The confidence and competence of primary school staff to administer an adrenaline auto-injector

Anaphylaxis is a severe life-threatening hypersensitivity reaction, with acute onset and multisystem involvement. A leading cause of anaphylaxis is immunoglobulin E (IgE)-mediated food allergy [1]. The estimated prevalence of food allergies is between 1.4% and 3.8% in school children [2]. The signs and symptoms of anaphylaxis are varied, with potential to involve the skin, respiratory and gastrointestinal tracts, and the cardiovascular and neurological systems [3]. The recommended treatment for anaphylaxis is adrenaline, given via the intramuscular route, usually delivered in the community with an adrenaline auto-injector (AAI) [4]. Early administration of adrenaline is recommended once anaphylaxis is recognised. However, there are often delays in treatment of anaphylaxis especially in the community setting [5]. Delay in administration can result in an increased risk of death [6, 7].

Due to the large amount of time children spend in education, it is not uncommon for anaphylaxis to occur within a school setting. Reported incidence of anaphylaxis in the school setting varies from 10% to 16% [5, 8, 9], with almost half of these occurring in children without a prior known food allergy [10]. However, research on the competence of school staff to use AAI is limited. A study carried out in Germany in the preschool setting found that levels of competence in AAI use were low—although there was a positive improvement following an educational intervention [11]. A weakness of some of the research in the area is that data on the confidence to use an AAI has tended to be based upon self-report [12, 13], rather than through observation of actual performance. This is a large limitation as it has been shown that people with a low ability in a specific area can give overly positive assessments of their ability. This phenomenon has been described as the Dunning-Kruger effect [14]. This effect has been identified in a range of studies—including with medical profession [14, 15].

Our study will address the shortcomings of previous research by: (1) assessing the observed and self-reported confidence of school staff in the Republic of Ireland (RoI) to use an AAI; (2) evaluating if there is an association between confidence and competence; and (3), based on the findings, make recommendations for future research and practice.

Food allergies are increasingly common in children [2]. Therefore, teachers are more likely than ever to be faced with the challenge of managing a pupil with anaphylaxis. However, teachers are often inadequately prepared to use an AAI, resulting in potentially dangerous treatment delays [4, 7, 17, 18]. Our study evaluated the confidence of teachers to use an AAI, the competence of teachers in performing the steps required to administer an AAI, and assessed if there is a link between confidence and competence.

The majority of the participants in our study reported little or no confidence in their ability to correctly use an AAI. This was even the case for the six participants who correctly performed the procedure, and for those respondents who had attended AAI training. Our findings agree with other studies that have found that even with training, parents and caregivers have low levels of confidence and high levels of anxiety in correctly using an AAI [19‐21]. This lack of confidence may lead to AAI use hesitancy in which treatment is delayed because caregivers are afraid to use the device. Evidence for this suggestion comes from a survey in which, of those respondents (parents or people with food allergies) who had experienced anaphylaxis, only a third had administered an AAI. The majority of the respondents reported that treatment was not given until the person experiencing anaphylaxis got to the Emergency Department [22]. This delay is concerning as early administration of adrenaline is important in the treatment of anaphylaxis, and a delay in administration has been found to be associated with an increased risk of death [23]. Therefore, it is important to address the lack of confidence that may lead to hesitancy in administering an AAI.

Only six participants in our study performed every step correctly in the administration of the AAI. The most frequently observed error was not holding the device in place for 10 seconds. Almost a third of participants held the device in place for less than 3 seconds. Eight participants incorrectly inverted the AAI such that they would have experienced a digital self-injection if a real AAI was used. Our findings are similar to other studies that have observed AAI administration with a teacher population [11, 16]. Although there was not an impact of prior training on confidence, there was a large effect of training on the number of steps performed correctly. The positive effect of training on AAI use is consistent with findings from other studies that have examined the impact of AAI training for preschool staff [11], medical staff [24], caregivers [25], and patients [26]. To some extent, the impact of training was a little unexpected. None of the participants had received AAI administration training in the last 6 months, and research has demonstrated skill decay in AAI administration 4 to 12 weeks after training [11]. Another interesting finding was that three of the participants who completed all of the steps correctly had not received any AAI training and were able to deliver the AAI successfully by using the instructions printed on the body of the device. Therefore, at least in simulated setting, there is utility in these instructions if they are noticed and used.

