Abstract
Background:
Stroke is a worldwide leading cause of disability, and around 50 million people are affected by stroke annually. Public stroke educational and awareness programmes can make a great deal of difference. Young children are in key position to seek urgent medical care if grandparents suffer a stroke, as grandparents are often their secondary caretakers.
Objective:
The objective of the current study was to design an educational intervention targeting children and, in parallel, directly involve extended family members.
Design:
Participatory action research.
Setting:
School-based education stroke intervention in Thessaloniki, Northern Greece.
Methods:
Over the course of 5 weeks, this intervention sought to educate 66 preschool children and their families through a series of novel activities that revolved around 4 superheroes, the FAST mnemonic and a medical emergency number. One superhero and their unique superpower was introduced every week through a Powerpoint presentation, a short animation video and a wide range of in-class and take-home activities, such as ‘phantom speech’, role-playing, funny face mimicking games, and rhyming poems. Children were also encouraged to identify their own family superhero so as to transfer their learning to real life.
Results:
Follow-up individual and group assessment pointed to encouraging results. Results from odd-one-out-tasks revealed that children were able to recognise the stroke symptoms in question. However, they performed more poorly on more complex tasks involving recall.
Conclusions:
Preschool children acquired knowledge of stroke symptoms which appears more solid when recognition is assessed. Assessment tasks involving substantial recall of information do not necessarily reflect the ability to detect stroke symptoms.
Introduction
Stroke is a worldwide leading cause of both morbidity and mortality (Ingall, 2004). Strokes can be broadly classified as haemorrhagic and ischaemic, with ischaemic strokes making up about 80% of all strokes (Aguilar, 2015). In the case of acute ischaemic strokes, thrombolysis can restore blood flow if patients present to the hospital within 45 hours post onset (Gonzalez-Aquines et al., 2018; Hacke et al., 2008). Urgent admission to the hospital is, therefore, key, but treatment delivery is often hindered by patient delay (Mellon et al., 2015). Public ignorance of stroke symptoms acts as a barrier to acute stroke treatment (Hodgson et al., 2007). Receiving treatment in a timely fashion can be also hampered by the lack of appreciating the importance of urgently seeking medical care. Importantly, the interpretation of the symptoms rather than their mere recognition is a critical factor leading to prompt hospital admission (Gonzalez-Aquines et al., 2018).
Mass media campaigns appear successful in raising stroke symptom awareness in adults, especially when they present participants with clear and continuous messages (Hodgson et al., 2007). In a systematic literature review, Mellon et al. (2015) explored the impact of public educational interventions designed to reduce patient delay. Results pointed to different success outcomes across studies, with particularly successful projects integrating different approaches, such as mass media campaigns, targeted community intervention, and professional education. Building on the first national stroke awareness campaign in Ireland, Hickey et al. (2018) conducted a follow-up study, comparing pre-campaign to post-campaign data. Results revealed significant improvements in terms of stroke knowledge, especially in relation to the FAST (facial droop, arm weakness, speech difficulties and time to seek medical care) symptoms.
Educational interventions targeting children also seem to have significant potential, as children spend a lot of time with their grandparents across various cultures (Pulgaron et al., 2016). Children are likely to witness their guardian undergo a stroke and may be in prime position to get in touch with emergency medical services (Davis, 2008). Amano et al. (2014) designed a high school educational intervention employing the FAST mnemonic and several items, such as posters, pens and sticky notes bearing the mnemonic. They also sought to motivate children to relay their learning to their families. Among other symptoms, students and their parents were found to have successfully learned the FAST message. In a different study, Williams and Noble (2008) designed a school-based intervention programme, entitled ‘Hip Hop Stroke’ (HSS) to educate children aged 9–11 years through music and dance, also involving their families. The programme included 1-hour sessions over three successive days and was found to be effective in terms of stroke recognition and urgent action-taking immediately post intervention but also 3 months post-intervention. In a more recent study, children of a similar age group and their parents participated in the same intervention (Williams et al., 2018). To assess child and parental stroke preparedness, HSS participants were compared to a control group. Overall, parents and children assigned to the HSS group displayed greater stroke preparedness compared to controls.
