Although literature on chronic pain treatment outcome has made substantial strides in improving the quality of the studies reported, there remain a number of factors that lead to qualification of the generally positive results. In the two previous papers in this series a set of migrating factors was discussed, namely, representativeness of the samples treated in these outcome studies, relapse, and non-compliance with therapeutic recommendations. Additional limitations include the lack of agreement on the criteria on which to base evaluation of the success of treatment outcome and the percentage of treated patients included in follow-up data. In this paper, the most common methods for determining success are described (group effects based on standard and quasi-standard outcome measures). The limitations of this approach are discussed and alternative strategies are presented that focus not only on traditional criteria based on group means but on additional criteria including: (a) importance of change (i.e., clinical vs. statistical significance), (b) proportion of patients who improve, (c) cost, (d) efficiency in treatment delivery, (e) and consumer acceptance and satisfaction.