Reliability and Sensitivity of the Power Push-up Test for Upper-Body Strength and Power in 6-15-Year-Old Male Athletes

REPEATABILITY OF SWAY MEASURES IN UPPER EXTREMITY WEIGHT-BEARING

Background

Analysis of upper extremity weight bearing ability is important for athletes as some function largely in a closed chain capacity (e.g., wrestling, football, gymnastics); also, all require closed chain upper extremity function during strength and conditioning. Additionally, in a rehabilitation setting, closed chain upper extremity functional testing is often used as a return to play criterion. Lower extremity sway measures (biomechanical and clinical) have been published widely and have established reliability and validity; however, the reliability of upper extremity sway biomechanical measures has not been investigated to date.

Hypothesis/Purpose

The purpose of this study was to determine the repeatability of a variety of force plate measurements during an upper extremity task in an athletic population. It was hypothesized that variables measuring upper extremity sway in a closed kinetic chain position would have excellent reliability.

Study Design

Cross-sectional.

Methods

All data were collected using a force plate system with commercially available software. Four hundred and ninety healthy Division I athletes were tested for both their dominant and non-dominant upper extremity at one of two testing sessions. Subjects were instructed to stay as still as possible while maintaining a full plank position with one upper extremity on the force plate and the contralateral upper extremity behind their back. Two, 20-second trials were performed for each extremity. Variables measured included average sway velocity (ASV), sway velocity in medial-lateral (SVML) and anterior-posterior (SVAP) directions, sway velocity at 1^st and 2^nd time intervals for AP (VAP1 and 2) and ML (VML1 and 2) directions, and sway frequency in the AP direction for 1^st and 2^nd time intervals (FreAP1 and 2). Intraclass correlation coefficients (ICC_(2,1)) and their 95% confident intervals were calculated for all force plate variables for 980 limbs.

Results

No difference was seen between left and right extremities for any measure (p > 0.05). ICC's ranged from 0.61-0.90 for all variables, indicating moderate to excellent reliability for all variables.

Conclusion

Upper extremity sway biomechanical variables using a force plate system have moderate to excellent reliability. These results are important prior to validation and clinical utilization of these measures including baseline testing, return to play guidelines, and injury prevention parameters.

Level of Evidence

3.

Test-Retest Reliability of Static and Countermovement Power Push-Up Tests in Young Male Athletes

Bohannon, NA, Gillen, ZM, Shoemaker, ME, McKay, BD, Gibson, SM, Cramer, JT. Test-Retest Reliability of Static and Counter-Movement Power Push-Up Tests in Young Male Athletes. J Strength Cond Res 34(9): 2456-2464, 2020-The primary purpose of this study was to evaluate test-retest reliability of the static (SP) and countermovement (CMP) power push-up test in young male athletes. The secondary purpose was to compare the reliability of vertical ground reaction forces versus torque measurements during the power push-up tests. Twenty boys (age = 11.60 ± 1.15 years) performed SPs and CMPs on force plates with the knees as the fulcrum on 2 laboratory visits separated by 2-7 days. Performance measurements included peak force (PF), peak rate of force development (pRFD), peak torque (PT), peak rate of torque development (pRTD), peak power (PP), average power (AP), eccentric impulse (ECC), and concentric impulse (CON) for both power push-up techniques. Age, maturity offset, height, body mass, fat-free mass, and estimated arm cross sectional area were obtained as measurements of growth. Intraclass correlation coefficients (ICC), SEM, coefficients of variation, and minimum detectable changes (MDC) were reported. Only PF (ICC = 0.87-0.88, SEM = 59-84 N) and PT (ICC = 0.89-0.90, SEM = 60-88 N·m) showed acceptable reliability. Neither pRFD, pRTD, PP, AP, ECC, or CON were reliable outcomes. There were no meaningful differences between force-time and torque-time curve measurements. The SP showed slightly lower CVs (33-34%) than the CMP (CVs = 39-40%). Coaches and practitioners would need to see 58-71% increases in upper-body strength measurements evaluated via power push-up on force plates to be 95% confident that the improvements exceeded the measurement variability.

Peak Torque Explains More Unique Variability in Growth Measurements than Rate of Torque Development in Young Boys and Girls

Gillen, ZM, Shoemaker, ME, McKay, BD, Bohannon, NA, Gibson, SM, and Cramer, JT. Peak torque explains more unique variability in growth measurements than rate of torque development in young boys and girls. J Strength Cond Res 34(9): 2507-2514, 2020-This study reported test-retest reliability and evaluated collinearity for isometric leg extension and flexion peak torque (PT) and rate of torque development (RTD) in young boys and girls. Measurements of growth included height, body mass, fat-free mass, maturity offset, and leg extensor and flexor muscle cross-sectional area. Maximal isometric contractions quantified PT and RTD. Intraclass correlation coefficients (ICCs), SEM, coefficients of variation, and minimum detectable changes quantified test-retest reliability. Zero-order correlations and first-order partial correlations evaluated collinearity. Peak torque from leg extension and flexion exhibited ICCs ≥ 0.90, RTD from leg extension and flexion exhibited ICCs ≥ 0.38. Partialing out leg flexion PT reduced the relationships between leg extension PT and growth (rPText, growth.PTflex = 0.392-0.605). Partialing out leg extension PT eliminated the relationships between leg flexion PT and growth (rPTflex, growth.PText = 0.098-0.263). Partialing out leg extension RTD reduced the relationships between PT and growth (rPText, growth.RTDext = 0.516-0.775). Partialing out leg extension PT eliminated the relationships between RTD and growth (|rRTDext, growth.PText| = 0.001-0.148). Leg extension PT was more reliable and explained the most unique variability in growth among young boys and girls. In contrast, RTD was less reliable and was fully accounted for by PT, indicating that RTD may be an unnecessary measurement in studies of young boys and girls.