Prior research shows that mass media intervention ceases once the campaign is over so that a sense of sustainability through family ownership is crucial to programme effectiveness. That is, to make the impact of interventions more sustainable, adherence to theoretical underpinnings as well as immersion of family involvement supplemented by home activities and a website illustrated of cartoons and computer-based activities, respectively (Miyashita et al., 2014; Morgenstern et al., 2007; Sakamoto et al., 2014).
Aims and hypotheses
Tapping into the FAST mnemonic, the present intervention, FAST Heroes 112, sought to educate preschool children and their extended families to recognise the main stroke symptoms and take action urgently in the event of a stroke. We hypothesised the following:
1a. Children whose parents completed some or all tasks at home would do better in the 1 week in class follow-up assessment than those whose parents did not complete any task at home.
1b. Children whose parents completed some or all tasks at home would do better in the 3-week follow-up assessment than those whose parents did not complete any task at home.
2. Children who attended all in-class programme activities would do better in the 3-week follow-up assessment than those who attended only some activities.
Materials and methods
Participants and simulation
The materials and interactive activities employed revolved around four cartoon superheroes and the intervention took place in a stimulating and entertaining environment. The educational team developed a systematic educational process based on cognitive research that was linked to the educational foundations, while teachers and team members ensured collaboration with the extended family and school classroom to create this stroke awareness curriculum. The high-quality graphics were used to capture children’s imagination and motivate them to engage in the activities, despite their young age.
Sixty-six kindergarten children, 38 boys and 28 girls, aged 5–6 years old participated in the intervention which took place in Thessaloniki, Northern Greece. Three of the children had learning difficulties. Participant information sheets and consent forms were handed out to participants’ parents prior to the commencement of the intervention. Informed consent was obtained from all the families that were contacted. Participants’ anonymity was preserved throughout the intervention and the study was conducted in accordance to the Declaration of Helsinki.
Ten children aged 5–6 years participated in a 3-hour simulation session, which took place before the experimental portion of the study. The session aimed to evaluate whether the cognitive load of the activities was age-appropriate and to set realistic educational goals.
Focus group meetings with parents were also set up one day before and during the simulation session. This was to discuss the intervention with the parents and to explore the potential of take-home activities as part of the programme. Their feedback was taken on board to fine-tune the activities and the material.
A team of professionals was established, called Super Grand League consisted of cognitive psychology researchers, kindergarten teachers, special education teachers, linguists, speech-pathologists, health care professionals, graphic designers and animators who helped develop, manage and implement the programme.
Study design
Named after the FAST acronym, four superheroes were introduced to participants: Franc (Face), Armando (Arm), Sophia (Speech) and Timmy (Time). Memorising the symptoms was also linked to the Pan-European emergency number 112. That is, the number itself was used as an additional mnemonic, with the first digit standing for one side of the face drooping, the second for one arm losing its strength and the third for the two lips becoming incoherent. To help the children transfer their learning into a real-world context, we asked them to identify their own hero, encouraging them to choose one of their grandparents and, if not, some other caregiver. Posters with the superheroes and their superpowers are shown in Figure 1.

FAST Heroes 112 posters for distribution to individual classrooms.
FAST Heroes 112 was tailored to children aged 5–6 years. Following Week 0 (Introduction), the rest of the programme (Weeks 1–4), introduced the four heroes and their superpowers to the children through weekly hourly sessions. In line with the popular belief that all heroes fight a villain, children were told that their superheroes’ enemy is called the Evil Clot and attacks their heroes by targeting their brain and causing them to suffer a stroke.
