Many behavioral outcomes are commonly uncovered in a systematic review of studies that focus on the same risk behavior. Although there have been efforts to standardize assessment measures, particularly in areas with substantial behavioral intervention research, such as tobacco use and cessation, more typically several valid measures are used for key behavioral outcomes. To complicate matters further, different assessment methods can be used to compute the same outcome measure. For example, with diet one can calculate the percentage of energy consumed from fats by using food-frequency questionnaires, brief dietary assessments, or 24-hour recalls (22). As evidence emerges about associations between dietary components and disease, different components are emphasized for different diseases, creating further complexity. For example, studies associating diet with incident cardiovascular disease often focus on saturated fat intake (23), whereas studies of diet and cancer focus on the percentage of calories from fat; grams of fiber; and, more recently, fruit and vegetable intake (24). Equally complex are measures of physical activity, which include self-reported surveys; job classification; motion sensors; and physiologic markers, such as doubly labeled water (25). Self-reported measures using methods as varied as detailed diaries, logs, checklists of activities, or global self-reported measures of physical activity are most commonly used (25). Efforts to pool results of behavioral counseling interventions can be hampered by noncomparability of interventions and outcomes and by varying validity of measured and reported outcomes.