While pharmacokinetic/pharmacodynamic relationships for opioids have not been consistently demonstrable or sufficiently predictive, there remain compelling reasons to pursue such relationships. Among the reasons for pursuing pharmacokinetic/ pharmacodynamic relationships is the prospect of predicting the time-action characteristics of new therapeutics on the basis of early studies in normals using pharmacodynamic surrogates for analgesia. The realization of such a model could improve the efficiency of development of analgesics. Four studies involving 98 normals were conducted to determine whether significant and reproducible relationships existed for oxycodone in the form of an oral controlled-release tablet. All studies were analytically blinded and utilized a validated gas chromatographic/mass spectrometric, sensitive (0.2 ng/ml), and specific method for oxycodone (four studies) and oxymorphone (two studies) quantitation in 17 to 20 serial plasma samples over 36 to 48 hours following a single 20 mg (or 40 mg) dose in each study. Concurrent assessments included vital signs and opioid effect VAS questionnaires. The studies demonstrated significant relationships between plasma oxycodone (but not oxymorphone) and pharmacodynamic surrogates (particularly VAS "drug effect") and were predictive of the 12-hour duration of pain control and prompt onset of analgesia subsequently demonstrated in multiple clinical studies involving patients with various pathological pain syndromes. The results suggest that investigators can make earlier, accurate predictions of opioid analgesic pharmacodynamics in patients based on pharmacokinetic/pharmacodynamic studies in normal volunteers.