The study followed a qualitative research design (Hissong, Lape & Bailey 2015) to describe and explore the feasibility of the PicPecc app as perceived by South African children. Data were collected through semistructured interviews and children were given a choice to participate either online (e.g., Google Meet) or in a natural setting (in the child’s school or home context).

The PicPecc app is an mHealth tool that was developed to allow children to communicate for symptom relief and improved well-being during hospitalisation (Wiljén et al. 2022). The app is based on the principles of universal design, meaning that all content is explained and implemented with support functions such as text, images, and sound (Thunberg et al. 2022). Children have the right to actively take part in decisions regarding their health. However, they require support to communicate their symptoms and also to function in their everyday life. Such communication support largely relies on information, a variety of communication methods, and technology.

Purposive convenience sampling was used, based on the following selection criteria: age (7–18 years); language (English or Afrikaans proficiency). Ethics approval was obtained after which data collection procedures commenced once parental consent and child assent were obtained. Table 1 provides detailed biographical information of the 20 participants. The different first languages (Afrikaans [n = 10], English [n = 4]; Zulu [n = 4]; Sepedi [n = 1] and Sesotho [n = 1]), of the 20 participants in this study represented diverse languages and cultural backgrounds. The biographical information was obtained from a questionnaire completed by the parents.

Nineteen participants lived in the provinces of Gauteng and the Western Cape. Eight participants did not own a smartphone, but they had frequent access to either their mother’s (n = 7) or their father’s phone (n = 1). All participants used their smartphones for leisure activities while only four indicated that they also used it for schoolwork. Their devices ranged from lower to higher price ranges (phone costs range between R500.00 to R30 000.00). The phones on the lower end of the price range included Ag (n = 1), Kika (n = 1), and Hisense (n = 1), with mid-range devices included Motorola (n = 1) and Nokia (n = 1). Phones in the higher price range included Samsung (n = 6), iPhone (n = 2), and Huawei (n = 2). Four participants were unsure of the brand names of the phones their parents used. A mean time of 2 h per day was spent on the devices during the week with a range of 1 h – 8 h. On the weekend, participants spent an average of 5 h per day on the phone with a range of 1 h – 10 h per day. Fifteen participants reported that they have Wi-Fi access at home with five who did not have any Internet access.

The following materials and equipment were utilised for data collection: mobile device with data or Wi-Fi (mobile phone or tablet with either iOS or Android), PicPecc app, interview script, and procedural checklist. The interview schedule was developed based on the five dimensions of mHealth implementation proposed by Sieck et al. (2021a). Table 2 provides an overview of the development of the interview script based on the five dimensions to consider when implementing an mHealth tool. (Refer to Appendix 1 for the complete interview schedule.)

Participants were recruited from healthcare practitioners who were invited to refer potential participants. These healthcare practitioners consulted the parents of children from their private practices and asked parents to provide contact details if they consented that their children could participate in the study. The potential pool of participants was increased by also extending the invitation to colleagues and friends with children who met the selection criteria.

Parents were then contacted electronically and provided with an information letter stating the aim of the study and what would be expected of them and their children when downloading and using the PicPecc app. Parents were further asked to complete the informed consent letter and upon receiving this written parental consent, all children were approached and given written assent before data collection commenced. Thereafter, the children were provided with a username and a pin to login into the PicPecc app and were requested to use the app for at least 5 days. Finally, they were asked to indicate their preference regarding an online or an in-person interview at their preferred location.

The last author conducted all the interviews with participants and made field notes to increase personal reflexivity. A procedural checklist was used to prepare the interview and an interview script was used to guide the interviews. This ensured that the same procedure was followed with all participants thereby increasing the trustworthiness of results. The interviews lasted approximately 60 min – 80 min and were audio recorded.

The audio recordings of the interviews were transcribed verbatim by the last author. To ensure the trustworthiness of the transcriptions, the first author reviewed all the transcripts to check for correctness. All transcriptions were imported into ATLAS.ti – an electronic software programme for qualitative analysis (Friese 2019). Inductive thematic analysis was used to analyse qualitative content obtained in the interviews and to aid in holistically understanding the data through pattern recognition of the separate parts (Nowell et al. 2017).

