Alan M. Nevill

Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine.

Minimal measurement error (reliability) during the collection of interval- and ratio-type data is critically important to sports medicine research. The main components of measurement error are systematic bias (e.g. general learning or fatigue effects on the tests) and random error due to biological or mechanical variation. Both error components should be meaningfully quantified for the sports physician to relate the described error to judgements regarding ‘analytical goals’ (the requirements of the measurement tool for effective practical use) rather than the statistical significance of any reliability indicators.Methods based on correlation coefficients and regression provide an indication of ‘relative reliability’. Since these methods are highly influenced by the range of measured values, researchers should be cautious in: (i) concluding acceptable relative reliability even if a correlation is above 0.9; (ii) extrapolating the results of a test-retest correlation to a new sample of individuals involved in an experiment; and (iii) comparing test-retest correlations between different reliability studies.Methods used to describe ‘absolute reliability’ include the standard error of measurements (SEM), coefficient of variation (CV) and limits of agreement (LOA). These statistics are more appropriate for comparing reliability between different measurement tools in different studies. They can be used in multiple retest studies from ANOVA procedures, help predict the magnitude of a ‘real’ change in individual athletes and be employed to estimate statistical power for a repeated-measures experiment.These methods vary considerably in the way they are calculated and their use also assumes the presence (CV) or absence (SEM) of heteroscedasticity. Most methods of calculating SEM and CV represent approximately 68% of the error that is actually present in the repeated measurements for the ‘average’ individual in the sample. LOA represent the test-retest differences for 95% of a population. The associated Bland-Altman plot shows the measurement error schematically and helps to identify the presence of heteroscedasticity. If there is evidence of heteroscedasticity or non-normality, one should logarithmically transform the data and quote the bias and random error as ratios. This allows simple comparisons of reliability across different measurement tools.It is recommended that sports clinicians and researchers should cite and interpret a number of statistical methods for assessing reliability. We encourage the inclusion of the LOA method, especially the exploration of heteroscedasticity that is inherent in this analysis. We also stress the importance of relating the results of any reliability statistic to ‘analytical goals’ in sports medicine.

A multidisciplinary approach to talent identification in soccer

The requirements for soccer play are multifactorial and distinguishing characteristics of elite players can be investigated using multivariate analysis. The aim of the present study was to apply a comprehensive test battery to young players with a view to distinguishing between elite and sub-elite groups on the basis of performance on test items. Thirty-one (16 elite, 15 sub-elite) young players matched for chronological age (15-16 years) and body size were studied. Test items included anthropometric ( n = 15), physiological ( n = 8), psychological ( n = 3) and soccer-specific skills ( n = 2) tests. Variables were split into separate groups according to somatotype, body composition, body size, speed, endurance, performance measures, technical skill, anticipation, anxiety and task and ego orientation for purposes of univariate and multivariate analysis of variance and stepwise discriminant function analysis.The most discriminating of the measures were agility, sprint time, ego orientation and anticipation skill. The elite players were also significantly leaner, possessed more aerobic power (9.0 - 1.7 vs 55.5 - 3.8 ml·kg -1 ·min -1 ) and were more tolerant of fatigue ( P ≪ 0.05). They were also better at dribbling the ball, but not shooting. We conclude that the test battery used may be useful in establishing baseline reference data for young players being selected onto specialized development programmes.

Scaling physiological measurements for individuals of different body size.

SummaryThis paper examines how selected physiological performance variables, such as maximal oxygen uptake, strength and power, might best be scaled for subject differences in body size. The apparent dilemma between using either ratio standards or a linear adjustment method to scale was investigated by considering how maximal oxygen uptake (1·min−1), peak and mean power output (W) might best be adjusted for differences in body mass (kg). A curvilinear power function model was shown to be theoretically, physiologically and empirically superior to the linear models. Based on the fitted power functions, the best method of scaling maximum oxygen uptake, peak and mean power output, required these variables to be divided by body mass, recorded in the units kg2/3. Hence, the power function ratio standards (ml·kg−2/3·min−1) and (W·kg−2/3) were best able to describe a wide range of subjects in terms of their physiological capacity, i.e. their ability to utilise oxygen or record power maximally, independent of body size. The simple ratio standards (ml·kg−1·min−1) and (W·kg−1) were found to best describe the same subjects according to their performance capacities or ability to run which are highly dependent on body size. The appropriate model to explain the experimental design effects on such ratio standards was shown to be log-normal rather than normal. Simply by taking logarithms of the power function ratio standard, identical solutions for the design effects are obtained using either ANOVA or, by taking the unscaled physiological variable as the dependent variable and the body size variable as the covariate, ANCOVA methods.

Home advantage in sport: an overview of studies on the advantage of playing at home.

