Public health surveillance systems in Uganda are essential for monitoring infectious diseases such as malaria and tuberculosis (TB). However, their efficiency varies over time, necessitating methodological evaluations. The study employs a Vector Autoregression (VAR) model to forecast trends in surveillance data, with robust standard errors accounting for uncertainty in the forecasts. A significant proportion (35%) of forecasting accuracy was achieved, indicating room for efficiency improvement. Variability across different time-series suggests targeted interventions are needed to optimise system performance. The VAR model demonstrated its potential as a tool for evaluating surveillance systems' effectiveness in Uganda, with specific recommendations for enhancing data collection and analysis strategies. Integrate real-time feedback loops into the surveillance process to improve timeliness and accuracy. Conduct targeted training sessions for surveillance staff to address identified knowledge gaps. Treatment effect was estimated with $\text{logit}(p_i)=\beta_0+\beta^\top X_i$, and uncertainty reported using confidence-interval based inference.