multicollinearity is a relationship or correlation between predictor variables. Multicollinearity can also occur in longitudinal data, which is a combination of cross-section data and time-series data. The impact of multicollinearity causes the influence of the predictor variable on the response variable to be insignificant, the least-squares estimator, and the error to be sensitive to changes in the data. Therefore, the procedure to overcome multicollinearity uses the principal component analysis method. This study aims to model PCA longitudinal data regression with a fixed-effect model that is applied to blood sugar data of diabetic patients with a time span of January 2019 to July 2019 at Ibnu Sina Hospital Makassar City. The results of this study indicate that there are two main components formed from PCA longitudinal data regression modelling with a fixed-effect model. Obtained variable values are systolic blood pressure of -0.007, diastolic blood pressure of -0,016, the body temperature of -0.098, and platelets of 0.005 which affect blood sugar in patients with diabetes.