Magnitude-based Inference is not Bayesian and is not a valid method of inference

Alternatives to the P value: connotations of significance

The statistical significance of a clinical trial analysis result is determined by a mathematical calculation and probability based on null hypothesis significance testing. However, statistical significance does not always align with meaningful clinical effects; thus, assigning clinical relevance to statistical significance is unreasonable. A statistical result incorporating a clinically meaningful difference is a better approach to present statistical significance. Thus, the minimal clinically important difference (MCID), which requires integrating minimum clinically relevant changes from the early stages of research design, has been introduced. As a follow-up to the previous statistical round article on P values, confidence intervals, and effect sizes, in this article, we present hands-on examples of MCID and various effect sizes and discuss the terms statistical significance and clinical relevance, including cautions regarding their use.

Frequentist, Bayesian Analysis and Complementary Statistical Tools for Geriatric and Rehabilitation Fields: Are Traditional Null-Hypothesis Significance Testing Methods Sufficient?

Null hypothesis significant testing (NHST) is the dominant statistical approach in the geriatric and rehabilitation fields. However, NHST is routinely misunderstood or misused. In this case, the findings from clinical trials would be taken as evidence of no effect, when in fact, a clinically relevant question may have a "non-significant" p-value. Conversely, findings are considered clinically relevant when significant differences are observed between groups. To assume that p-value is not an exclusive indicator of an association or the existence of an effect, researchers should be encouraged to report other statistical analysis approaches as Bayesian analysis and complementary statistical tools alongside the p-value (eg, effect size, confidence intervals, minimal clinically important difference, and magnitude-based inference) to improve interpretation of the findings of clinical trials by presenting a more efficient and comprehensive analysis. However, the focus on Bayesian analysis and secondary statistical analyses does not mean that NHST is less important. Only that, to observe a real intervention effect, researchers should use a combination of secondary statistical analyses in conjunction with NHST or Bayesian statistical analysis to reveal what p-values cannot show in the geriatric and rehabilitation studies (eg, the clinical importance of 1kg increase in handgrip strength in the intervention group of long-lived older adults compared to a control group). This paper provides potential insights for improving the interpretation of scientific data in rehabilitation and geriatric fields by utilizing Bayesian and secondary statistical analyses to better scrutinize the results of clinical trials where a p-value alone may not be appropriate to determine the efficacy of an intervention.

KeepRunning: A MoCap-Based Rapid Test to Prevent Musculoskeletal Running Injuries

The worldwide popularisation of running as a sport and recreational practice has led to a high rate of musculoskeletal injuries, usually caused by a lack of knowledge about the most suitable running technique for each runner. This running technique is determined by a runner's anthropometric body characteristics, dexterity and skill. Therefore, this study aims to develop a motion capture-based running analysis test on a treadmill called KeepRunning to obtain running patterns rapidly, which will aid coaches and clinicians in assessing changes in running technique considering changes in the study variables. Therefore, a review and proposal of the most representative events and variables of analysis in running was conducted to develop the KeepRunning test. Likewise, the minimal detectable change (MDC) in these variables was obtained using test-retest reliability to demonstrate the reproducibility and viability of the test, as well as the use of MDC as a threshold for future assessments. The test-retest consisted of 32 healthy volunteer athletes with a running training routine of at least 15 km per week repeating the test twice. In each test, clusters of markers were placed on the runners' body segments using elastic bands and the volunteers' movements were captured while running on a treadmill. In this study, reproducibility was defined by the intraclass correlation coefficient (ICC) and MDC, obtaining a mean value of ICC = 0.94 ± 0.05 for all variables and MDC = 2.73 ± 1.16° for the angular kinematic variables. The results obtained in the test-retest reveal that the reproducibility of the test was similar or better than that found in the literature. KeepRunning is a running analysis test that provides data from the involved body segments rapidly and easily interpretable. This data allows clinicians and coaches to objectively provide indications for runners to improve their running technique and avoid possible injury. The proposed test can be used in the future with inertial motion capture and other wearable technologies.

