2.1 Protocol and Registration
A meta-analysis was conducted following the latest guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) (27) and was registered with the PROSPERO database in General interest, on 15 November 2022 (CRD42022373126). Additional file 1 (Table S1) provides a description of the relevant adaptations.
2.2 Eligibility Criteria
We used PICOS (participants, intervention, comparators, outcomes, and study design) approach (28) for reference to rate studies for eligibility. Since the studies we chose were single-arm trials, only PIO can be used. Table 1 indicates our inclusion/ exclusion criteria.
After removing duplicates, titles and abstracts from relevant articles were reviewed according to the inclusion and exclusion criteria. Full texts of all potentially relevant studies were obtained and processed for inclusion by two independent investigators (HC. L. and WP. Y). Data was blindly collected and recorded in a special worksheet. Study characteristics were recorded.
Table 1. Modified selection criteria (PICOS) used in the meta-analysis
Category
|
Inclusion criteria
|
Exclusion criteria
|
POPULATION
|
Healthy athletes aged under 18 years old
|
Participants with concomitant pathology
|
INTERVENTION
|
Used both s-RPE scales (CR-10 or CR-100) and physical methods (TRIMP) to monitor the TL
|
TL monitoring without s-RPE and HR; without Edward’s TRIMP
|
COMPARATORS
|
Not used
|
Not used
|
OUTCOME
|
Provided data established a correlation between s-RPE and physiological outcome measures (TRIMP) as reference criteria
|
Lack of correlation
|
STUDY DESIGN
|
Single-arm trials
|
Other types of study design
|
OTHERS
|
Original and full-text studies written in English
|
Non-English
|
2.3 Search Strategy
We conducted a systematic literature search in the following databases PubMed, Web of Science, Ovid, Embase, Cochrane, EBSCO, Scopus, and ProQuest. Medical Subject Headings (MeSH) (adolescents) and free terms (s-RPE, RPE session, Rating Perceived Exertion session, RPE, Rate of Perceived Exertion, Rated of Perceived Exertion, young, teen*, validity, concurrent validity, correlation) were used. The Boolean operators AND and OR were also used. (The specific search strategy of different databases was in the supplementary files). The search covered the period from the first publication to November 2022. In addition, reference lists of primary articles were reviewed.
2.4 Data Extraction and Methods of the Review
The following data were extracted from the included articles: authors with the year of publication, study sample size (sex (males and females)), age range (mean and standard deviation (SD)), exercise protocol (type, intensity, total time), exercise modality (the sports that athletes engage in), RPE familiarity (whether pre-familiar with RPE scales), s-RPE scale (e.g., CR-10 and CR-100), physiological criterion, validity criterion and results (i.e., the coefficient between s-RPE and validity criterion). Two reviewers independently appraised papers (WP. Y. and HC. L); a third reviewer (Q. L.) was consulted to resolve disputes. Contact was made with the principal authors if the data to be extracted could not be found. When no response was obtained from the principal authors, data were extracted from the figures of the studies using the WebPlotDigitizer software. Studies that were not written in English, adopted findings from adults, diseases, or animals, and did not use heart rate and s-RPE were excluded.
The PRISMA flowchart was adopted for illustrating the search results. In addition, we also manually checked the reference lists of eligible papers for the possibility of potentially suitable studies.
2.5 Methodological quality and risk of bias assessment
Methodological quality (MQ) was assessed independently by two reviewers (HC. L and WP. Y) using the methodological index for non-randomized studies (MINORS) (29). The MINORS is suitable for single-arm tests and is used to access the methodological quality of criterion-related validity studies. The total score for each study was used to rank the risk of bias as low (13–16), moderate (9–12), or high (0–8).
2.6 Certainty assessment
Two evaluators (WP. Y. and HC. L) independently assessed the certainty of evidence using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach (30) through the GRADE PRO website (https://gradepro.org). GRADE specifies four categories: high, moderate, low, and very low, applied to a body of evidence.
2.7 Statistical analysis
The relationships between s-RPE and TRIMP were computed using the Pearson correlation coefficient (r-value). The strength of relationships was classified as very r ≤ 0.19, very weak; 0.2 ≤ r ≤ 0.39, weak; ≤ 0.4 r ≤ 0.59, moderate; 0.6 ≤ r ≤ 0.79, strong; r ≥ 0.8, very strong (31). For correlation comparison, Fisher's r-to-z transformation was calculated for correcting the r values such that the normal distribution is satisfied. Therefore, after standardization z-values could be used for addition, subtraction, or comparison between data of different units or magnitude. The related formula is: z’ = 0.5[ln(1 +r)—ln(1-r)] (32). The Cochran test was used to determine statistical heterogeneity among publications, and inconsistency was assessed using I2 statistics, defined as I2 = 100% (Q-DF)/Q, where Q is Cochran’s heterogeneity index and DF accounts for degrees of freedom. A value of 0% indicates a lack of heterogeneity, whereas a larger value shows the existence of heterogeneity. For analyses where I2 was below 50%, a fixed effects model was applied and, if I2 was above 50%, a random effects model was employed (33). General criterion-related validity was established using calculated Fisher’s z between s-RPE and HR, and different scales of s-RPE were analyzed in the subgroup. A funnel plot (34) was constructed to evaluate publication bias, and the Begg and Egger’s tests were used. A 95% confidence interval (95% CI) was reported and the significance level was set at p<0.05 according to the analysis conducted in Stata software version 15.1 (Stata Corp, College Station, TX, USA).