Search

Background: There is growing awareness and clinical interest in athletes with affective symptoms after sport-related concussion (SRC), as these symptoms may contribute to overall symptoms and represent a modifiable risk factor of longer recovery. However, evidence of their effects on the entire return-to-play (RTP) trajectory, particularly among women and men, is limited. Purpose/Hypothesis: To examine the relationship between affective symptom reporting and RTP progression after SRC among a cohort of Division 1 student-athletes. We hypothesized that those endorsing affective symptoms, specifically nervous-anxious symptoms, spend more time in RTP progression and recovery. Study Design: Cohort study; Level of evidence, 3. Methods: Using SRC data from the Ivy League–Big Ten Epidemiology of Concussion Study among varsity athletes through February 2020, we identified the 4 affective symptoms from the Sport Concussion Assessment Tool symptom inventory. We modeled the relationship between a 4-category affective symptom variable and time to symptom resolution, RTP, and RTP progression, adjusting for nonaffective symptom prevalence and concussion history. Cox regressions were used to estimate hazard ratios for time to event outcomes, and linear regressions were used to evaluate mean differences for continuous outcomes. Results: Among 2077 student-athletes (men, 63.5%) with SRC symptoms, affective symptom prevalence was 47.6% and 44.3% in women and men, respectively, and nervous-anxious prevalence was 24.2% and 22.5%, respectively. When comparing women with and without co-occurring affective symptoms, rates of symptom resolution and RTP were significantly lower in those with affective symptoms, and women with nervous-anxious symptoms spent significantly longer in RTP progression. When comparing men with and without co-occurring affective symptoms, rates of symptom resolution and RTP were significantly lower in those with co-occurring affective symptoms, and affective symptoms were not associated with time in RTP progression. Conclusion: Student-athletes with affective symptoms and nervous-anxious symptoms exhibited delayed clinical recovery and RTP timelines, particularly for time in RTP. Symptom prevalence and concussion history contributed to this; however, unmeasured confounding remains, as indicated by the poor model fit. This study motivates future work to explore affective symptoms and RTP timelines, considering anxiety and risk/protective factors over time.

This book has presented principles, methodologies, and examples for integrating predictive analytics and data science to many current applications. In the implementation of each of these applications, there are ethical aspects that relate to the question of whether or not to proceed and how to do it. We should integrate these ethical aspects into our decision-making operations for reasons that are both practical and theoretical reasons. In a discussion of the ethical aspects of any subject, it is helpful to begin with the presentation of a concrete situation from the real world. The next step is to work backward to derive a list of best practices with reference to basic principles that connect to one or more worldviews for help in determining the right things to do ( Sandel, 2010 ). We will review briefly the secular systems of ethics that serve as a relevant background to ethical aspects of decisions enabled by data science studies of today. The three secular academic perspectives for making ethical systems and decisions are (1) the normative, (2) the situational, and (3) the existential. This chapter will join a needed discussion on ethics in the new discipline of data analytics ( Fung, 2015 ).

Predictive analytics with Big Data in education will improve educational programs for students and fund-raising campaigns for donors ( Siegel, 2013 ). Research in both educational data mining (EDM) and data analytics (LA) continues to increase ( Siemens, 2013 ; Baker and Siemens, 2014 ). The key elements of recruitment, learning, and retention can be tracked and increased over time in both cases for administrators and donor development. A few early examples of Big Data programs in these key areas of formal education will be reviewed. The paradigms of educational psychology can inform the construction of systems that maximize “engagement.” Contemporary educational technology enables the personalization of education whether face to face or at a distance. More responsive evaluation systems will allow the school to demonstrate the effectiveness of its services. There is more opportunity for education as a sustainable enterprise with the implementation of Big Data, both locally and globally.