Prediction of the abuse liability of a drug before it reaches the market is complicated by the fact that there are many factors that influence the actual abuse of a drug. Laboratory methods used in humans to assess the abuse liability of the opioids are reviewed and illustrative studies of morphine and the agonist-antagonist opioids, pentazocine, butorphanol, nalbuphine and buprenorphine, are presented. Three assessment methods, subjective effect measurement, self-administration and drug discrimination, provide information relevant to measuring reinforcing efficacy, a major determinant of the degree to which a drug is sought and self-administered by abusers. Physical dependence capacity, which can contribute to sustained drug use, is evaluated in direct addiction and substitution/suppression studies. Withdrawal precipitation studies measure antagonist activity which might limit abuse. The results of testing the agonist-antagonist opioids are generally consistent across these various methods and consistent with historical experience with these drugs, suggesting that these methods are useful in predicting abuse liability of novel opioids.