Correlations Between Hamstring Muscle Architecture, Maturation, and Anthropometric Measures in Academy Soccer Players

PURPOSE

Muscle architecture is associated with motor performance and muscle injury. While muscle architecture and knee-flexor eccentric strength change with growth, the influence of anthropometric measures on these properties is rarely considered. This study aimed to investigate the relationship between hamstring muscle architecture and knee-flexor eccentric strength with anthropometric measurements.

METHODS

Sixty male footballers (16.6 [1.05] y) from the U16, U17, and U19 teams of an elite soccer club were included in this study. Fascicle length, pennation angle, and muscle thickness of the biceps femoris long head (BFlh) and semimembranosus muscles were measured in both legs using ultrasound. Knee-flexor eccentric strength, height, body mass, leg length, femur length, and peak height velocity (PHV) were measured within 1 week of the ultrasound images. A stepwise regression and 1-way analysis of variance tests were used to evaluate the effects of age, maturity, and anthropometric measurements on muscle properties.

RESULTS

Variance within BFlh and semimembranosus muscle thickness (r < .61), semimembranosus pennation angle (r < .58), and knee-flexor eccentric strength (r = .50) were highly related to body mass. We observed no significant correlations between muscle architecture and age (P > .29). However, moderately greater BFlh muscle thickness was shown for the post-PHV compared with the PHV group (effect size ± 90% CI: 0.72 ± 0.49).

CONCLUSIONS

In conclusion, weak correlations between muscle architecture and anthropometric measurements suggest that other factors (ie, genetics, training regimen) influence muscle architecture. The moderate effect of maturity on BFlh muscle thickness strongly suggests post-PHV hypertrophy of the BFlh muscle. Our results confirmed previous findings that eccentric knee-flexor strength is influenced by body mass.

Effects of creatine monohydrate timing on resistance training adaptations and body composition after 8 weeks in male and female collegiate athletes

Background

Limited research is available on the potential impact of creatine monohydrate administration before or after workouts among athletes. This study aimed to investigate the effects of pre- vs. post-exercise creatine monohydrate supplementation on resistance training adaptations and body composition.

Methods

In a randomized, double-blind, placebo-controlled, parallel design, 34 healthy resistance-trained male and female athletes were randomly assigned and matched according to fat free mass to consume a placebo, or 5-g dose of creatine monohydrate within 1 h before training, or within 1 h after training for 8 weeks, while completing a weekly resistance training program. Participants co-ingested 25-gram doses of both whey protein isolate and maltodextrin along with each assigned supplement dose. Body composition, muscular strength, and endurance, along with isometric mid-thigh pull were assessed before and after the 8-week supplementation period. A 3 × 2 mixed factorial (group x time) ANOVA with repeated measures on time were used to evaluate differences.

Results

All groups experienced similar and statistically significant increases in fat free mass (+1.34 ± 3.48 kg, p = 0.04), upper (+2.21 ± 5.69 kg, p = 0.04) and lower body strength (+7.32 ± 10.01 kg, p &lt; 0.001), and decreases in body mass (−1.09 ± 2.71 kg, p = 0.03), fat mass (−2.64 ± 4.16 kg, p = 0.001), and percent body fat (−2.85 ± 4.39 kg, p &lt; 0.001).

Conclusions

The timing of creatine monohydrate did not exert any additional influence over the measured outcomes.

Replacing statistical significance and non-significance with better approaches to sampling uncertainty

