A series of CTE-related questions were included in the Porter Novelli (PN) Public Service’s SummerStyles public opinion survey that was fielded in June and July of 2017. All of PN’s consumer surveys utilize GfK’s KnowledgePanel^® (32), which is a type of multi-purpose panel (33). Panels are composed of individuals who have agreed to participate in periodic surveys about consumer, marketing, lifestyle, and health issues. These types of panels are generally thought to convey certain advantages in terms of conducting surveys, including relatively high response rates, lower attrition rates in longitudinal research, and easy access to respondent and household information (i.e. it is collected in the beginning and does not have to be collected for each survey) (34). Panel members for each specific survey are randomly recruited using probability-based sampling by home address to reach respondents regardless of their land-line phone or Internet access. If needed, households are provided with a laptop computer and access to the Internet to complete the survey. The panel is continuously replenished and maintains approximately 55,000 panelists. Porter Novelli has been conducting these surveys since 1995. The response rates for the SummerStyles panels have ranged between 66% and 74% for the last five iterations. While SummerStyles’ respondents should rightly be considered a convenience sample, researchers within and outside of the CDC were able to support the reliability and validity of the panel data by comparing survey responses to a separate nationally representative dataset (35,36).

The 2017 SummerStyles survey was sent to a random sample of 5586 panelists ages 18 or older who had participated in the spring iteration of the survey as well. Email reminders were sent to non-responders on the 3^rd, 15^th, 21^st and 23^rd day of the field period. The survey contained dozens of questions covering a variety of consumer and general health topics – ranging from general health questions to questions on tobacco use and air quality. Median survey completion time was approximately 26 minutes. Respondents were not required to answer any of the questions and had the opportunity to exit the survey at any time. Respondents received 5000 cash-equivalent points for their participation (approximately $5). Respondents who did not answer at least half of the questions were removed from the data as incomplete (n = 14). A total of 4107 adults qualified and completed the survey for a survey response rate of 74%. We focused our analysis on the sub-sample of respondents who had children age 5–18 who play sports in a youth or school sports programme (n = 674). Analyses of SummerStyles data are exempt from institutional review board approval because data do not include personal identifiers.

A series of 12 yes/no, Likert scale, and multiple-choice questions were included in this survey. The CTE questions were developed based on a review of the latest science outlined through the NIH CTE consensus conference (10,15) and a search for current CTE educational products available online. We identified four websites (37–40) (not inclusive of media websites or reports) for the public on CTE. We reviewed the information on each website to assess the type of content presented, understand if content on the website was consistent with the science, and whether the materials were written using plain language. After asking whether the respondent has a child 5–18 years who plays a contact sport, we asked parents a series of knowledge, opinion, information needs, and preference questions related to CTE. Topics included:

Age respondents think children should begin to participate in contact sports

The full question wording and response options are included in the tables.

Respondents were asked to self-report on a series of demographic variables at the beginning of the survey. Categorical variables for parent demographic characteristics included: sex (men, women), age group (18–34, 35–49, 50–64, ≥65 years); race/ethnicity (white non-Hispanic, black non-Hispanic, other); education (high school graduate or less, some college, college graduate); annual household income (less than $25,000, $25,000–49,999, $50,000–74,999, $75,000–99,999, $100,000 or more); marital status (currently married, currently cohabiting, formerly married, never married); metropolitan statistical area (MSA; non-metropolitan, metropolitan, other); and region (Northeast, Midwest, South, West)(41).

The analysis focused on the 674 respondents who indicated that they were the parent of a child age 5–18 who plays in a youth or school sports programme. We used SAS 9.3 to perform standard descriptive frequencies and bivariate associations. We performed chi square tests to compare categorical variables to determine whether there were significant differences between groups, specifically between parents who have children who play contact sports and parents who do not have children who play contact sports, and between men and women. Previous research has shown (a) concussions are more common in children who play contact sports (42) and (b) that men and women have different levels of experience, knowledge, and concern regarding concussions (43,44). For these reasons, we were interested in seeing if the level of knowledge and concern about CTE also varied by these factors. For the purposes of this analysis for the applicable questions, “agree” comprises those who responded “agree” and “slightly agree” while “disagree” comprises those who responded “disagree” and “slightly disagree.” A statistically significant difference was noted when two-sided P values < 0.05.

