Regular tuberculosis (TB) screening is required for healthcare employees. TB is a potentially deadly disease. Early detection prevents the spread of the disease and helps with treatment. Two types of TB diagnostic tests, which vary in terms of cost and accuracy, are available on the market. Thus, hospitals need to carefully consider the available screening options to effectively monitor TB prevalence at low cost. We develop the first optimization model and solution approach in the literature that can be used to obtain effective TB screening policies where testing decisions are differentiated based on employee characteristics. We categorize healthcare employees based on the department they work in, the specific job they do, and their history of vaccination against TB. We develop a Markov decision process (MDP) model to determine which TB test should be utilized for each employee category to minimize the total costs related to testing, undetected infections, and employees’ time lost due to testing. Due to the size of the problem, we use approximate dynamic programming (ADP) to obtain near-optimal solutions. We analyze the ADP solutions under varying assumptions to develop a screening policy that specifies not only the type of the tests that should be used but also the frequency with which each test should be administered. We further examine the impact of system parameters such as costs and TB prevalence on our policy recommendation via a sensitivity analysis. The results indicate that our approach yields a simple policy that can be used by healthcare providers that do not have the expertise or the resources to develop and solve sophisticated optimization models on an ongoing basis.