A sample provides only an approximate estimate of the magnitude of an effect, owing to sampling uncertainty. The following methods address the issue of sampling uncertainty when researchers make a claim about effect magnitude: informal assessment of the range of magnitudes represented by the confidence interval; testing of hypotheses of substantial (meaningful) and non-substantial magnitudes; assessment of the probabilities of substantial and trivial (inconsequential) magnitudes with Bayesian methods based on non-informative or informative priors; and testing of the nil or zero hypothesis. Assessment of the confidence interval, testing of substantial and non-substantial hypotheses, and assessment of Bayesian probabilities with a non-informative prior are subject to differing interpretations but are all effectively equivalent and can reasonably define and provide necessary and sufficient evidence for substantial and trivial effects. Informative priors in Bayesian assessments are problematic, because they are hard to quantify and can bias the outcome. Rejection of the nil hypothesis (presented as statistical significance), and failure to reject the nil hypothesis (presented as statistical non-significance), provide neither necessary nor sufficient evidence for substantial and trivial effects. To properly account for sampling uncertainty in effect magnitudes, researchers should therefore replace rather than supplement the nil-hypothesis test with one or more of the other three equivalent methods. Surprisal values, second-generation p values, and the hypothesis comparisons of evidential statistics are three other recent approaches to sampling uncertainty that are not recommended. Important issues beyond sampling uncertainty include representativeness of sampling, accuracy of the statistical model, individual differences, individual responses, and rewards of benefit and costs of harm of clinically or practically important interventions and side effects.

The Effect of Mat Pilates Training Combined With Aerobic Exercise Versus Mat Pilates Training Alone on Blood Pressure in Women With Hypertension: A Randomized Controlled Trial

OBJECTIVE

The purpose of this study was to determine the effects of mat Pilates (MP) versus MP plus aerobic exercise (AE) compared with the effects of no intervention on ambulatory blood pressure (BP) in women with hypertension.

METHODS

This 3-arm, parallel-group randomized clinical trial assessed 60 women who had hypertension and were 30 to 59 years old. The intervention lasted 16 weeks, and the participants were allocated into 3 groups: MP only (MP group), MP with alternating bouts of AE on a treadmill (MP + AE group), and control group (CG) with no exercises. Primary outcomes were the effects of the interventions on ambulatory BP assessed in the 24-hour, awake, and asleep periods of analysis.

RESULTS

A 2-way analysis of variance did not reveal statistically significant differences in between-group comparisons in the 24-hour period of analysis for systolic BP (CG vs MP = 3.3 [95% CI = -7.1 to 13.8]; MP vs MP + AE = 0.7 [95% CI = -4 to 5.4]; CG vs MP + AE = 4.0 [95% CI = -5.2 to 13.4]), diastolic BP (CG vs MP = 2.2 [95% CI = -5.6 to 10.0]; MP vs MP + AE = 1.1 [95% CI = -4.3 to 6.5]; CG vs MP + AE = 3.3 [95% CI = -3.8 to 10.4]), and heart rate (CG vs MP = 3.4 [95% CI = -2 to 8.8]; MP vs MP + AE = 2.0 [95% CI = -3.4 to 7.5]; CG vs MP + AE = 5.4 [95% CI = -0.8 to 11.8]). The awake and asleep periods of analyses also showed similar behavior and did not reveal statistically significant between-group differences. Furthermore, in the responsiveness analysis based on the minimal clinically important difference, no differences were observed between groups.

CONCLUSION

The magnitudes of the decrease in systolic BP during the 24-hour period of analysis were -3 and -5.48 mm Hg for the MP and MP + AE groups, without differences for responsiveness between groups. The results suggest that MP supplemented with AE or not may be an alternative adjuvant treatment for women who have hypertension and are using antihypertension medication.

IMPACT

Sixteen weeks of MP training reduced ambulatory BP in women who had hypertension. The MP + AE group displayed a BP reduction similar to that of the MP group. A reduction in ambulatory BP can decrease the risk of cardiovascular disease.

The dose-response relationship between training-load measures and aerobic fitness in elite academy soccer players

