Confidence in eating disorder knowledge does not predict actual knowledge in collegiate female athletes

Introduction

Eating Disorders (EDs) are serious mental illnesses affecting millions of individuals worldwide regardless of race, age, nationality, or sex, and are associated with high levels of morbidity and mortality (Chesney, Goodwin & Fazel, 2014; Keshaviah et al., 2014; Micali et al., 2015; Schaumberg et al., in press). The cumulative lifetime risk of an ED is around 4.6% and an even larger number of individuals (nearly 10%) meet criteria for a subthreshold ED diagnosis ( Hudson et al., 2007). EDs (including anorexia and bulimia) and subthreshold EDs (not fully meeting diagnostic criteria for either anorexia and bulimia) are more prevalent among female athletes compared to non-athletes, and these disordered behavior can dramatically affect an athlete’s health and performance (Bratland-Sanda & Sundgot-Borgen, 2013; Joy, Kussman & Nattiv, 2016; Segura-García et al., 2010; Sundgot-Borgen & Torstveit, 2010). For female athletes, this is a particular concern because energy restriction increases the risk for the female athlete triad, referring to three interrelated health threats, consisting of inadequate energy availability, menstrual disorders, and decreased bone mineral density (De Souza et al., 2014; Nattiv et al., 2007). Considering these pathologies can largely impact future bone and musculoskeletal health as well as quality of life, (Mueller et al., 2015; Palacios et al., 2014) it is imperative to identify and treat disordered eating behaviors and EDs as early as possible (Montenegro, 2006; Papathomas & Lavallee, 2012; Sundgot-Borgen & Torstveit, 2010).

A substantial amount of research exists on ED risk and prevalence within athletes at large universities (e.g., NCAA Division I and II) (Abood, Black & Birnbaum, 2004; DiPasquale & Petrie, 2013; Gutgesell, Moreau & Thompson, 2003; Nagel et al., 2000; Rosen et al., 1986; Sanford-Martens et al., 2005); (Sherman et al., 2005; Sundgot-Borgen & Torstveit, 2010; Turner & Bass, 2001). However, little research has been completed at institutions with smaller athletic programs, for example National Association of Intercollegiate Athletics (NAIA) universities. NAIA universities (<2,000 students) often lack the resources that are readily available at larger universities, such as access to team-specific physicians, team-specific athletic trainers, or dieticians. Furthermore, recent literature suggests that small universities may have higher prevalence of mental illnesses, including disordered eating (Eisenberg, Hunt & Speer, 2013). Therefore, athletes at small universities represent an underrepresented research population with unique requirements for identification and treatment of disordered eating in athletes compared to larger universities.

In general, teammates and coaches spend a significant amount of time together, often times consuming multiple meals a week together. As such teammates and coaches may have the greatest opportunity to identify inappropriate eating behaviors and advocate professional treatment by allied health provider (Sherman et al., 2005). However, athletes and coaches generally do not have sufficient knowledge of ED behaviors or appropriate nutrition knowledge (Govero & Bushman, 2003; Torres-McGehee et al., 2011; Turk & Prentice, 1999) . A significant gap in ED knowledge is evident in prior research in coaches (Torres-McGehee et al., 2012; Turk & Prentice, 1999). This trend of inadequate knowledge is comparable with that of student-athletes (Torres-McGehee et al., 2011). These studies suggest that lack of knowledge may be a potential problem for collegiate athletics and amplify the difficulty with identifying problematic eating behaviors before they develop into pathological EDs.

One potential mechanism for lack of knowledge may be due to overconfidence (Dunning Kruger effect). The Dunning Kruger effect, first described in the 1990s (Kruger & Dunning, 1999) describes the phenomena where an individual may perceive high confidence in their knowledge on a subject when in fact, the individual does not have the ability to recognize the limitation of their knowledge. Coaches or teammates who are unable to recognize problematic eating behavior but having confidence in their ability to do can have serious ramifications. However, to our knowledge, no study has specifically investigated how athletes’ confidence in their knowledge relates to their actual knowledge in NAIA athletes. Therefore, the purpose of this study was to NAIA female athletes’ ED knowledge, confidence in their knowledge, and if confidence positively or negatively predicts ED knowledge. We hypothesized that female athletes would not have sufficient ED knowledge, but despite insufficient knowledge, would have high confidence in their ability to address EDs.

Methods

Results

A total of 51 college athletes (mean age = 19.7, SD = 1.3) enrolled in the study and completed all measures. Demographics of sport participation can be found in Table 1. Of the 51 college athletes, the average score of eating disorder knowledge was 69.1%, SD = 10.8% and in general, individuals were fairly confidence in their knowledge (mean = 3.67/5, SD = 0.33). Twelve participants (23.4%) scored an adequate score (>80%), with the remaining participants 39 participants receiving unsatisfactory scores (<80%).

