Prevention of diabetic ketoacidosis at diagnosis of type 1 diabetes: Are certified athletic trainers an untapped resource?

Abstract Aims To assess the knowledge of certified athletic trainers (ATs) on the presenting signs and symptoms of type 1 diabetes (T1D). Methods We conducted a 31‐question survey of secondary school ATs recruited from the National Athletic Training Association that established demographic information, knowledge of presenting signs and symptoms of T1D, and previous personal or professional exposure to individuals with T1D. We report descriptive statistics and univariate analyses evaluating the characteristics associated with T1D knowledge. We then report a multivariable model incorporating age, gender, years of experience and education level with T1D knowledge as the dependent variable. Results 128 participants (92f:34m) met inclusion criteria and were included in this study. The majority of participants correctly identified frequent thirst (96.1%, n = 123) and frequent urination (85.9%, n = 110) as common presenting signs and symptoms of T1D, while fewer participants identified weight gain (58.6%, n = 75) or joint pain (39.1%, n = 50) as incorrect presenting signs and symptoms of T1D. Participants with over ten years of experience or previous exposure to individuals with T1D had increased T1D knowledge. Participants with advanced education (Master's degree or Doctorate) had no statistically significant difference in T1D knowledge compared to those with a Bachelor's degree. The only factor that demonstrated a significant association with T1D knowledge on multivariable analysis was the female gender. Conclusions Educational awareness campaigns of T1D symptoms to reduce the rate of DKA at diagnosis of T1D have never included ATs. This study illustrates the importance of targeting future educational interventions on newly trained ATs.

• Educational interventions on the common presenting symptoms of type 1 diabetes (T1D) to childhood contacts have successfully reduced the rate of diabetic ketoacidosis at diagnosis of T1D.
• No previous educational interventions have included athletic trainers (ATs). No literature reported ATs knowledge of the presenting symptoms of T1D and characteristics associated with increased knowledge.
• Experienced ATs (>10 years) or previous exposure to an individual with T1D was positively associated with T1D knowledge in univariate analysis. The only factor that demonstrated an association in multivariable analysis was the female gender. Our study highlights the importance of targeting newly trained ATs in future T1D education.

| INTRODUC TI ON
The overall incidence rate of type 1 diabetes (T1D) in United States youth has increased 1.9% each year from 2001 to 2015, with the greatest incidence rate being in children aged 10-14 and 15-19. 1 The majority of children with new-onset T1D present with one or two of the four main symptoms of diabetes: polyuria, polydipsia, weight loss and tiredness. 2 It has been reported in the United States that approximately 30% of new-onset T1D cases progressed to diabetic ketoacidosis (DKA) before diagnosis. 3 Presentation during DKA has devastating complications including increased morbidity, mortality and psychological distress, and is a predictor of poor long-term glycaemic control. 4 In addition, those diagnosed during DKA have a significantly higher use of healthcare services and increased medical costs. 5 Lack of awareness by patients and physicians of the presenting signs and symptoms of diabetes accounts for the majority of DKA cases at time of T1D diagnosis. 6 Attempts to improve public awareness of the presenting signs and symptoms of T1D have been successful in decreasing rates of DKA during diagnosis of T1D. 7 In the United States, these T1D awareness campaigns have focused on patients, parents, school teachers and healthcare providers. 8 Certified athletic trainers (ATs) often spend considerable time with student athletes,therefore, we hypothesize that it is especially important that they are targeted in future T1D awareness campaigns. 9 ATs, as a member of the healthcare team, should identify and refer athletes presenting with signs and symptoms of T1D. 10 There is a gap in the literature regarding ATs ability to identify presenting signs and symptoms of T1D and factors, including educational background, that influence AT knowledge of T1D.
This study aimed to bridge that gap by assessing the knowledge of presenting signs and symptoms of T1D and identifying factors that influence this knowledge through a national survey of ATs in the United States.