The aim of the current study is to examine the dose-response relationships between training load (TL) measures and the consequent changes in aerobic fitness. Data were collected over the 6-week pre-season period in elite youth soccer players. Participants completed a lactate threshold test to identify changes in treadmill speed at 2 mmol · l^-1 (S2) and 4 mmol · l^-1 (S4). Internal TL was quantified with the following training impulse (TRIMP) methods: Banister TRIMP, Edwards TRIMP, Lucia TRIMP, individual TRIMP (iTRIMP) and rate of perceived exertion was also collected. External TL measures were total distance, PlayerLoad, high speed running (14.4-19.8 km · h^-1), very high-speed running (19.8-25.2 km · h^-1) and maximal sprint distance (>25.2 km · h^-1). Individual high-speed distance was derived from each participants treadmill speed at S4. Different Bayesian regression models were run with different likelihood functions. The best-fitting models with both the lowest out-of-sample prediction error and the highest variance explained (R²) were used. iTRIMP had the strongest relationships with changes in S2 (r = 0.93, R² = 0.90) and S4 (r = 0.88, R² = 0.82). Explained variance ranged from 10%-69% and 11%-38% for all other internal TL measures and external measures, respectively. In summary, the iTRIMP method demonstrates a dose-response relationship with changes in aerobic fitness in elite youth soccer players.

A Systems Analysis Critique of Sport-Science Research

PURPOSE

The broad aim of sport-science research is to enhance the performance of coaches and athletes. Despite decades of such research, it is well documented that sport-science research lacks empirical evidence, and critics have questioned its scientific methods. Moreover, many have pointed to a research-practice gap, whereby the work undertaken by researchers is not readily applied by practitioners. The aim of this study was to use a systems thinking analysis method, causal loop diagrams, to understand the systemic issues that interact to influence the quality of sport-science research.

METHODS

A group model-building process was utilized to develop the causal loop diagram based on data obtained from relevant peer-reviewed literature and subject-matter experts.

RESULTS

The findings demonstrate the panoply of systemic influences associated with sport-science research, including the existence of silos, a focus on quantitative research, archaic practices, and an academic system that is incongruous with what it actually purports to achieve.

CONCLUSIONS

The emergent outcome of the interacting components is the creation of an underperforming sport-science research system, as indicated by a lack of ecological validity, translation to practice, and, ultimately, a research-practice gap.

Timing of Creatine Supplementation around Exercise: A Real Concern?

Creatine has been considered an effective ergogenic aid for several decades; it can help athletes engaged in a variety of sports and obtain performance gains. Creatine supplementation increases muscle creatine stores; several factors have been identified that may modify the intramuscular increase and subsequent performance benefits, including baseline muscle Cr content, type II muscle fibre content and size, habitual dietary intake of Cr, aging, and exercise. Timing of creatine supplementation in relation to exercise has recently been proposed as an important consideration to optimise muscle loading and performance gains, although current consensus is lacking regarding the ideal ingestion time. Research has shifted towards comparing creatine supplementation strategies pre-, during-, or post-exercise. Emerging evidence suggests greater benefits when creatine is consumed after exercise compared to pre-exercise, although methodological limitations currently preclude solid conclusions. Furthermore, physiological and mechanistic data are lacking, in regard to claims that the timing of creatine supplementation around exercise moderates gains in muscle creatine and exercise performance. This review discusses novel scientific evidence on the timing of creatine intake, the possible mechanisms that may be involved, and whether the timing of creatine supplementation around exercise is truly a real concern.

Directions for Exercise Treatment Response Heterogeneity and Individual Response Research

Treatment response heterogeneity and individual responses following exercise training are topics of interest for personalized medicine. Proposed methods to determine the contribution of exercise to the magnitude of treatment response heterogeneity and categorizing participants have expanded and evolved. Setting clear research objectives and having a comprehensive understanding of the strengths and weaknesses of the available methods are vital to ensure the correct study design and analytical approach are used. Doing so will ensure contributions to the field are conducted as rigorously as possible. Nonetheless, concerns have emerged regarding the ability to truly isolate the impact of exercise training, and the nature of individual responses in relation to mean group changes. The purpose of this review is threefold. First, the strengths and limitations associated with current methods for quantifying the contribution of exercise to observed treatment response heterogeneity will be discussed. Second, current methods used to categorize participants based on their response to exercise will be outlined, as well as proposed mechanisms for factors that contribute to response variation. Finally, this review will provide an overview of some current issues at the forefront of individual response research.

