This survey included 265 children aged 4–10 years and their parents with a 95% response rate. With Children Centered Care, our multi-faceted concept, 88% of children reported to feel comfortable during their unsedated MRI. This proportion was even higher than in the standard setup, where only selected children expected to succeed had unsedated MRI (77%), P = 0.02. Children liked to use the app and toy-scanner. Parents felt more prepared and secure during their child’s MRI with the Children Centered Care concept.

Competent care of children in healthcare requires specific approaches aimed at the relevant age group, especially for anxiety-provoking or painful procedures. To enable children to cooperate during unsedated MRI, the following elements should be considered when designing a setup: information, recognizability, motivation, child-friendly communication, and positive distraction [21, 22]. Individual children may benefit from some elements more than others. In some studies, the youngest children were found to benefit the most from various child-friendly features [23, 24].

The finding that 88% of children felt comfortable with the Children Centered Care concept is very satisfactory, as the study population included all children aged 4–10 years referred to an MRI in the intervention period, including inpatients, with few exclusion criteria and no further selection based on child age or maturity. Also, MRI was completed without GA or sedation by 98% of children. We believe that the multi-faceted approach is the basis for this, each of the four elements contributing to meet the child’s needs when placed in a demanding situation in a foreign environment, as discussed below. Likely, the interventions work in synergy to empower the child to successfully complete the scan and feel comfortable. The effect of individual elements cannot be evaluated from this study.

When planning the project, we only found a few other studies of children’s experiences with MRI, especially involving children aged 4–6 years. In this age group, the use of anesthesia is common [25,26,27], which makes the question of comfort during the scan irrelevant. Some studies use healthy volunteers assessing anticipated anxiety for an imagined or simulated MRI scan which may introduce substantial selection bias with results not applicable to a clinical setting [28, 29]. Others introduce an intervention and measure anxiety before and after the MRI, but as no control group is included, the effect of the intervention cannot be assessed [16, 19].

The app is central in the Children Centered Care concept and has several obvious advantages. Firstly, it delivers age-appropriate information essential for the preparation of the child. Also, it is used at home in familiar surroundings and without the time limits of a hospital schedule. The gamification elements motivate the child to explore the app in a familiar way known from leisure gaming. It connects individual elements of the Children Centered Care concept and introduces them to the child.

“Rumble” the troll character, who many children know from the pediatric department at our hospital, was used in the app and in physical wall decorations in the MRI section. Thereby, recognizability and continuity during the process of an MRI were secured, optimizing the app for local use. Using “Rumble” may have made the app less generic though, as other hospitals might have their own character/mascot. Also, “Rumble” might not suit the taste of all children world-wide. In the future, app developers could consider using exchangeable or more generic characters. The app was liked by 98% of the children who used it. Among enrolled children, 77% had received information of the app, and of these, 95% had used it. Thus, parents and children are well-motivated to use an app for at-home preparation. Ensuring that all families are informed of the app should be of high priority in the future.

Preparation in the children’s lounge was done in just 15 min, leaving the MRI scanner available to other patients. In practice, the toy-scanner acts as a great tool for the radiographer to connect with the child through play. Having used the app at home, children are familiar with the situation and seem empowered to take a lead at the toy-scanner, motivated to scan the four characters they know from the app to help them get a diagnosis. Parents may naturally withdraw a bit, leaving space for the radiographer to lead the dialogue with the child.

Using the toy-scanner, the need for an actual mock session is eliminated, which is otherwise a common approach in preparation of children for MRI [24, 25, 30]. Importantly, a mock session can be time-consuming and often requires an extra visit to the hospital. Preparation with the toy-scanner saves time for families as well as staff and simplifies patient flow. Morel et al. evaluated the impact of a teddy bear-scale mock MRI scanner on the anxiety level of 91 children aged 4–16 years using a visual analogue scale of stress-level from 0 to 100 [18]. They found that children’s anxiety levels were significantly lower after a mock MRI than before. This was not the case in the control group. Actual test values and reductions were not reported. Overall, these findings confirm that a mock session with a toy-scale scanner can reduce anxiety.