Each superhero was introduced through a short narrative, a power point presentation and a short animation video. The Evil Clot was also incorporated in the narrative and so were the symptoms that compromised the heroes’ superpowers. To further familiarise themselves with the material, children took part in a number of activities (including phantom speech, funny face mimicking games, role-playing and kinaesthetic games). Phantom speech (Bluemel, 1957) involves talking without voicing but exaggerated lip and tongue movement or what is also called verbal imagery. Another activity involved saying words with scrambled syllables, reflecting phonemic paraphasic errors of an aphasia-related stroke survivor. Experiential games and rhyming poems were employed to carry out an interim evaluation of what children had taken in. The children also had the opportunity to test their knowledge in real-life practice conditions through role-playing focusing on calling 112. Practice scenarios required them to provide their name, surname, home address and the stroke symptom that they had recognised. Additional activities including handicrafts, music and acting were used (Table 1).
Programme overview (Weeks 0–4) describing aims and activities.
The sessions concluded with a recap of the main learning points and a take-home task. Children were also offered a superhero poster as a reward for their participation and a reminder of their take-home task. They were also handed over take-home worksheets aiming at collecting feedback on what they had absorbed from the oral presentations. Children were promised an additional reward on the condition that they completed the take-home task and the questionnaire with their parents. At a later stage of the study, the take-home material was processed by our team and forwarded to the grandparents who were identified as the children’s superheroes and a questionnaire was handed over to parents, requesting them to report the level of their contentment in the programme.
Programme assessment
In order to assess transfer of stroke information, for week-by-week data collection, we used postcards with the task Circle the Correct Picture (see Figures 2(b) and (d)), which required them to work individually in the classroom. Alongside the aforementioned postcard activity, children’s hand dominance was also assessed, through an in-class activity (Figure 2(c)); they had to hold up both arms until one drifted down. This was to assess arm strength in conjunction with hand dominance. One week later, we handed over a review worksheet, with an odd-one-out task (online supplemental Appendix 1, note: the acronym FAST in the Greek language is XOΠΑ, Χ-Χέρι [Arm], O-Ομιλία (Speech), Π-Πρόσωπο (Face), Α-Αμέσως (Time)). To complete the task, children had to distinguish between situations in which the heroes had intact superpowers and others in which they manifested a stroke symptom.

FAST Heroes 112 postcards for the in-class activities: a) Card drawing activity completed in the first week; b) ‘Circle the correct answer’ card completed in the second week; c) Bottles arm task card completed in the third week; d) ‘Circle the correct answer’ card completed in the fifth week.
A threefold follow-up test was administered individually to each student three weeks after the end of the intervention (online supplemental Appendix 2). The first subtest was a triadic comparison task in which children were expected to link either the Evil Clot or the Hero with the matching target picture. This subtest consisted of six sets of pictures. The second subtest was a sentence completion task in which children heard four sentences and had to fill in the last word (target words: speech, Franc, help/ambulance/stroke, arm). The third subtest presented children with three different cards displaying the superheroes manifesting the FAST symptoms and required them to describe, at first in a spontaneous verbalisation and then by providing some hints, the appropriate course of action (recognition of the symptom, calling 112 and asking for help). Each child was tested individually in a quiet room.
We did not use a pretest/posttest design to test the effect of the intervention. In fact, pilot data from Williams and Noble (2008) suggests that much older participants, aged 9–11 years, were not aware of the FAST acronym prior to their intervention. Hence, such a design would be also inappropriate for the purposes of the present study.
Results
From a total of 66 children, 38 parents returned the questionnaire and two did not complete it. At the initial contact, we collected information regarding two grandparents or caregivers they considered as their superhero. Twenty-five children (37.9%) picked their grandmother as their own real-world superhero in the school postcard (see Figure 2(a)) and 12 children (18.2%) in the home Week 1 Face activity (Figure 2(b)) revealed that 43 (65.2%) out of 61 children answered correctly. Error analysis shows that 2 children (3.0%) circled the sad face, 2 (3.0%) circled the scared face, 1 (1.5%) circled the angry face and 13 (19.7%) circled more than 2 faces. In the home postcards, 19 (28.8%) children and parents answered correctly, 1 (1.5%) circled the tongue out face and 1 (1.5%) circled the happy face.