Effect of high-velocity and traditional resistance exercise on serum antioxidants and inflammation biomarkers in older women: A randomized crossover trial

INTRODUCTION

Contrary to the production of oxidant and inflammation biomarkers, antioxidant enzymes decrease with age and the adaptive response to oxidative stress is reduced. The effects of the different resistance exercise protocols to attenuate these conditions are largely unexplored.

OBJECTIVE

We compared the acute effects of traditional resistance exercise (TRE) and high velocity RE (HVRE) on inflammatory biomarkers and antioxidant activity in elderly women.

METHODS

Fourteen elderly women (67 ± 7 years) were randomly assigned to TRE or HVRE. Blood samples were collected to measure inflammatory markers and antioxidant enzymes at three time-points (pre-exercise, post-exercise, 30 min post-exercise).

RESULTS

TRE and HVRE induced acute reductions over time on tumor necrosis factor-alpha, interleukin-6 (IL-6), soluble receptor of IL-6 (sIL-6R), as well in catalase, glutathione and superoxide dismutase antioxidant levels without differences between groups. In addition, TRE and HVRE groups displayed acute increments in interleukin-10 (IL-10) and IL-10/IL-6 ratio over time, while the HVRE protocol displayed higher IL-10 values post-exercise and 30 min post-exercise time points as compared with TRE. Based on magnitude based inference, subjects from the HVRE group demonstrated superior responsiveness for IL-6 and IL-10 as compared with the TRE group. All participants from the HVRE group displayed a minimal clinical important difference on IL-10 levels as compared to only two persons from the TRE group.

CONCLUSION

Both RE protocols were capable of positive changes in inflammatory and antioxidant status in elderly subjects, but HVRE demonstrated a superior response on IL-10. The HVRE may be incorporated to exercise recommendation in this population.

The Effects of Caffeine Mouth Rinsing on Exercise Performance: A Systematic Review

Caffeine ingestion can improve performance across a variety of exercise modalities but can also elicit negative side effects in some individuals. Thus, there is a growing interest in the use of caffeine mouth rinse solutions to improve sport and exercise performance while minimizing caffeine’s potentially adverse effects. Mouth rinse protocols involve swilling a solution within the oral cavity for a short time (e.g., 5–10 s) before expectorating it to avoid systemic absorption. This is believed to improve performance via activation of taste receptors and stimulation of the central nervous system. Although reviews of the literature indicate that carbohydrate mouth rinsing can improve exercise performance in some situations, there has been no attempt to systematically review the available literature on caffeine mouth rinsing and its effects on exercise performance. To fill this gap, a systematic literature search of three databases (PubMed, SPORTDiscus, and Web of Science) was conducted by two independent reviewers. The search resulted in 11 randomized crossover studies that were appraised and reviewed. Three studies found significant positive effects of caffeine mouth rinsing on exercise performance, whereas the remaining eight found no improvements or only suggestive benefits. The mixed results may be due to heterogeneity in the methods across studies, interindividual differences in bitter tasting, and differences in the concentrations of caffeine solutions. Future studies should evaluate how manipulating the concentration of caffeine solutions, habitual caffeine intake, and genetic modifiers of bitter taste influence the efficacy of caffeine mouth rinsing as an ergogenic strategy.

The effects of strength and conditioning interventions on golf performance: A systematic review

There is a growing body of literature on strength and conditioning (S&C) interventions for golfers of various skill levels. The aim of this systematic review was to evaluate the effects of S&C interventions on measures of golf performance (clubhead speed, ball speed, distance, etc.). Three databases (PubMed, SPORTDiscus, Web of Science) were searched and twenty-five studies identified that evaluated the effects of a S&C intervention on at least one golf performance measure compared to a control or comparison group. Most studies used combinations of strength training, plyometrics, stretching or core exercise, with many finding a benefit. Though it varied across studies and outcomes, average increases in clubhead speed, ball speed and distance measures were 4-6.4% when significant findings were synthesized. Four studies also found significant changes to golf swing kinematics, while three others found positive effects on measures of accuracy or consistency. Future research should compare different S&C interventions, explore the role of training status, skill level and intervention duration on the effects of S&C interventions, and report individual responses in addition to group data. Further, research should continue to evaluate effects on swing kinematics, accuracy and direct golf performance measures (e.g., handicap index).