Of the five subscales (signs/symptoms, risk factors, etiology, prevention, education, management), identifying risk factors had the highest average percentage (mean = 81%, SD = 14.4%) compared to all other subscales (p < 0.05) (Fig. 1), whereas, identifying signs and symptoms subscale was lower than every domain other than the etiology subscale (mean = 57.2%, SD = 21.1%, p < 0.05). There were no other differences noted between groups (Fig. 1). Interestingly, confidence scores were not different on any subscales (Fig. 2) with the exception of etiology compared to education and prevention.

Regression analysis demonstrated no relationship between confidence and total knowledge scores (F(1, 49) = 1.94, R² = 0.038, b = 0.063, p > 0.05) (Fig. S1). Through an exploratory analysis we also found a small but statistically significant relationship between participant age and total knowledge scores (F(1, 49) = 9.17, R² = 0.16, b = .0321, p < 0.05) was noted (Fig. S2).

Discussion

To our knowledge, we are the first group to analyze eating disorder knowledge specifically in NAIA athletes. The student athletes surveyed lacked eating disorder knowledge; specifically, they scored the lowest on the identification of ED signs and symptoms domain. Further, confidence in perceived knowledge was not predicative of actual knowledge regarding ED knowledge. These results provide preliminary evidence that student athletes do not recognize their lack of ED knowledge, and therefore, may be ill equipped to recognize unhealthy eating behaviors.

After a preliminary analysis of the data, we noticed a relationship between age and knowledge in our sample; therefore, we conducted a post hoc exploratory relationship between age and overall ED knowledge. We found a small relationship between athlete age and increased scores; to our knowledge this is the first study to compare knowledge in relation to athletes’ age. It is possible that older students have received knowledge through an outside source during their time at college, such as a nutrition course or a seminar on EDs not specifically sponsored by athletics or some educational program. However, due to limitations of the survey, it is impossible to determine why the older students tended to have higher knowledge compared to younger. However, because this was an exploratory analysis, not the original aims of the study, we urge a cautious interpretation of the data and encourage future research to investigate mechanisms by which age may influence knowledge of EDs.

Overall, many participants did not have appropriate ED knowledge (>80%), however participants still tended to be confident in the accuracy of their knowledge (average confidence > 3.5/5). The results are consistent with previous research utilizing this survey (Torres-McGehee et al., 2011; Turk & Prentice, 1999), and others that have investigated similar domains (Torres-McGehee et al., 2012) demonstrating little knowledge, but high confidence in perceived knowledge. This discrepancy in knowledge and confidence has been seen across many psychological domains (Pennycook et al., 2017; Simons, 2013). Specifically, for this population, their lack of knowledge in relation to their confidence represents a potential health risk for female student athletes. Depending on the university’s resources, there may be limited information available to athletes regarding healthy weight loss and nutrition (Sherman et al., 2005). Thus, athletes often ask teammates for nutritional advice (Bond et al., 2008; Reel & Gill, 1996; Turocy et al., 2011). However, the accuracy of this advice may be questionable (Torres-McGehee et al., 2012; Turk & Prentice, 1999; Turner & Bass, 2001; Wiita & Stombaugh, 1995), yet our data suggests that they may still be confident in their knowledge, which could contribute to unhealthy eating behaviors.

The ability to identify signs and symptoms was the lowest subdomain in this population. This is concerning as teammates are often in a position to recognize problematic eating behaviors and suggest further treatment due to amount of time spent together (Sherman et al., 2005). This may be especially true of athletes at NAIA universities, which may share practitioners with training in nutrition, such athletic trainers, across multiple teams. Athletic trainers monitoring multiple teams may limit opportunities to identify unhealthy eating behaviors. If teammates do not know what behaviors are associated with EDs, then other teammates engaging in pathological eating behaviors may go unnoticed. Moreover, unhealthy eating behaviors may even by female athletes are not familiar with symptoms of EDs and suggests they may not recognize these unhealthy behaviors in teammates. Taken together, a lack in knowledge by fellow teammates as well as limited athletic training staff may lead to unnoticed unhealthy eating behaviors, placing female athletes in NAIA athletic programs at probable risk. However, we should note that this study was only conducted at one institution, therefore the results are fairly preliminary and this research topic should be further investigated across multiple NAIA universities.

In conclusion, this study reiterates there is poor knowledge of EDs among female collegiate athletes. This is the first study to investigate participants at small collegiate athletic programs, an underrepresented population in the field. Specifically, our data demonstrate that female student athletes clearly lack the knowledge to identify the signs and symptoms of disordered eating in their peers, yet have high confidence in their capacity to do so. Coaches, athletic directors and athletic trainers should be educated on these conditions and educate their respective athletes on EDs. While it should not be the expectation that athletes understand the intricacies of nutrition and EDs, due to the large time spent with teammates, they are likely in the optimal position to notice unhealthy eating behaviors in teammates, and it would be beneficial for them to understand early indicators of problematic eating behaviors and advocate treatment; this may help alleviate the occurrence of full blown EDs or catch disorders early.

Supplemental Information