For week 2 Arm activity (Figure 2(c)), strength in relationship to hand dominance for writing was also examined through an in-class activity. Twenty-nine children (43.9%) were right-handed with their right hand/arm drifting down first, 24 were left-handed (36.4%) with their left arm/hand drifting down first, while 6 children (9.1%) were right-handed with their left arm/hand drifting down first. In the home activity, 28 parents (42.4%) filled in the same postcard.
The scores of correct answers for the school and home postcard activity of Week 4 with all symptoms (see Figure 2(d)) revealed that 38 children (57.6%) out of the 51 (numbers of children participating in activities changed in accordance to school attendance) responded correctly in the school activity, while 26 (39.4%) out of the 30 children gave a correct answer in the same home activity. School activity error analysis showed that 5 children (7.6%) made a mistake in Time, 3 children (4.5%) in Face, 2 children (3%) in Speech and 3 children (4.5%) made more than 1 mistake. Home activity error analysis captured the following mistakes; 3 children (4.5%) in Speech and 1 (1.5%) in Speech and Time.
Using a postcard for odd-one-out tasks for all three symptoms, we elicited 60 correct answers (90.9%) out of 62 children for face, 58 correct answers (87.9%) out of 58 children for arm, 50 correct answers (75.7%) out of 54 children for speech and 49 correct answers (74.3%) out of 53 for all stroke symptoms (online supplemental Appendix 1). After the education session, we conducted a 3-week follow-up, with three tests examining stroke knowledge retention.
In the triadic comparison task (online supplemental Appendix 2a), 39 children (59.1%) responded correctly in task scoring the highest rate, while the lowest percentage was found in the Comparison Task f with 29 correct answers (43.9%).
A four-sentence completion task was administered to the children (online supplemental Appendix 2b). Fifteen children (22.7%) gave all four correct answers, for children who made 1, 2 or 3 mistakes, the scores are 33.4%, 16.7% or 1.5% respectively. The most common mistake was made in Help (15.2%).
Results from the FAST 112 Sequence Task (online supplemental Appendix 2c) show that 21 children (31.8%) responded correctly; however, 27 children (40.9%) did not complete fully the appropriate course of action. In total, out of the 13 correct answers a child could score in the three tests conducted in the follow-up assessment, the mean number of correct responses was 6.53 (SD = 4.62).
A Kruskal–Wallis test was used for our first hypothesis, showing that there was no statistically significant difference in regard to parental involvement both for the in-class, χ2(3) = 1.999, p = .573, and the follow-up activities, χ2(3) = 2.291, p = .514. Differences in the mean ranks in each scoring in regard to graded parental involvement are depicted in Table 2.
Mean ranks of children’s in-class recapitulation and follow up scores in regard to graded parental involvement.
There was no statistically significant difference, χ2(2) = 0.205, p = .903, between children who attended every in-class activity and children who attended some. Table 3 contains data on the ranks of children’s follow-up scores with respect to their in-school activities participation.
Mean ranks of children’s follow up scores in regard to in-school activities participation.
Discussion
Given the opportunities young children have to contribute to the wellbeing of their grandparents (Davis, 2008; Pulgaron et al., 2016) and previous evidence that educational programmes can have a positive impact on children’s knowledge of stroke (Amano et al., 2014; Miyashita et al., 2014; Morgenstern et al., 2007; Sakamoto et al., 2014; Williams et al., 2018), the present study aimed to educate preschool children and their extended families on the FAST symptoms through a wide range of novel activities.
Using high-quality graphics, the designed activities represent an original approach to capturing the imagination of children and inviting them to identify their heroes but also to assume for themselves the ‘role’ of heroes. Children participating in the intervention worked with a number of specialists over the course of 5 weeks. In particular, the four FAST heroes and their enemy, the Evil Clot, were introduced to the children through narratives, Powerpoint presentations and animation videos. Learning activities involved role-playing, kinaesthetic games, poetry and music.
