Modeling count and binary data collected in hierarchical designs have increased the use of Generalized Linear Mixed Models (GLMMs) in medicine. This article presents a systematic review of the application and quality of results and information reported from GLMMs in the field of clinical medicine. A search using the Web of Science database was performed for published original articles in medical journals from 2000 to 2012. The search strategy included the topic “generalized linear mixed models”,“hierarchical generalized linear models”, “multilevel generalized linear model” and as a research domain we refined by science technology. Papers reporting methodological considerations without application, and those that were not involved in clinical medicine or written in English were excluded. A total of 443 articles were detected, with an increase over time in the number of articles. In total, 108 articles fit the inclusion criteria. Of these, 54.6% were declared to be longitudinal studies, whereas 58.3% and 26.9% were defined as repeated measurements and multilevel design, respectively. Twenty-two articles belonged to environmental and occupational public health, 10 articles to clinical neurology, 8 to oncology, and 7 to infectious diseases and pediatrics. The distribution of the response variable was reported in 88% of the articles, predominantly Binomial (n = 64) or Poisson (n = 22). Most of the useful information about GLMMs was not reported in most cases. Variance estimates of random effects were described in only 8 articles (9.2%). The model validation, the method of covariate selection and the method of goodness of fit were only reported in 8.0%, 36.8% and 14.9% of the articles, respectively. During recent years, the use of GLMMs in medical literature has increased to take into account the correlation of data when modeling qualitative data or counts. According to the current recommendations, the quality of reporting has room for improvement regarding the characteristics of the analysis, estimation method, validation, and selection of the model.

Different active and passive post-exercise recovery techniques such as massage, foam rolling, stretching or ice baths among others, are used by elite athletes to promote effective physiological, ph...

Objective We assessed past-season, pre-season and in-season risk factors to investigate their association with an in-season groin problem in male amateur football players. Methods Past-season groin-pain information and pre-season short-lever and long-lever adductor squeeze strength were obtained at baseline, together with anthropometrics (weight, lower limb lever length) and player age. In-season hip-related and groin-related sporting function was monitored every 4 weeks using the Sports and Recreation (Sport) subscale from the Hip And Groin Outcome Score questionnaire (HAGOS (Sport)). Groin problems, including time-loss groin injuries and groin pain irrespective of time loss, were collected over a 39-week competitive in-season. We estimated relative risk (RR), and 95% credibility interval (ICr) from logistic regressions fitted in a Bayesian framework. Results Players (n=245) suffering from groin pain during the past-season had 2.4 times higher risk of experiencing a groin problem in the new season (2.40 RR; 95% ICr 1.5 to 3.7). This risk was reduced by 35% (0.65 RR; 95% ICr 0.42 to 0.99) per unit (N·m/kg) increase in the long-lever adductor squeeze test. Player age, short-lever squeeze test and the HAGOS (Sport) scores were not associated with the risk of a groin problem. Conclusions Past-season groin pain increased the risk of a groin problem in the new in-season. This risk was reduced by higher pre-season long-lever adductor squeeze strength. Past-season groin-pain information and long-lever adductor squeeze strength can be quickly obtained during pre-season to identify players with an elevated risk of in-season groin problems. This may be key to reduce these problems in the new season.

The aims of this study were to identify variables which may potentially influence player performance, and to implement a statistical model to study their relative contribution in order to explain two outcomes: points and win score. We used all the possible variables affecting player performance creating a comprehensive database from two sources of statistical information about the NBA 2007 regular season: www.basketball-reference.com and www.nbastuffer.com. The data employed for the analysis were composed of 2187 cases (27 players * 81 games), having followed a filtering process. We dealt with a balanced study design with repeated measurements given that each player was observed the same number of games, and therefore the player was considered as a random effect. We carried out mixed models to quantify the variability in points and win score among players. Minutes played, the usage percentage and the difference of quality between teams were the main factors for variations in points made and win score. The interaction between player position and age was important in win score. We encourage managers and coaches of sports teams to choose appropriate methods according to their aims. Future research should take into consideration the use of models with random effects on players’ characteristics.