| PARTI CIPANTS
A national survey was powered to include at least 125 participants to capture the participants estimated to obtain statistical significance for differences in AT knowledge of T1D presenting symptoms. The survey was distributed electronically to 2975 secondary school ATs at a cost of $0.15 per survey, thus collecting data beyond what was statistically necessary was beyond the scope of the budget. The survey was subsequently closed after reaching the desired sample size.
The survey was distributed in accordance with the National Athletic Training Association internal process to obtain a broad representative sample of ATs throughout a three-week span in October 2019.
Inclusion criteria included ATs in the United States, able to read, write and communicate in English, have Internet access, and willing to consent and complete study requirements.

| ME THODS
The cross-sectional survey study was approved by the Institutional Review Board at Penn State College of Medicine. Before proceeding to the survey, participants read a summary of explanation of research and were asked if they agreed to participate in the study.
They were also asked eligibility questions, which either allowed them to continue to the survey, or notified them that they were not eligible, and thanked them for their time.
A 31-question survey was constructed using Qualtrics survey generation software. 11 The survey featured questions on the presenting signs and symptoms of T1D (T1D knowledge assessment), certified athletic trainer-specific information, and demographics.
The T1D knowledge assessment gave participants a clinical presentation that included either a true symptom of T1D (frequent thirst, frequent urination, changes in behaviour/mood and changes in appetite) or a false symptom of T1D (joint pain and weight gain). They were given three answer options: true, false or unsure. They were also asked to rate their confidence level on their answer choice (very confident, somewhat confident, or not confident). Certified athletic trainer-specific information included professional and personal encounters with individuals living with T1D, frequency ATs reported undiagnosed T1D in an athlete, and ATs opinion on continuing education in type 1 diabetes. Demographic information included age, sex, zip code, race and education level.

| Statistical analysis
The T1D knowledge assessment (range: −18 to 18) was scored based on the ability to correctly identify true and false symptoms of T1D. The scoring incorporated the ATs confidence level, as shown in Figure 1.
ATs were dichotomized into two groups (low or high knowledge) based on their T1D knowledge score in relation to the median score.
First, differences in T1D knowledge were examined using two-sample t tests for demographics (race, gender and education level), professional and personal encounters with individuals living with T1D, and years of experience as an AT. An analysis of variance test determined the temporal impact of years of experience on T1D knowledge. We then report a multivariable model incorporating age, gender, years of experience and education level with T1D knowledge as the dependent variable.

| RE SULTS
A total of 128 participants (response rate = 6.4%) completed the survey, including representation from 35 states. Demographic information is presented in Table 1. The participants were primarily female  Figure 2 shows the distribution of participant scores on the T1D knowledge assessment (median = 9). Participants were assigned to a high (n = 60) or low (n = 68) T1D knowledge group based on their T1D knowledge score relative to the median score.
An increased age (P = .005) and years of experience as an AT (P = .003) were positively associated with T1D knowledge, particularly among ATs with 11-15 years (P = .02) or greater than 15 years (P = .001) experience. Both previous professional (P = .04) and personal (P = .02) exposure to an individual with T1D was positively associated with T1D knowledge. Participants with advanced education (Master's degree or Doctorate in athletic training) had no significant difference in T1D knowledge compared to those with a Bachelor's degree in both univariable and multivariable analysis. There were no statistically significant differences in T1D knowledge based on race or area of work (rural, suburban or urban). The only factor that demonstrated a significant association with T1D knowledge on multivariable analysis was the female gender (P = .03). While the majority of participants (71.6%, n = 91) never raised concern of T1D in an undiagnosed athlete, those with increased T1D knowledge were more likely to have reported their concern of undiagnosed T1D in an athlete (P = .03).

| D ISCUSS I ON
In efforts to improve early detection of T1D, there have been attempts to educate child contact points on the presenting signs and symptoms of T1D (polyuria, polydipsia, weight loss and tiredness). 2,3 The early identification of these symptoms may prevent the development of

ACK N OWLED G EM ENTS
The authors would like to thank Katie Scott, The National Athletic Trainers' Association and the athletic trainers that completed our pilot study. The study has institutional IRB approval.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.