Effectiveness of Physical Activity Interventions on Return to Work After a Cancer Diagnosis: A Systematic Review and Meta-analysis.

Only 8 studies were included, with 41 to 240 participants each. The authors point to small sample sizes and a lack of intention-to-treat analysis, and downgraded the quality of the evidence by two levels for risk of bias, ending at moderate. The pooled result covered only 4 of the 8 studies, and 5 of the 8 studies found no significant difference. Both trials where return to work was the primary outcome were null. Usual care was rarely described and was not the same across studies, so it is not clear what the exercise was being compared against. Participants had to be well enough to exercise, so patients too unwell to take part are not represented in these results. Most participants were women with breast cancer, all studies were done in high-income countries, and head and neck, thoracic, brain and testicular cancers were not covered, so the results cannot be generalised to all cancer survivors. Return to work was defined differently in every study and measured by self-report in all of them, and most studies treated it as a secondary outcome. The longest follow-up was 18 months, so nothing is known about effects beyond that.

The review was funded by the Institut National du Cancer, France's national cancer institute. The authors report that none of the included studies was sponsored by industry, and used that together with their grey-literature searching to argue there was no serious risk of publication bias.

Do not read the pooled result as proof that exercise gets working cancer survivors back to work. It combined only 4 of the 8 studies, the two trials where return to work was the primary outcome found no difference, and the dose recommendation rests on an analysis that did not reach statistical significance (P = 0.0703).