AbstractThis review identifies the most likely causes of home advantage. The results of previous studies have identified 4 factors thought to be responsible for the home advantage. These can be categorised under the general headings of crowd, learning, travel and rule factors. From the accumulated evidence, rule factors were found to play only a minor role (in a limited number of sports) in contributing to home advantage. Studies investigating the effect of learning factors found that little benefit was to be gained from being familiar with the local conditions when playing at home. There was evidence to suggest that travel factors were responsible for part of the home advantage, provided the journey involved crossing a number of time zones. However, since high levels of home advantage are observed within countries where travel distances are not great, travel factors were not thought to be a major cause of home advantage. The evidence from studies investigating crowd factors appeared to provide the most dominant causes of home advantage.A number of studies provide strong evidence that home advantage increases with crowd size, until the crowd reaches a certain size or consistency (a more balanced number of home and away supporters), after which a peak in home advantage is observed. Two possible mechanisms were proposed to explain these observations: either (i) the crowd is able to raise the performance of the home competitors relative to the away competitors; or (ii) the crowd is able to influence the officials to subconsciously favour the home team. The literature supports the latter to be the most important and dominant explanation. Clearly, it only takes 2 or 3 crucial decisions to go against the away team or in favour of the home team to give the side playing at home the ‘edge’.

Selected issues in the design and analysis of sport performance research

The aim of this review is to discuss some issues in the design and statistical analysis of sport performance research, rather than to supply an authoritative ‘cookbook’ of methods. In general, we try to communicate some possible solutions to the conundrum of how to maintain both internal and external validity, as well as optimize statistical power, in applied sport performance research. We start by arguing that some sport performance research has been overly concerned with physiological predictors of performance, at the expense of not providing a valid and reliable description of the exact nature of the task in question. We show how the influence of certain factors on competitive performances can be described using linear or logistic regression. We discuss the choice of analysis for factorial repeated-measures designs, which is complicated by the assumption of ‘sphericity’ in a univariate general linear model, and the relatively low statistical power of the multivariate approach when used with small samples. We consider a little-used and simpler technique known as ‘analysis of summary statistics’. In multi-group pre- and post-test designs, a useful technique can be to pair-match individuals on their performance scores in a counterbalanced fashion before the intervention or control has been introduced. Finally, we outline how confidence intervals can help in making statements about the probability of the population difference in performance exceeding the value designated as being worthwhile or not.

Hormonal factors in the development of differences in strength between boys and girls during adolescence: a longitudinal study

The development of elbow flexor (biceps) and knee extensor (quadriceps) strength has been followed in a mixed longitudinal study of 50 boys and 50 girls from the age of 8 to 17 years. Sex differences in strength emerged at the time of peak height velocity and were especially marked for the biceps. Data for individual children were aligned to the time of peak height velocity and associations between strength, height, weight and circulating testosterone were investigated using multi-level modelling. The results show that, for girls, quadriceps strength is proportional to height and weight while for boys there is an additional factor which can be fully attributed to increasing levels of testosterone. Testosterone is important in explaining differences in biceps strength between the sexes but an additional factor is also required. It is suggested that, in addition to a direct effect on muscle, testosterone could have a second indirect action on biceps strength by promoting growth in length of the humerus as part of the general development of the male upper limb girdle.

Association of physical inactivity with increased cardiovascular risk in patients with rheumatoid arthritis.

Objective Patients with rheumatoid arthritis (RA) are characterized by reduced physical activity and increased morbidity and mortality from cardiovascular disease (CVD). The aim of this study was to investigate associations between levels of physical activity and CVD risk profile in RA patients. Methods Levels of physical activity were assessed in 65 RA patients (43 females). Using the International Physical Activity Questionnaire, patients were allocated into three groups: active, moderately active and inactive. Anthropometric characteristics, RA activity/severity, multiple classical and novel CVD risk factors and 10-year CVD event probability were assessed and compared among the three groups. Results Significant differences were detected among groups in systolic blood pressure (P = 0.006), cholesterol (P < 0.001), low-density lipoprotein (P = 0.01), homeostasis model assessment (P = 0.001), type-1 plasminogen activator inhibitor antigen (P < 0.001), tissue-type plasminogen activator antigen (P = 0.019), homocysteine (P = 0.027), fibrinogen (P = 0.001), apolipoprotein B (P = 0.002) and von Willebrand Factor (P = 0.001), with a consistent deterioration from the physically active to the physically inactive group. Multivariate analysis of variance revealed that levels of physical activity were significantly associated with the differences in all of the above variables (P < 0.05) after adjustment for age, weight, sex, smoking status, as well as RA disease activity and severity. Conclusion This cross-sectional study suggests that physically inactive RA patients have significantly worse CVD risk profile compared with physically active patients. The possible beneficial impact of increased physical activity, including structured exercise, to the CVD risk of RA patients needs to be accurately assessed in prospective studies. Eur J Cardiovasc Prev Rehabil 16:188-194

