Employing Google Trends to Assess Concussion Search Popularity, Seasonality, and Association With High School Sports Participation

Neurology. 2022 Jan 4;98(12 Suppl 2):S9-S10. doi: 10.1212/01.wnl.0000801820.38637.38.

Abstract

Objective: This investigation employs 2004-2020 Google Trends data to correlate concussion search popularity and high school sports participation while assessing search seasonality and factors associated with search escalation.

Background: Events linked to concussions, from news to entertainment, altered public perception of concussions and discouraged high-risk activities. Such sentiment likely promoted high school football's 2004-2018 8% decrease in participation, despite simultaneous 8% increase for sports overall. Google Trends (GT) analysis calculates distinct subjects' popularity scores (0-100) by normalizing Google search frequency with other subjects in the same topic. With 35% of patients researching their conditions using Google, tracking concussion popularity scores allows robust analysis of event-driven fluctuation, seasonal variation, and sports impact.

Design/methods: Worldwide GT analysis was conducted for the disease topic "Concussion" with the categorical filter "Health" from 2004 to 2020. This period's monthly popularity scores with annual averages and annual high school sports participation data were subsequently obtained from Google Trends and the National Federation of State High School Associations, respectively.

Results: Concurrence of significant score changes with CTE studies' publication, release of 2015 film Concussion, and the COVID-19 pandemic validates GT's measurement of public interest. From 2006 to 2016, Pearson's correlation coefficients demonstrated strong negative correlation between GT popularity scores for "Concussion" with total number of high school football participants overall (R2 = 0.8553) and participants per school (R2 = 0.9533). Confirming football-related seasonality, one-way ANOVA regression analysis concluded 2004-2020's mean change in month-specific popularity score is not the same (p = 3.193E-08), and months during football season had statistically significant variability (September: p = 4.389E-05) with elevated average z-scores.

Conclusions: Coupled with ANOVA regression and normalized analyses, strong negative correlations between concussion popularity score and high school football participation produce quantitative measures of a long-suspected relationship. Tracking concussion popularity and seasonality with Google Trends provides insight into how, when, and why patients are educating themselves online.

MeSH terms

  • Athletic Injuries* / epidemiology
  • Brain Concussion* / epidemiology
  • COVID-19
  • Humans
  • Pandemics
  • Schools
  • Search Engine*
  • Seasons
  • Youth Sports