We conducted a multicenter, prospective, quasi-experimental study in five state-subsidized private schools.

Five schools in the autonomous community of Galicia, Spain, participated in the study between January 2025 and January 2026, with an initial sample of 330 children aged 5–8 years enrolled in first and second grade of primary education. For organizational and methodological reasons, and to avoid cross-contamination between groups through sharing of information, the schools were divided into 2 groups: a control group (CG), which received training using traditional resources (a slide presentation and a manikin), and an intervention group (IG), which used 2 interactive manipulatives adapted to the developmental level of the students (the Endless Book and a telephone simulator).

The inclusion criteria were age 5–8 years and having no prior training in BLS. Children with any type of physical or visual disability were excluded from the analysis. In addition, informed consent from the parents and assent from the child were required for participation.

The intervention consisted of two 50-minute sessions delivered by the physical education teacher of each school, who had previously been trained by the research team, and followed the 2021 ERC guidelines.7

The methodology applied in both groups was identical in terms of the student-to-instructor ratio (25:1) and the manikin-to-student ratio (1:5), as well as the duration and structure of the sessions. To ensure consistency across sessions, the teachers were provided with a manual that included three standardized simulation scenarios.

The first session consisted of a theoretical explanation, which was delivered in the CG using slides and the teacher’s telephone simulator; at the IG, the students manipulated and explored the Endless Book (Fig. 1A) with the help of the teachers (there was one Endless Book per student) and simulated a phone call using a laminated cardboard phone simulator (Fig. 1B) (there was one phone per two students).

Figure 1.

(A) Scenes from the Endless book. (B) Low-cost phone simulator.

In the second session, the teachers gave a practical demonstration of the covered topics, and, last of all, the students participated in three simulations during which they received feedback from the teachers and discussed their performance (Fig. 2).

Figure 2.

Study design.

• •

  Endless Book: picture book with a short accompanying text designed to teach children how to recognize cardiac arrest and contact emergency services immediately. It features four scenes arranged in a cyclical pattern, so the story is always read in the same order (Fig. 1A).

• •

  A minimum-cost telephone simulator made from laminated cardboard (Fig. 1B). The goal was for the students to practice calling emergency services, just as they would with a real phone.

• •

  Instructor Manual: Teachers were provided with a manual outlining the methodological components, schedule, and classroom arrangements for the training, as well as the implementation of the three simulation scenarios.

Two assessments were conducted: the first in the 15 days following the training (E0) and the second at 12 months (E1).

In both groups, the students were presented with the following scenario: “You’re in your room playing, and suddenly you hear a noise in the living room. When you get there, you find your grandfather lying on the floor, face up, with a phone next to him. How would you react?”

A member of the research team played the role of the emergency services dispatcher once the student made the call.

The evaluation was performed with a checklist (Table 1) to asses yes/no dichotomous variables that has been used in previous studies.8

The research protocol was approved by the Ethics Committee of the University of Vigo, Spain (code 09-170123).

The data were analyzed with the Statistical Package for the Social Sciences (SPSS) software. We summarized quantitative data as absolute and relative frequencies. Groups were compared by means of the χ2 or the Fisher exact test, and changes over time (E0 vs E1) with the McNemar test. We calculated odds ratios (ORs) with the corresponding 95% confidence intervals (CIs) and assessed the size of the effect using the Cramer V: small, 0.10−0.30; moderate, 0.30−0.50; large, ≥0.50. For all tests, we considered p values of less than 0.05 statistically significant.

As can be seen in Table 3 and Fig. 3, more than 80% of children in both groups were able to perform 7 of the 10 assessed actions. There were statistically significant differences in favor of the IG for the following actions: emergency recognition, pressing the emergency SOS button, calling 112, describing the emergency, and providing the address (Table 3).

Figure 3.

Comparison of the results of the control and intervention groups in the evaluation conducted 15 days after the training. Asterisks indicate variables with statistically significant differences in favor of the intervention group (P < .005).

The actions with the lowest performance rates (fewer than 70% of participants) were providing a parent’s phone number (CG, 56.10 vs IG: 50.50%; P = .095) and turning on the speakerphone (CG, 65.80 vs IG, 61.00%; P = .241).

Table 3 and Fig. 4 show the results of the follow-up evaluation at one year for both groups. We found that, compared to the CG, students in the IG achieved better results in half of the evaluated actions, with statistically significant differences in pressing the emergency SOS button, calling 112, providing their name, and providing a parent’s phone number.

Figure 4.

Comparison of the results of the control and intervention groups in the evaluation conducted one after the training. Asterisks indicate variables with statistically significant differences in favor of the intervention group (P < .005).

In both groups, fewer than 60% of students turned on the speakerphone to enable hands-free use of the phone (CG, 50.80% vs IG, 55.90%; P = .234).

Table 3 presents the learning outcomes. At the one-year follow-up, the CG exhibited poorer retention for the following actions: pressing the emergency SOS button, calling 112, and turning on the speakerphone (P = .117). The lowest retention corresponded to providing a parent’s phone number.

In the IG (Table 3), the actions with the poorest retention were also pressing the emergency SOS button and turning on the speakerphone. In contrast, the outcome with the greatest increase was providing a parental phone number (E0, 56.10% vs E1: 67.40%; P =  .008).

We believe that our findings could be applicable to the fields of education and public health. The incorporation into the school curriculum of manipulatives and other interactive resources, such as the Endless Book and the low-cost phone simulator, could promote a more effective and sustainable approach to the teaching of basic life support skills.

These materials make it possible to tailor the content to the developmental stage of the students and facilitate understanding of essential concepts of the chain of survival from an early age. Furthermore, their low cost and ease of use make them ideal tools for integration into educational curricula, as they are more affordable. The integration of these resources can help raise future generations that are better prepared and more committed to providing early care in emergency situations.

Our study has certain limitations that should be taken into account.

The observed differences between the control and intervention groups could be attributed to the lack of randomization in the study. Schoolchildren were assigned to groups based on organizational and logistical criteria related to the program’s implementation. Priority was given to maximizing the number of participants exposed to the intervention while maintaining a sufficiently large control group for comparison. The study was conducted in a specific geographical area, in a particular type of center (state-subsidized private schools), and in a population with specific sociodemographic characteristics. The intervention was delivered by schoolteachers who had been trained beforehand using a standardized training and evaluation model; however, we cannot rule out the possibility of some variability in the implementation of the training program. Both the training and the evaluations were conducted in a controlled and secure simulation environment, a factor to consider when extrapolating the results to a real-world situation. The low-cost phone used in the simulation ran a specific operating system (iOS). Since the interface, navigation, and steps for making a call may differ across operating systems, the results obtained may not be directly applicable to devices running other systems, such as Android.