Factors associated with home advantage in English and Scottish soccer matches

Using the results from the end-of-season (1992-93) league tables, overall home advantage was confirmed in the eight major divisions of the English and Scottish football leagues. The degree of home advantage was found to vary significantly across the divisions. Furthermore, these divisional differences in home advantage were found to be significantly associated with the mean attendance of each division. In an attempt to understand these findings, every occurrence of two influential events (either a sending-off or penalty scored) reported in a national Sunday newspaper was recorded. The overall frequency of both sendings-off and penalties scored favoured the home side, but again this was not constant across the divisions. In divisions with large crowds, the percentage of home sendings-off was relatively small (30%), compared with no difference (50%) in divisions with smaller crowds. Similarly, the percentage of penalties scored by home sides in divisions with the largest crowds was large ( > 70%), in contrast to little or no advantage in divisions with smaller crowds. Two possible explanations for these findings were proposed. Either larger crowds were able to provoke the away player into more reckless behaviour (real fouls), or influence the referee into believing that the away player had committed more fouls (perceived fouls).

Physical activity, risk of death and recurrence in breast cancer survivors: A systematic review and meta-analysis of epidemiological studies

Abstract Strong evidence exists supporting the effect of lack of physical activity on the risk of developing breast cancer. However, studies examining the effects of physical activity on breast cancer outcomes, including survival and prognosis have been inconclusive. Therefore, the aim of the current study was to provide a systematic review and meta-analysis of studies investigating the association between physical activity and breast cancer recurrence and death. Methods. PubMed, EMBASE, and CENTRAL databases were searched up to 18 October 2014. Reference lists of retrieved articles and relevant previous reviews were also searched. Observational studies that reported risk estimates for all-cause and/or breast cancer-related death and/or breast cancer recurrences by levels of physical activity, were included in the review. Random effects models were used to calculate pooled hazard ratios (HR) and 95% confidence intervals (CI) and to incorporate variation between studies. The Newcastle-Ottawa scale was used to critically appraise the risk of bias across studies. Results. Twenty-two prospective cohort studies were eligible in this meta-analysis. During average follow-up periods ranging from 4.3 to 12.7 years there were 123 574 participants, 6898 all-cause deaths and 5462 breast cancer outcomes (i.e. breast cancer-related deaths or recurrences). The average Newcastle-Ottawa score was six stars (range 4–8). Compared to those who reported low/no lifetime recreational pre-diagnosis physical activity, participants who reported high lifetime recreational pre-diagnosis physical activity levels had a significantly lower risk of all-cause (HR = 0.82, 95% CI 0.70–0.96, p < 0.05) and breast cancer-related death (HR = 0.73, 95% CI 0.54–0.98, p < 0.05). Significant risk reductions for all-cause and breast cancer-related death was also demonstrated for more recent pre-diagnosis recreational physical activity (HR = 0.73, 95% CI 0.65–0.82, p < 0.001; and HR = 0.84, 95% CI 0.73–0.97, p < 0.05, respectively), post-diagnosis physical activity (HR = 0.52, 95% CI 0.43–0.64, p < 0.01; and HR = 0.59, 95% CI 0.45–0.78, p < 0.05, respectively) and meeting recommended physical activity guidelines (i.e. ≥ 8 MET-h/wk) post-diagnosis (HR = 0.54, 95% CI 0.38–0.76, p < 0.01; and HR = 0.67, 95% CI 0.50–0.90, p < 0.01, respectively). However, there was evidence of heterogeneity across lifetime recreational pre- and post-diagnosis physical activity analyses. Both pre-diagnosis (lifetime and more recent combined) and post-diagnosis physical activity were also associated with reduced risk of breast cancer events (breast cancer progression, new primaries and recurrence combined) (HR = 0.72 95% CI 0.56–0.91, p < 0.01; and HR = 0.79, 95% CI 0.63–0.98, p < 0.05, respectively). Conclusion. There is an inverse relationship between physical activity and all-cause, breast cancer-related death and breast cancer events. The current meta-analysis supports the notion that appropriate physical activity may be an important intervention for reducing death and breast cancer events among breast cancer survivors.

Greek adolescents, fitness, fatness, fat intake, activity, and coronary heart disease risk

Background: Percutaneous central venous lines (long lines) are commonly used in neonatal practice. The position of these lines is important, because incorrect placement may be associated with complications. Aims: To determine whether the addition of radio-opaque contrast material improves the localisation of long line tips over plain radiography. Methods: Radiographs taken to identify long line position were identified in two periods; 106 radiographs without contrast taken between October 1999 and August 2000, and 96 radiographs with contrast between September 2001 and July 2002. Two observers independently reviewed each radiograph to identify the position of the line tip. The formal radiology report was recorded as a third observer. Results: The use of contrast increased the proportion of radiographs in which all observers reported they could see the long line tip (53 (55%) v 41 (39%)). It also increased the proportion where they agreed on anatomical position (57 (59%) v 39 (37%)) and there was a higher kappa coefficient for agreement (0.56 v 0.33). Conclusions: The use of contrast while taking radiographs for the localisation of long line position improves the likelihood that an observer can see a long line tip and reduces inter-observer variability. Even using contrast, precise localisation of a long line tip can be difficult.