During data analysis, an analytic process was followed to ensure the trustworthiness of the findings (Clarke & Braun 2013). The five phases of thematic analysis that were used include: (1) reviewing and rereading transcripts to increase familiarisation with data and obtain a deep understanding of the content; (2) searching for themes; (3) generating initial themes independently and reviewing the generated themes; (4) organising a meeting to compare and review themes and compile a codebook to increase consistency of coding; and finally (5) defining themes and linking these themes to the theoretical framework in a collaborative manner (Clarke & Braun 2013; Nowell et al. 2017; Roberts, Dowell & Nie 2019). A mutual agreement between the two coders (the first and the last author) was required for all steps of thematic analysis.

In this study, four components of trustworthiness were targeted, specifically credibility, transferability, conformability, and researcher bias. Credibility refers to data that truthfully represents the population of this study (Roberts et al. 2019). Credibility was increased through reviewing transcripts multiple times and by multiple authors and although collective data analysis to ensure that the findings were accurate and reflective of the participant’s experiences (Harper & Cole 2012). Transferability refers to the applicability of information representative of a larger population (Roberts et al. 2019). Thick and rich descriptions were provided in different settings. This assisted in increasing transferability, as the different settings may be more representative of a larger population. Conformability refers to the clarity of data collection procedures and documentation (Harper & Cole 2012). To maintain conformability in this study the authors made field notes and conducted a within-group discussion. The aim of these steps were to reduce researcher bias and possible assumptions that could influence the findings.

Ethical approval to conduct this study was obtained from the Research Committee of the Faculty of Humanities, University of Pretoria and the University of South Africa (UNISA) College of Education (No. HUM003/0219; 2023/03/08/90438574/02/AM).

All the participants mentioned that they liked the app, adding comments such as the ‘app is cool’ (P6; P7; P10; P13) or the ‘app is fun’ (P9; P11; P16; P18), ‘it is cute’ (P7; P8), ‘it is a good app’ (P1) and ‘I like everything’ (P13; P18). The participants further indicated that they enjoy the app features, particularly the avatar. Some participants mentioned that the avatar could have been created based on their personal preferences (P1; P2; P4; P13; P16) or ‘… I can choose their preferred facial expression and hairstyle’ (P4; P17). The participants specifically favoured the icons of pets and the adaptability of the pets (e.g., changing the pets’ names; P15, P16), which encouraged their frequent use of the app (P1; P4; P8; P20). In contrast, some children did not find the app appealing and engaging because of performing the same activity repetitively (‘commenting on how you feel over and over again’ [P5]; ‘little boring because it doesn’t have games to play’ [P11]). Despite these comments, the participants mentioned that this will not hinder their use of the app when they are sick.

Although many participants (P1, P2, P9; P10; P13, P15; P20) did not provide any recommendations to improve the app, some had innovative suggestions for app development. For example, participants suggested changing the background to include different scenes, such as at the beach (P11; P18); at home (P12), jungle (P18), a field with dogs (P11) or a friend’s house (P12) ‘to help the sick child to feel better’ (P4). Apart from adjusting the background to visual scenes, some participants suggested changing the background colour to their preference. Currently, the background is green in colour, but other colours were mentioned such as ‘yellow or lime green’ (P20); ‘pink’ (P7; P2; P19); ‘purple’ (P2; P19), blue (P2; P9; P18). These colour adaptations were also suggested for the hair colours of the avatar (at the time of data collection this option was limited to different shades of brown, black, reddish brown, pink and white or the option of no hair).

Generally, the participants enjoyed the app features, but recommended making the app more interactive especially referring to the avatars and pets (e.g., feeding, playing or dressing the avatar or pet [P4, P6, P8; P12; P14; P16, P17, P18, P19]). Furthermore, participants suggested adapting the app to allow them to access their favoured games (P5, P14, 17) and watch YouTube videos for entertainment (P11).

