Aerobic Exercise and Cancer-Related Fatigue in Adults: A Reexamination Using the IVhet Model for Meta-analysis.

Abstract

Although the results of a recent meta-analysis using the traditional random effects model yielded a statistically significant standardized mean difference (SMD) reduction in cancer-related fatigue (CRF) as a result of aerobic exercise, a recently developed inverse heterogeneity (IVhet) model has been shown to be more valid than the traditional random effects model. The purpose of this study was to compare these previous meta-analytic results using the IVhet model. Using data from a previous meta-analysis that included 36 SMD effect sizes (ES) representing 2,830 adults (1,426 exercise, 1,404 control), results were pooled using the IVhet model. Absolute and relative differences between the IVhet and random effects results for CRF were also calculated as well as influence analysis with each SMD ES deleted from the IVhet model. Nonoverlapping 95% confidence intervals (CI) were considered statistically significant. A statistically nonsignificant reduction in CRF fatigue was found as a result of aerobic exercise using the IVhet model (SMD, -0.08; 95% CI, -0.31-0.14; P = 0.46). The IVhet model yielded a SMD ES that was 0.14 (63.6%) smaller than the random effects model. With each study deleted from the IVhet model once, results remained statistically nonsignificant with SMD ESs ranging from -0.11 (95% CI, -0.33-0.11) to -0.06 (95% CI, -0.28-0.16). Insufficient evidence currently exists to support the use of aerobic exercise for reducing CRF in adults. Additional studies are needed to determine the certainty of aerobic exercise on CRF in adults. Cancer Epidemiol Biomarkers Prev; 26(2); 281-3. ©2016 AACR.