Background: Trials evaluating behavioral interventions often report numerous outcomes. Synthesizing evidence across multiple outcomes can be problematic. Objectives: To describe methods of prioritizing outcomes and pooling across heterogeneous behavioral and related outcomes; to address concerns about selective reporting of outcomes. Methods: We conducted a systematic review of behavioral counseling interventions to improve diet and/or physical activity for the United States Preventive Services Task Force. Results: We included 64 counseling trials with a large range of outcomes. The number of relevant outcomes ranged from 4 outcomes in the physical activity counseling literature to 50 outcomes in the dietary counseling literature. We worked with the USPSTF, to define a hierarchy of acceptable outcomes, prioritizing intermediate outcomes (e.g., blood pressure, lipids, and weight) over self-reported behavioral outcomes. Thirty-six of the 64 trials reported intermediate outcomes, 32% of physical activity trials, 64% of the dietary trials, and 82% of the combined physical activity and dietary trials reported intermediate outcomes. All trials included some type of behavioral outcome. In the absence of robust reporting of intermediate outcomes, we focused on behavioral outcomes. An extensive audit of outcomes reported in each trial revealed a large variation in behavioral outcomes in both bodies of literature. Using physical activity trials as an example, there were 4 commonly reported categories of self-reported physical activity measures, objective measures were not commonly reported (see Table 1). However, within these 4 categories of outcomes, there was a large variation in the types of measure used (see Table 1). We used a measure of this heterogeneity in our meta-regression to determine if the measurement type influenced on effect size. To address concerns about selective reporting of outcomes, we evaluated the consistency of direction and magnitude of effect across behavioral and intermediate outcomes, as well as across counseling trials. Conclusions: An a priori hierarchy of outcomes is needed when conducting quantitative and qualitative synthesis across trials reporting numerous behavioral outcomes. An audit of trial outcomes is an important step to determine ability and acceptability of pooling outcomes.