Systematic review of the use of "magnitude-based inference" in sports science and medicine

Magnitude-based inference (MBI) is a controversial statistical method that has been used in hundreds of papers in sports science despite criticism from statisticians. To better understand how this method has been applied in practice, we systematically reviewed 232 papers that used MBI. We extracted data on study design, sample size, and choice of MBI settings and parameters. Median sample size was 10 per group (interquartile range, IQR: 8-15) for multi-group studies and 14 (IQR: 10-24) for single-group studies; few studies reported a priori sample size calculations (15%). Authors predominantly applied MBI's default settings and chose "mechanistic/non-clinical" rather than "clinical" MBI even when testing clinical interventions (only 16 studies out of 232 used clinical MBI). Using these data, we can estimate the Type I error rates for the typical MBI study. Authors frequently made dichotomous claims about effects based on the MBI criterion of a "likely" effect and sometimes based on the MBI criterion of a "possible" effect. When the sample size is n = 8 to 15 per group, these inferences have Type I error rates of 12%-22% and 22%-45%, respectively. High Type I error rates were compounded by multiple testing: Authors reported results from a median of 30 tests related to outcomes; and few studies specified a primary outcome (14%). We conclude that MBI has promoted small studies, promulgated a "black box" approach to statistics, and led to numerous papers where the conclusions are not supported by the data. Amidst debates over the role of p-values and significance testing in science, MBI also provides an important natural experiment: we find no evidence that moving researchers away from p-values or null hypothesis significance testing makes them less prone to dichotomization or over-interpretation of findings.

Pacing and Performance in the 6 World Marathon Majors

The main goal of this study was to analyse the pacing strategies displayed by the winners of the six World Marathon Majors in order to determine which race offers the greatest potential for future world record attempts. For data analysis, the total distance of the marathon was divided into eight sections of 5 km and a final section of 2.195 km, and time needed to complete each section was calculated in seconds. When we analyzed the mean winning time in the last 13 editions of each of the World Marathon Majors, we observed differences between New York and London (ES = 1.46, moderate effect, p = 0.0030), New York and Berlin (ES = 0.95, small effect, p = 0.0001), London and Boston (ES = 0.08, small effect, p = 0.0001), Boston and Berlin (ES = 0.10, small effect, p = 0.0001), Boston and Chicago (ES = 0.16, small effect, p = 0.0361), Berlin and Tokyo (ES = 0.20, small effect, p = 0.0034), Berlin and Chicago (ES = 0.27, small effect, p = 0.0162). This study shows that Berlin and London are likely candidates for future world record attempts, whilst such a performance is unlikely in New York or Boston.

Using differential ratings of perceived exertion to assess agreement between coach and player perceptions of soccer training intensity: An exploratory investigation

We aimed to assess the coach-player agreement of subjective soccer training loads via differential ratings of perceived exertion (dRPE). The coach initially underwent quantifiable familiarisation (blackness test) with the Borg CR100 scale. Data were collected from 16 semi-professional soccer players across seven consecutive training sessions. For the measurement of subjective training load, the coach and players provided dRPE (CR100) for legs (RPE-L), breathlessness (RPE-B) and technical exertion (RPE-T). Coach-prescribed dRPE were recorded prior to training, with coach observed and player reported dRPE collected post training. Statistical equivalence bounds for agreement between coach (prescribed and observed) and player reported dRPE scores were 4 arbitrary units on the CR100 and we used a probability outcome of likely (≥75%) to infer realistic equivalence. Following three familiarisation sessions, the coach improved their blackness test score from 39% to 83%. Coach observed and player reported RPE-T scores were likely equivalent, with all other comparisons not realistically equivalent. Since training prescription is coach-led, our data highlight the importance of accurate internal load measurement and feedback in soccer. The improved accuracy and precision of coach intensity estimation after three attempts at the blackness test suggests that this method could be worthwhile to researchers and practitioners employing dRPE.