The interest in developing synthetic glucocorticoids (GCs) arises from the utility of endogenous steroids as potent anti-inflammatory and immunosuppressant agents. The first GCs to be discovered, such as cortisol or dexamethasone, still represent the main treatment for conditions of the inflammatory process, despite the fact that they carry a significant risk of side effects. Hence, there is a continuing need to find drugs that preserve the immune effects of GCs without the side effects, such as those on metabolism (diabetes), bone tissue (osteoporosis), muscles (myopathy), eyes and skin. In this review, we focus on the recent use of ligand-based computational approaches in glucocorticoid receptor (GR) drug-design efforts for the determination of novel GR ligands. We examine a number of ligand-based (similarity searches, pharmacophore screens and quantitative structure-activity relationships) approaches that have been implemented in recent years. A recent virtual high-throughput screening similarity search was successful in developing a novel series of nonsteroidal GR antagonists. Additionally, there has been considerable success in ligand-based structure-analysis relationship generation and lead optimization studies for the GR. Future trends toward integrated GR ligand design incorporating ligand- and structure-based methodologies are inevitable.