Overall, FAST Heroes 112 yielded encouraging results, with children performing successfully when individual symptoms were assessed and when assessment focused on recognition rather than recall. More specifically, children achieved high scores for in-class assessments and subsequent take-home activities. Despite some students’ non-attendance, and consequently their absence from some of the in-class activities, results indicate that the educational message was delivered to the children. Likewise, children seemed to have absorbed the information presented regardless of parental involvement. Although we believe that parental involvement may have indirectly affected the educational process, results following the Kruskal–Wallis test showed that omitting the home activities of the programme did not affect children’s in-school and follow-up performance.
As regards the 3-week follow-up assessment, and more specifically, the triadic comparison task, the majority of children demonstrated that, from all three symptoms, they understood very well the symptom of arm weakness, superhero Armando. By contrast, the poorest performance was found for the speech impairment symptom, superhero Sophia. It is known that speech disturbances are ‘invisible’. As such, they are difficult to illustrate and assess visually in a two-dimensional fashion, in contrast to arm weakening and one side of the face drooping.
The sentence completion task (also part of the 3-week follow-up) pointed to a more complex picture, as less than one third of children completed all sentences successfully. In particular, when children were prompted to fill in the sentence ‘I call 112 and I say’, they had great difficulty providing the target word (‘Help/Ambulance’).
The FAST 112 sequence task also proved to be challenging for children, with fewer than half of them having successfully memorised the correct action sequence (i.e. symptom recognition, calling 112 and asking for help). The sentence completion task and the FAST 112 sequence task brought the performance average down. However, while 40% of children did not recall all aspects of what needed to be said, most did show significant knowledge gains. The findings discussed above suggest that naming stroke symptoms was more challenging for children than recognising them. Although we did not design equivalent recall and recognition tasks, it could be argued that the tasks involving recall were more demanding in terms of both short-term and long-term memory processing (Camina and Güell, 2017; Uytun, 2018).
FAST Heroes 112 is a stroke educational campaign that integrates educational theories and scientific evidence, while enabling teachers and researchers to raise stroke awareness in the kindergarten classroom and beyond with possible long-lasting gains. The current intervention design could benefit from additional activities on calling 112, the Pan-European emergency number, and asking for help. In particular, isolating the act of calling for help from other learning points is likely to boost children’s performance by reducing working memory demands when it is incorporated in target course of action sequences.
Limitations
This study has some limitations. The follow-up evaluation session was conducted 3 weeks after the presentation of the programme because of limited time, due to the school summer break. There was a limited sample size, due to participants’ and the school’s availability at that time. Parental involvement was rather poor. One explanation for this could be that, similar to the parents of other private school students, the families taking part in this study tended to spend less time with their children at home, leaving their education to ‘experts’ (Gauthier et al., 2004). In addition, the current programme was a new initiative, and perhaps not so popular among members of the local community, which may also have hindered parents’ participation.
Concluding comments
Informed by the encouraging findings, we aim to extend this pilot study, with further testing in a larger randomised control trial in Thessaloniki and possibly other cities in Northern Greece. Future large-scale studies assigning preschool children to interventions involving different types of input, activities and testing can enhance our understanding and help fine-tune the most appropriate educational methods for this age group.
Supplemental Material
Online_supplemental_Appendices_1 – Supplemental material for FAST 112 HEROES: A kindergarten-based educational stroke intervention for the whole family
Supplemental material, Online_supplemental_Appendices_1 for FAST 112 HEROES: A kindergarten-based educational stroke intervention for the whole family by Kalliopi Tsakpounidou, Ariadne Loutrari, Freideriki Tselekidou, Maria Baskini and Hariklia Proios in Health Education Journal
Footnotes
Acknowledgements
We thank Lucan Visuals, everybody who participated in the study and members of the Super Grand League Team.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Boehringer Ingelheim Angels Initiative (grant number 395479, 2019).
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
