Public health surveillance systems in Nigeria have been established to monitor infectious diseases such as tuberculosis (TB). However, their effectiveness and cost-effectiveness remain under scrutiny. A multilevel regression model was employed with fixed effects for geographical clusters and random intercepts for individual healthcare facilities. Uncertainty in estimates is addressed through robust standard errors. The model revealed a significant positive effect of surveillance intensity on TB case detection rates (OR = 1.05, $p$-value < 0.001), with moderate precision around the estimate. Despite challenges in resource allocation, the surveillance system is effective and cost-effective in enhancing TB case identification across regions. Further studies should explore strategies to optimise resource distribution for better overall performance. Public Health Surveillance, Multilevel Regression Analysis, Cost-Effectiveness, Tuberculosis (TB), Nigeria