What are the potential challenges and benefits of implementing z score thresholds in clinical practice for COPD severity classification, and how might this affect patient outcomes?