Our study did not find a significant correlation between observed performance and self-reported confidence. Therefore, the Dunning-Kruger effect [14]—in which people tend to overestimate their performance—was not observed. Our findings contrast with an evaluation of AAI training carried out with preschool teachers that found that the effect an education intervention had on confidence remained high even 4 to 12 weeks after the AAI training. However, the percentage of correct AAI administrations was 3% before the training, 35% immediately after the training, and 16% 4 to 12 weeks after the training [11]. Either way, both studies demonstrate that, although determining competency based upon self-reported confidence is attractive as it requires little resources to complete, it is also generally invalid. Training in AAI administration would appear to be necessary, but not sufficient, for teachers to be confident in administering an AAI.

It is possible to make recommendations based upon the findings of our study for researchers, schools, education authorities, and AAI manufacturers. It is suggested that researchers consider the efficacy and efficiency of different approaches to training on both observed competence and confidence. Precision teaching using simulation is suggested as an educational approach to build fluency in AAI administration. Once fluency has been attained, it is possible to complete the task with little conscious effort—even when the training was completed a long time ago [27]. Precision teaching, utilising simulation, has been found to be an efficient and effective approach to building fluency in a number of clinical skills (e.g. venepuncture [28], lumbar puncture [28]). Schools and education authorities must recognise the increased likelihood of teachers encountering anaphylaxis and develop robust policies, and training, so that teachers are adequately prepared to manage anaphylaxis. Consideration should also be given to keeping AAI devices in schools. Although this is not done in the RoI, this is the case in countries such as USA, Canada, and England [29]. Teachers would then only have to be trained to use the AAI that is stocked by the school. This would address the issue of not knowing how to use all four types of AAI prescribed in Ireland. It would also mean that the teachers could administer a second dose if required. A survey of adults found that 82% did not carry a second AAI—despite the fact that a second dose of adrenaline is commonly required [5]. Finally, it is suggested that AAI manufactures should use the principles of human factors engineering to consider how the design of AAI could be changed in order to make the devices easier (and safer) to administer and address issues with variability in operations between devices. To illustrate, the voice-prompted AUVI-Q device, developed with patient input, has shown promising results in comparison to other devices—even with device-naive users [30, 31].

There are a number of limitations that should be acknowledged. Only staff from four primary schools participated in the study from one city in the RoI. Therefore, the generalisability of the findings, particularly outside the RoI, could be questioned. The administration of only one AAI was assessed in this study. However, similar levels of competency have been found in using both the Jext^® and Epipen^® AAIs [30], so we believe that the findings generalise to other AAIs. We only assessed the ability of the participants to use an AAI. We did not assess their knowledge of the symptoms, or management, of anaphylaxis. Finally, our recruitment method could have led to selection bias. Previous research has shown that if teachers are aware that they are going to be assessed administrating an AAI, then their performance is superior to teachers who were not informed of the assessment [16]. However. the ethics board did not deem it appropriate to blind the participants to the assessment. Therefore, the level of performance may be lower in the general teacher population.

Due to the increasing prevalence of food allergies. teachers are more likely than ever to be faced with the challenge of managing a pupil with anaphylaxis. However, although not a complex task, teachers largely lack the confidence and competence for AAI administration. Therefore, action is required to ensure teachers are confident and competent to manage this potentially life-threatening condition. This can be achieved through the application of approaches: teach clinical skills such as precision teaching, changes to school policies and practices, and consideration of the design of AAIs in order to make their operation simpler.