Test-retest reliability of physical fitness tests among young athletes: The FITescola® battery

Monitoring physical fitness in young athletes is essential to improve physical performance, identify talents and develop injury prevention programs. The aim of this investigation was to analyse the reliability of the physical fitness tests from the FITescola^® battery in young athletes with different sports backgrounds and competition levels. Participants comprised of 138 young athletes (boys n = 92) aged 9 and 18 years old. Eight physical fitness tests assessing six components of physical fitness were performed: cardiorespiratory fitness (PACER); upper body muscular fitness (push-up); flexibility (sit and reach); lower-body muscular fitness (horizontal and vertical jump); agility (4 × 10 shuttle run); and speed (sprint at 20 m and 40 m). Each test was performed twice, with a one-week interval (7 days) between duplicate tests. No differences between duplicate tests were found (p > .05), except for the speed at 20 m in boys (p < .001) and speed at 40 m in girls (p = .006). The battery of tests had good or excellent reliability and concordance of the ICC between the two trials (ICC ≥ 0.75) with exception for the 20 m speed run in girls that presented moderate reliability (ICC = 0.57). The Bland-Altman plots showed high reliability for all the fitness tests for both sexes. Hence, The FITescola^® battery may be a novel tool to assess the physical fitness of large groups of young athletes from different sports backgrounds.

Performance Differences between National Football League and High School American Football Combine Participants

PURPOSE

The purpose of this study was to determine magnitudes of differences for anthropometric and athletic performance scores between high school and elite college-level American football players.

METHOD

Participants included high school-age (n = 3,666) athletes who participated in American football combines, as well as elite college-level (n = 5,537) athletes who participated in the National Football League (NFL) scouting combine. Combine data included position; height; weight; 10-, 20-, and 40-yard dash; pro-agility (PA); L-cone drill (LC); vertical jump (VJ); and broad jump (BJ). Athletes were separated into their respective position group, defensive back (DB), wide receiver (WR), linebacker (LB), quarterback (QB), running back (RB), tight end (TE), defensive line (DL), and offensive line (OL) for analysis of performance differences. Percent differences for each dependent variable were calculated to quantify magnitudes of differences.

RESULTS

NFL combine participants scored 3% to 25% better on all measurements, with the largest differences between weight and VJ (14%-25%).

CONCLUSION

The largest measurement-specific differences between high school-age and elite college-level American football players were body size and power. Although it may seem intuitive that elite college-level players would perform better, these data provide a unique perspective to high school players, parents, and coaches, giving new information to use when designing measurement-specific athletic development programs. Thus, strength and conditioning professionals may benefit from emphasizing increases in muscle mass and power output in strength and conditioning programs.

Normative Reference Values for High School-Aged American Football Players

McKay, BD, Miramonti, AA, Gillen, ZM, Leutzinger, TJ, Mendez, AI, Jenkins, NDM, and Cramer, JT. Normative reference values for high school-aged American football players. J Strength Cond Res 34(10): 2849-2856, 2020-The purpose of the present report was to provide test- and position-specific normative reference values for combine test results based on a large, nationally representative sample of high school-aged American football players in their freshman, sophomore, and junior classes. Cross-sectional anthropometric and performance data were obtained from 12 different high school American football recruiting combines between March 7, 2015, and January 9, 2016, across the United States. Subjects included a sample (n = 7,478) of high school-aged American football athletes in their junior (n = 3,779), sophomore (n = 2,514), and freshman (n = 1,185) classes. The database included combine date, school state, position, class, height, body mass (BM), 40-yard dash, pro-agility, 3-cone, vertical jump, broad jump, and power push-up. Each player self-classified their American football positions as defensive back, defensive end, defensive linemen, linebacker, offensive linemen, quarterback, running back, tight end (TE), or wide receiver. Test- and position-specific normative values were generated by aggregating data from freshman, sophomore, and junior classes. Mean differences were found among classes for all positions and all measurements (p ≤ 0.05), except for TE BM (p > 0.05). Greater differences for all variables were observed from freshman to sophomore classes than from sophomore to junior classes. These normative reference values may provide realistic comparisons and evaluations in performance for young American football players, parents, and coaches with collegiate football aspirations. High school strength and conditioning professionals should use these norms to set attainable goals and reward accomplishments for young football players.