Because different definitions and criteria of a diagnosis lead to different outcomes, the researchers chose to define depression and anxiety by using the definitions in the Diagnostic and Statistical Manual of Mental Disorders (4th ed.) and the criteria of the International Classification of Disease, 9th Revision, Clinical Modification (ICD-9-CM) and ICD-10. In addition, to purify issues of physical activity dose-response in depression and anxiety, researchers only reviewed studies in which depression and anxiety are primary diseases rather than secondary illness or caused by substance abuse. Diseases with strong genetic linkage were also excluded since it is unlikely they can be treated effectively by using only exercise. Therefore, depression includes major depressive disorder, dysthymic disorder, and depressive disorder not otherwise specified; anxiety includes panic disorder without agoraphobia, generalized anxiety disorder, and anxiety disorder not otherwise specified. As for the criteria of clinical outcomes, they are response, remission, prevention of relapse, recovery, and prevention of recurrence.

The studies were collected from Medline, PsychLit, and the Internet, and were conducted using combinations of the following keywords: depressive disorders, depression, anxiety, anxiety disorder, physical activity, and exercise. Reference lists from review articles, book chapters, and personal retrieval systems were also searched. However, articles that did not have specific methods of measurement, only measured acute effects, or studied secondary depression and anxiety were excluded. After all exclusions, 37 articles were selected.