During the time of data collection, there were challenges with the downloading of the app, which resulted in technical glitches. Some participants complained that it was time-consuming to request a pin code to log into the app or that they had limited access to data and/or Wi-Fi (‘I do not have my own phone or access to data’ [P2]).

Participants had older model phones or phones in the lower price range, which affected the amount of storage available for downloading the PicPecc app as well as the compatibility of the device with the app (‘My device wasn’t compatible and I couldn’t download it from Play Store’ [P4]). Some participants could not download the app because of outdated software or found the spelling of the word ‘PicPecc’ difficult to remember. Adult assistance was required as mentioned by one participant who stated, ‘my mother had to help me’ (P2) when installing the app, although some mentioned that it was easy to install the app – ‘press install button’ (P7, P11). Even though downloading the app was generally considered to be easy, participants commented that the consistent loadshedding (the interruption of electricity supply during specific times of a day) and weak and/or slow Wi-Fi connection negatively impacted their use of the app (P8, P3).

Some participants commented that the fact that they needed a pin code to access the app made it slightly difficult to login: ‘It was hard due to the login details and the registration of your account and pin code’ (P18); ‘It [download] was easy, but the login was difficult because you needed the password and pin code’ (P17). Comments on the navigation of the app ranged from ‘First bit difficult – get used to it, easier’ (P1; P13); ‘learning how to use it’ (P15) to ‘easy’ (P2 – P9; P11; P17 – P19).

Participants mentioned that they would have loved it if the app could be used as a communication aid to communicate with somebody else in the hospital as a:

Participant P9 commented that one can ‘show the pictures to someone else to tell them that you are sick when you cannot talk’, and Participant P13 mentioned ‘It helps you when you not feel well I can just give it to my mom or brother’. Participant P12 also noticed, ‘This app can help me communicate and talk. If I am not in the mood to talk, I can just show the picture in the app’.

The participants’ socio-economic status impacted the use of the app, with comments such as ‘data’ or ‘Wi-Fi’ when asked about challenges or stating their phone’s incompatibility with the app (P2, P6, P8, P12 – P16, P19). Participants utilised the app for different reasons, for example, to report anxiety (P15) or to report when the child was feeling sick at home, at the hospital or at school (P14).

Participants shared ideas on how the app could be adapted for various contexts and different users. They particularly enjoyed the idea that symptoms such as emotions, pain and nausea could be expressed with a link to a diary, and how the use of statistics could reveal how their current feelings compared to their previous ones, for example, ‘You can see that you feel today better than yesterday, e.g., my headache is better – it decreased from an 8 to a 2’ (P1).

The participants agreed that the app can be used by sick children at home; in the hospital; and in school and that ‘they [children in hospital] can share their stories over and over’ (P3) or that they can share the app with others ‘I can show the app to tell my mom how I am feeling’ (P2, P4, P9; P13, P14).

Acceptability refers to participants’ perceptions of the app’s features, which affect motivation to use the app. Although most participants believed that the ‘app is cool’, this study highlighted the importance of interactive and gamification features in apps to maintain and increase children’s engagement with the app (e.g., children suggested features to dress the avatar or links to access games in the app). Gamification builds on a person’s instinct to play (Treiblmaier et al. 2018a, 2018b). The aim of gamification is, therefore, to use game-design elements in non-game contexts (such as healthcare) to boost a person’s extrinsic and intrinsic motivation by helping them to analyse information to achieve better goals, and/or to change their attitude and subsequent behaviour (Treiblmaier et al. 2018a). Gamification should introduce a novelty component and avoid unnecessary repetition to appeal to this age group. The children preferred features such as interactive avatars and pets, for example, feeding, playing or dressing the avatar or pet. Additional comments addressed interactivity by allowing children to access their favoured games and YouTube videos for entertainment. Children utilise YouTube to watch videos on topics such as humour, nursery rhymes, television shows, and animal videos (Neumann & Herodotou 2020b). Neumann and Herodotou (2020a) also evaluated the content of YouTube videos of interest to children and found that learning increased when educational videos included interactive elements. This interactivity element was confirmed by the work of Linder et al. (2021) who found that interactivity can help children improve their creativity, specifically when they are sick and are required to report their symptoms. Although the children in this study highlighted the importance of interactivity, they stated that regardless of this feature, they would still utilise the app to report their symptoms when they are sick. Previous research shows the need to design a meaningful experience to increase the value of gamification. The quality of gamification is more important than quantification (Auf et al. 2021).

