Post-activation performance enhancement (PAPE) has demonstrated efficacy in acutely improving athletic performance. However, its distinction from general warm-up (GW) effects remains ambiguous, and experimental designs adopted in most PAPE studies exhibit important limitations. The aims of this work are to (i) examine the effects of research methodology on PAPE outcomes, (ii) explore PAPE outcomes in relation to comparison methods, performance measures, GW comprehensiveness, recovery duration, participants' characteristics, conditioning activity (CA) parameters, and (iii) make recommendations for future PAPE experimental designs on the basis of the results of the meta-analysis. Four databases were searched for peer-reviewed English-language literature. Risk of bias was assessed using a modified Cochrane Collaboration's tool and PEDro scale. PAPE groups were compared with control groups, pre-conditioning activity (pre-CA) performances were compared with post-conditioning activity (post-CA) performances throughout a verification test in PAPE groups, and control groups were compared before and after the "rest" period using a three-level meta-analysis. Further analyses, including subgroup analysis and both linear and nonlinear meta-regression methods, were used to explore the effect of different moderating factors on PAPE magnitude. A subgroup analysis of GW comprehensiveness was conducted using four classification methods. One method classified GW as non-comprehensive (stretching or jogging only), partially comprehensive (stretching, jogging, and low-intensity self-weighted dynamic exercises), and comprehensive (adding maximal or near-maximal intensity CAs to a partially comprehensive GW). The other three classifications were adjusted according to the type and number of GW exercises. Certainty of evidence was assessed using the GRADE approach. The final analysis included 62 PAPE studies (1039 participants, male: n = 857, female: n = 182) with a high risk of bias and low certainty of pooled evidence. A trivial PAPE effect was observed from pre- to post-CA (effect size [ES] = 0.12, 95% CI [0.06 to 0.19], prediction intervals [PI] = - 0.29 to 0.54); a small PAPE effect was observed when compared with a control group (ES = 0.30, 95% CI [0.20 to 0.40], PI [- 0.38 to 0.97]). The slightly greater effect against control resulted from a small decrease in performance in control groups (ES = - 0.08, 95% CI [- 0.13 to - 0.03], PI [- 0.30 to 0.14]), but there was no relationship with between PAPE recovery time (β = - 0.005, p = 0.149). Subgroup analyses showed that PAPE magnitude was greater for non-comprehensive GWs (ES = 0.16) than comprehensive (ES = 0.01) and partially comprehensive GWs (ES = 0.11). In contrast, the control group showed a decline in performance after comprehensive GW (ES = - 0.20). An inverted U-shaped PAPE was noted as a function of recovery time. In some cases, PAPE appeared to manifest at < 1min post CA. Additionally, participants with longer training experience (ES = 0.36) and higher training levels (ES = 0.38) had larger PAPE magnitudes. PAPE effect was higher in females (ES = 0.51) than males (ES = 0.32) and mixed groups (ES = 0.16) but did not reach a significant difference (p > 0.05). Plyometric exercise (ES = 0.42) induced greater PAPE amplitude than traditional resistance exercise (ES = 0.23), maximal isometric voluntary contraction (ES = 0.31) and other CA types (ES = 0.24). Although the overall pooled results for both PAPE pre- versus post-CA and PAPE versus control group comparisons showed significant improvement, the wider and past-zero prediction intervals indicate that future studies are still likely to produce negative results. The comprehensiveness of the GW, the time between GW and the pre-CA test, participant sex, training level, training experience, type of CA, number of CA sets, and recovery time after CA all influence the PAPE magnitude. The PAPE magnitude was trivial after comprehensive GW, but it was greater in studies with a control group (i.e., no CA) because performance decreased over the control period, inflating the PAPE effect. Finally, two theoretical models of PAPE experimental design and suggestions for methodological issues are subsequently presented. Future studies can build on this to further explore the effects of PAPE. The original protocol was prospectively registered (osf.io/v7sbt) with the Open Science Framework.
Read full abstract