Nearly all children in the Children Centered Care group used the toy-scanner (90%). Thus, this form of preparation worked well in practice and was easier to implement than the app.

Highly motivated radiographers were specifically recruited for the Children Centered Care concept. With our training program, skills of communication and cooperation with children were further improved. By selecting a smaller group, individual radiographers continuously gained more experience and know-how. The size of the team must fit the desired capacity and be larger if an out-of-hours service is needed.

Our study indicates that trained radiographers may be as effective as child life specialists, with the advantages of not needing additional staff. In our experience, the radiographer’s approach towards the child is very important. Castro et al. inspected the effect of adding patient-centered communication to preparation with an MRI toy-scanner including 30 children aged 4–10 years in each group [31]. Decreased anxiety was found when the preparation included patient-centered communication, as measured on a 5-point facial image scale.

Establishing a pediatric team is affordable and readily accessible compared to the technical solutions. It is thus a great place to start, when working to improve the pediatric patient’s experience in MRI.

A hospital environment can be frightening and intimidating for a child, not least when it includes large, foreign-looking technical equipment such as an MRI scanner. Presenting the MRI scanner to the child in the app and children’s lounge is likely to already make it more familiar. We believe that child comfort was increased even further in the appealing atmosphere created with the ambient lighting, sound, and movie themes in the scanner room. During the scan, the movies create positive distraction, helping the child to stay comfortable and remain still. The light- and movie theme is selected by the child. Giving children choices in healthcare is considered important and a part of best practice [32].

Some recent studies evaluate the effect on children’s anxiety after adding an additional intervention to an existing multi-faceted, child-friendly setup. Results cannot be compared directly between different studies including ours, due to different anxiety measures.

In their randomized-controlled trial of 122 children aged 3–7 years, Fletcher et al. added a mock session to home-based preparation materials and training with a child life specialist and found overall similar PedsQL VAS scores between groups. However, lower self-reported fear and parent-reported sadness were found in the play-based setting using a mock scanner [33].

Ozdemir et al. evaluated the preparatory video made by McGlashan et al. together with child-friendly communication inspired by Raschle et al. in a randomized-controlled study of 66 children aged 4–15 years and found lower anxiety scores during MRI in the intervention group [16, 22, 34, 35]. Anxiety levels were measured comprehensively with the State-Trait Anxiety Inventory, the Modified Yale Pre-Operative Anxiety Scale, the Children’s Anxiety Meter-State, and the Children’s Fear Scale by children, parents, and professionals. They also found lower state anxiety levels of parents in the intervention group. Results cannot be directly compared to ours, but the findings are in consistency.

Geuens et al. included 82 children aged 4–10 years in a study evaluating a smartphone app with educational mini-games for use at home, in a setup including a child specialist and a movie in the MRI room [24]. Parent-reported anxiety scores were used and reduced anxiety after preparation with the app was found for 4–6-year-olds. Thus, findings confirm that a preparatory app can reduce pre-scan anxiety for young children. Furthermore, with the app, less than 5 min face-to-face preparation by a child specialist was needed, compared to the 30–60 min previously used. The potential for a preparatory app as part of a practically manageable, child-friendly setup was thus confirmed.

Various methods of assessing child comfort are found in the literature, of which some are validated for specific clinical settings or age groups not relevant for this study. The commonly used State-Trait Anxiety Inventory for Children is validated only from the age of 5 and was found too comprehensive for this context [14]. We preferred asking the children themselves, as we found the child’s perceived experience to be most relevant. Feeling comfortable remains subjective.

Visual analogue scales can be used directly by pre-school children. Using a whole-body figure allowed our illustrator to emphasize the different emotional states by using “Rumble’s” body language. It remains uncertain if the scale can be used across cultures.