Participants suggested that they preferred changes to the colour of the background, which is currently a shade of green. A variety of other colours such as yellow, lime green, pink, purple, and blue, were suggested. This may be related to the fact that colours have specific affective meanings. For example, Jonauskaite et al. (2019a) found that a lighter shade of yellow was associated with joy, yellow-green with relaxation, and that a darker yellow was perceived as a less pleasant colour. Although pink and blue have traditionally been linked to gender – blue for boys and pink for girls (Jonauskaite et al. 2019b) these authors found that both pink and blue are linked to positive emotions. Therefore, this may be the reason why participants in this study opted for these colours.

Participants also suggested that the background should be changed to specific visual scenes ‘to help the child to feel better’. Since the participants believed specific happy scenes that are unrelated to the medical event might help the children to feel better, it may also be related to the fact that these types of scenes are often linked to specific colours. For example, Jonauskaite et al. (2019a) found that imaginary scenes, such as the beach scene may be linked to the colour blue and a sunset to red.

Non-medical factors that influence health (social determinants of health) (Richardson et al. 2022) related to the availability and accessibility dimensions were mentioned in this study. The World Health Organization (https://www.who.int/health-topics/social-determinants-of-health#tab=tab_1) defined social determinants of health as the environmental conditions (i.e., the circumstances within the surroundings where individuals are born, reside, acquire knowledge, engage in employment, participate in recreational activities, practice religion, and grow older) that impact various aspects of children’s health, abilities, and overall well-being, as well as the likelihood of encountering risks. Social determinants of health are also responsible for health disparities as they are influenced by the dispensation of money, power, and resources (Richardson et al. 2022). For example, in this study, these disparities were observed as some participants had to borrow a phone from a parent while some owned their own, some had older phones that were incompatible with the PicPecc app, while some had the latest high-end phones. Some had limited data or experienced poor or no Wi-Fi access while some had access to uncapped, unshaped Wi-Fi. Another challenge was that nationally scheduled power cuts (loadshedding) negatively influenced participants’ Wi-Fi access resulting in long download times. Challenges because of loadshedding have also been reported by various authors who confirmed that the continuous loadshedding in South Africa has a huge impact on healthcare, including the use of mHealth tools (Githaiga et al. 2023; Malange 2023; Romski et al. 2023).

The availability and accessibility dimensions also highlighted the importance of acknowledging the participants’ digital literacy. Some participants also complained that they misspelled the name of the app when searching for it on Google Play Store or had to ask someone to help them search and download the app. They also believed it was time-consuming to request the pin code to log into the app. Therefore, Sieck et al. (2021b) mentioned that healthcare practitioners should be aware that some participants may need support on the use of digital tools, for example, how to create an email account or log into an app. These recommendations should be followed when the PicPecc app (or any other app) is introduced to patients in a clinical setting.

There was a difference in the frequency with which the five dimensions to consider regarding the implementation of mHealth were reported in the findings. For example, the ‘affordability’ dimension was mentioned the least. Previous research reported that the affordability dimension was more difficult to interpret than the other four dimensions (Adolfsson, Johnson & Nilsson 2017).

A possible limitation of this study is that the participants’ exposure to the app was relatively short, which might have influenced their perceptions of the feasibility of the app. Research has shown that the longer a person uses an app, the interest in the app decreases as the novelty wears off (Romski et al. 2023). Another limitation could include the socioeconomic inequality in South Africa as many South African children may not have access to smartphones. However, as found in this study, most children use their parents’ or relatives’ phones if they do not own one. Therefore, the findings may impact generalisation to the larger representation of South African children.

Suggestions for future research include replicating this study with parents and healthcare practitioners as they are often enablers for children to access mHealth. Furthermore, the PicPecc app could also be utilised as a reflective journal to express children’s emotions related to medical procedures.