Table 1 shows that approximately half of the SummerStyles subsample (respondents who have children between the ages of 5–18 who play sports) was between the ages of 30–44 and another 42.6% was between the ages of 45–59. About 55.2% were women and 50.6% had a bachelor’s degree or higher. This was a relatively affluent sample, with 44.2% of respondents reporting a household annual income of $100,000 or more and another 17.4% reporting an income of $75,000–99,999 a year. Approximately 84.7% of respondents were currently married. About 86.5% lived in a metropolitan area and the sample was fairly evenly disbursed across regions of the USA.

Of the 674 respondents who had at least one child between the ages of 5 and 18 who plays in a youth or school sports programme, about half (49.3%) play contact sports (analysis not shown). Nearly half (47.7%) of the respondents who have children between the ages of 5 and 18 who play sports believed that children should start playing contact sports between the ages of 5 and 10 and another quarter (25.4%) believed they should start between the ages of 11 and 13 (Table 2). The responses to this question varied significantly (χ(2)= 89.1, P < 0.0001) by whether the respondent’s child currently plays a contact sport: a majority (70.2%) of parents who have children who play contact sports say that kids should start playing contact sports by age 10, compared to 36.5% of parents who do not have kids who play contact sports. This question did not show differences by gender; men and women were equally likely to report each age group as a proper starting place for contact sports.

When we asked how familiar respondents were with CTE on a 5-point Likert scale, about 38% reported being somewhat or very familiar. Parents whose children play contact sports were significantly more likely to be very or somewhat familiar with CTE (42.6%) than parents whose children do not play contact sports (34.3%) (χ(2)= 4.6, P = 0.0320). About 1 in 4 parents (28.6%) agreed that they were concerned that their child will get CTE from playing sports. Parents whose children play contact sports were more likely to agree that they were concerned that their child will develop CTE as a result of playing sports (37.0%) than parents who do not have kids who play contact sports (21.0%) (χ(2)= 20.2, P < 0.0001). Concern about CTE did not vary by gender.

Only 18.3% of respondents reported ever having received educational materials from a school or sports programme with information about CTE; another 14.7% were not sure whether they had received materials (Table 3). Doctors and nurses were the preferred source of information about CTE as 69.5% of respondents said they would like to receive educational materials or information about CTE from them. About 54% would like to receive information from their child’s coach or sports programme and 40.1% from their child’s school. In terms of the vehicle for information about CTE, respondents indicated a strong preference for handouts or fact sheets (76.5%). The second most common vehicle was through a website or online content (39.0%).

About 47.0% of respondents agreed that their child’s sports programme or school is taking the right steps to protect athletes from developing CTE (Table 4). A sizeable proportion chose the “neither” response option (as in neither agree nor disagree) for this question: 43.3%. Two-thirds of respondents knew that a person cannot get CTE exclusively from playing sports and 47.3% thought that changing the rules of play for contact sports can protect athletes from getting CTE. A little over half of the respondents (52.7%) knew that signs and symptoms of CTE do not show up right after a head or brain injury and only 10.5% believed that doctors and nurses have treatments to cure CTE. Finally, 12.0% of respondents believed that they can never protect their child from getting CTE while the vast majority believed they could protect their child at least some of the time.

While all of Porter Novelli’s surveys attempt to recruit a relatively representative sample, we utilized a select group of individuals who have children who play sports. Therefore, the representativeness of their responses is unclear and the sample of 674 parents should be considered a convenience sample. It is possible that these selected respondents or those who choose to participate in this survey more generally are different from the general population in ways that bias the results. Because this survey relied on self-assessment of familiarity with CTE, it is likely that there is some degree of inter-respondent variability on this measure and thus does not generalize to a wider population’s familiarity with CTE. This survey was also limited in scope; the focus was only on sports-related causes of CTE. Future research is needed to determine general awareness of CTE, and information and resource gaps among Americans overall, including key audiences.