Parental emotions before and during pediatric radiologic procedures affect the child [21]. Parents request thorough information before the scan to be able to support their child optimally [27, 36, 37]. Thus, parents’ needs should be met in a pediatric setup. In the Children Centered Care concept, all interventions were aimed mainly at children. Still, parental sense of security increased to a very high level. Thus, when seeing their child’s needs taken care of, psychological security is facilitated for parents. Meeting professional, child-friendly staff allows the parent to withdraw somewhat from the responsibility of the scan being successful. This creates space for the radiographer to lead the situation and interact with the child, in a manner known from the “One Voice” approach in pediatric nursing [38].

More parents felt very well prepared, likely via watching the app with their child. The children’s lounge was found child-friendly by almost all parents, while the waiting room of the standard setup was found “neutral” and “clinical.” By fulfilling their needs, the Children Centered Care concept optimizes parents’ impact on their child, as well as their own experience.

This study is large compared to other studies in the field and includes a standard group to compare with. Inpatients and acute patients were included and exclusion criteria were narrow.

However, differences between the Children Centered Care and standard groups are likely to have led to an underestimation of the effect of the Children Centered Care concept. In the standard group, only children expected to be capable of MRI without GA by the referring clinician were included. These were likely less anxious and/or more mature than children referred to GA (and thus not included in this study). Accordingly, the standard group included a smaller proportion of 4– and 5-year-olds, who are likely to be more anxious due to young age. Thus, the selected children in the standard group were more likely to feel comfortable during their MRI. Still, the proportion of comfortable children increased with Children Centered Care and to a very high level.

Randomization to MRI without GA in the standard setup would not have been ethically safe nor feasible. The effect of the Children Centered Care concept is likely to have had a larger effect on the younger children, but a subanalysis of children aged 4–6 would not be valid due to too few subjects.

Fewer children in the Children Centered Care group had a blood test or intravenous access on the scan day. Having a painful procedure before the scan may lead to greater anxiety during MRI. During the observations of children going through MRI (mounted camera sessions), it became evident that some children got an intravenous access before the exam “just in case” contrast agent was needed. In the Children Centered Care concept, meticulous care was taken to avoid any intravenous access if not strictly necessary. Also, blood tests were performed after the MRI, not before, as part of this strategy. This might have added somewhat to the effect of the concept, and should be considered in any child-friendly MRI setup. There was a trend towards a smaller proportion of inpatients in the Children Centered Care group. This may be due to logistic challenges of enrolling children with acute illness in the study. Children in severe pain may not be able to cooperate during MRI without GA. Also, an MRI tech from the pediatric team was not always available during out-of-hours service. For some acutely ill children, access to MRI without GA might actually have reduced the waiting time for their scan because fasting and anesthetic staff were not needed.

The range of MRI examinations done was comparable in the standard and Children Centered Care groups, as the scanned anatomical regions and the net scan time did not differ. We found no association between comfort level and scan duration. It seems that if the child is comfortable enough to enter the scanner and initiate the scan, the child is unlikely to become anxious later.

Informing all families of the app proved challenging, as only 77% knew about the app. If all children had received the full concept, the effect is likely to have been larger.

The high response rate of 95% is a strength of this study and was obtained by having the families answer the questionnaire while still in the department. However, the timing of the survey may have influenced the answers by both children and parents [39]. The perception of the experience may change over time. Answering while sitting in the department, and knowing the iPad should be handed back to the staff afterwards, could lead to more positive feedback out of politeness or a wish to stay on good terms with staff. As all data were collected this way though, results are comparable.

We previously found that with the Children Centered Care concept, the need for GA for MRI of children aged 4–6 was reduced to 5% [12]. The present study validates this concept even further. Thus, multi-faceted, child-friendly concepts for unsedated MRI should be strongly encouraged.

Our department is a secondary center not specialized in pediatric care. Future studies should assess the feasibility in tertiary centers, where higher patient flows create even more potential for improvements, while the patient population is more complex and may have special needs.