| Patient selection |
| Predominant recruitment of breast and prostate cancer survivors |
| Exclusion of patients with poor baseline functional status |
| Lack of individual physical activity needs-based approach to patient selection |
| Racially and ethnically homogenous population |
| Limited number of studies during active treatment of cancer |
| Lack of information regarding reasons for attrition |
| Lack of information regarding barriers to patient accrual to exercise studies |
| Intervention |
| Not tailored to the type, and stage of cancer or treatment |
| Tends to be of short duration (typically 4–12 weeks) |
| Often cumbersome and expensive |
| Often not personalized to patient needs |
| Not integrated into cancer care or patient lifestyle |
| Not reinforced after the completion of the study |
| Adverse events often not considered or known |
| Not designed with patient input |
| Outcomes |
| Functional outcomes (i.e. driveway walk) not routinely measured |
| Impact of exercise intervention on cancer outcomes not regularly assessed |
| Patient-centered outcomes not prioritized |
| Individualized change from baseline not assessed |
| Biologic pathways are non-existent |
| Not sustained after intervention is complete |
| Long-term effect of the intervention on outcomes is unknown |
| Relation of functional outcomes to the quality of cancer survivorship, cancer treatment, and health care are largely unknown. |
