Reliability, validity and sensitivity of a submaximal exercise test to monitor recovery and performance in elite rowers

Abstract

Background: Monitoring recovery and performance indirectly indicates how rowers are coping with heavy training demands, but established tests are impractical for frequent use throughout a season. Submaximal rowing ergometer tests (SRETs) are potential alternatives, but evidence on their monitoring value has not been collated. Using an ergometer training session as an SRET represents a feasible monitoring strategy, but measurement properties of training test sets have yet to be investigated in rowers. Objectives: To synthesise evidence on SRETs used in rowers, and to investigate the reliability, validity and sensitivity of a 6-km submaximal rowing ergometer training test set (6SRT) for monitoring recovery and performance in elite rowers. Study designs: One systematic review, and three measurement studies including repeated measures or cross-sectional observational designs. Setting: Rowing South Africa training centre at the Hillcrest Campus, University of Pretoria. Participants: Highly-trained to world class rowers. Methods: A systematic review was conducted using Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. For measurement studies, the 6SRT was incorporated into the training programme of observed weeks. Reliability and usefulness of 6SRT responses were assessed in 43 rowers using three observations separated by 7 d. To assess its predictive value, multiple linear regression models were developed for estimating 2-km time trial (TT) time, lactate threshold power output (LT PO) and self-reported total recovery (TR), and cross-validated in 11 rowers. Sensitivity and responsiveness of the 6SRT was assessed based on concurrence and correlation with changes in TR over short-term changes in training volume, and 2-km TT time measured across preparation periods in 13 rowers. Main results: Various SRETs have been used in research, with fixed-intensity formats more common than training test sets. Power output (PO)- and heart rate (HR)-derived metrics in SRETs lasting ~5–25 min at ~65–85% 2-km TT PO (~70–90% maximal HR [HRmax]) are most useful for evaluating performance and fitness, while evidence for monitoring recovery is limited. Mean time (~22–24 min), PO (~64–73% 2-km TT PO), exercise HR (HRex, ~82–87% HRmax), and rating of perceived exertion (RPE, 15–16) showed rowers appropriately self-regulated the 6SRT as a heavy aerobic but submaximal (non-exhaustive) training test set. Gross 6SRT intensity ratios PO:HRex and PO:RPE showed excellent reliability (ICC3,1 ≥ 0.95, CV ≤ 3.5%) and ‘acceptable’ to ‘good’ usefulness (typical error ≤ smallest worthwhile change) for identifying small effects. The 6SRT enabled reasonably accurate estimations of 2-km TT time and LT PO (SEE: ~5 s and ~12 W), although 95% limits of agreement were too wide for detecting small differences. The PO:HRex and PO:RPE ratios were sensitive to change across three weeks of higher or lower training volumes, and between preparation periods, and showed responsiveness to seasonal improvements in 2-km TT time. Conclusions: Evidence on SRETs has been collated for researchers or practitioners seeking alternatives to existing tests of performance, aerobic fitness or recovery in rowers. The 6SRT represents a feasible method for monitoring recovery and performance in elite rowers. Further research is required on ergometer training test sets in rowers.