Hospitalized children's experience of a Fairy Garden in Northern Thailand

Abstract Aim To explore through draw, observation and talk hospitalized children's experience of a Fairy Garden in Northern Thailand. Design A participatory visual arts method of draw, observations and talk along with Clandinin's narrative inquiry framework of three‐dimensional space (sociality, temporality and place). Methods This study was conducted in a paediatric ward in a large tertiary hospital in Northern Thailand over 12 months (July 2017–2018). A purposive sample of 17 hospitalized children aged between 4 and 13 years participated in drawings and face‐to‐face interviews. The study adhered to the COREQ checklist for qualitative research (see Supplementary File S1). Results The analysis resulted in five interrelated threads: (1) representations of the illness experience; (2) imagination and fantasy; (3) connection to place, family and home; (4) moments of social representations of play; and (5) happiness and enjoyment.


| INTRODUC TI ON
This study explored the experiences of sick children hospitalized in a Northern Thailand hospital who had access to and were able to take part in recreational activities in a natural and play area titled a Fairy Garden (FG) located adjacent to the two children's wards in the hospital. An increasing body of research indicates the value of green space such as gardens and recreational facilities to the overall health and well-being of hospitalized children (McCormick, 2017).
How children respond to the facilities of an FG and the impact of such a facility on their well-being as hospital patients can contribute to our knowledge in better catering for children in hospital.
These green spaces have sometimes been referred to as healing environments (Ananth, 2008), and an evaluation of healing gardens in a hospital setting for paediatric cancer patients indicates potential for this "restorative environment" to overcome the pressures of hospitalization for children (Sherman, Varni, Ulrich, & Malcarne, 2005).
Healing environments can be described in terms of both nonphysical and physical to facilitate the healing process impacting through psychological and spiritual dimension of health (Abbas & Ghazali, 2012).
Findings from the study by Sherman et al. (2005) concluded that while adults prefer to sit, socialize and walk about the garden, sick children will actively engage with facilities in the garden.
The focus of several studies has been on hearing the voice of children as they reveal their views of their experiences while in hospital (Brady, 2009;Corsano et al., 2012Corsano et al., , 2015Wilson, Megel, Enenbach, & Carlson, 2010). Wilson et al.'s (2010) study looked at the emotions of children through story telling of their hospitalized experience. Brady's (2009) study used draw and talk and addressed hospitalized children's views on what makes a good nurse. Corsano's et al. (2015) study looked at the emotional reaction to hospitalization, and Corsano's et al. (2012) study looked at the relationship with nurses and doctors.
The purpose of the study was not to investigate children's views of what makes a good nurse and their relationships with healthcare professionals but, instead, to explore the experiences of children hospitalized in a Northern Thailand hospital who were able to engage in recreational activities in a natural garden and play area titled a Fairy Garden (FG). The FG is a facility with gardens of colourful plants and flowers, short walks through the gardens, play equipment, a cubby house, bridge, gazebo, concrete blue whale, ceramic animals and seating for quiet times with parents or carers (van der Riet, Jitsacorn, Junlapeeya, Thursby, & Thursby, 2014). All hospitalized children had access to the FG throughout the day providing they did not have an infectious illness. The Fairy Garden was easily accessed from both wards so children could engage freely in activities during the day providing they were not receiving any scheduled treatments such as chemotherapy or blood transfusions. Parents, grandparents and nurses would accompany both small and any unwell children while some children entered the garden unaccompanied. Nursing staff were always nearby to monitor activity in the FG.

| BACKG ROUND (LITER ATURE )
The value of nature in urban design through the inclusion of green spaces to benefit human psychological and physical well-being (Hoyle, Hitchmough, & Jorgensen, 2017) has also been advocated in the healthcare industry through the provision of natural environments and outdoor spaces being incorporated into hospital design for the potential healing benefits afforded to patients (Dijkstra, Pieterse, & Pruyn, 2006;Geary, 2003;Horowitz, 2008;Huisman, Morales, Hoof, & Kort, 2012;Ingulli & Lindbloom, 2013;Morrison, 2011;Pasha, 2013;van der Riet, 2014;Walker, 2012;Zborowsky & Kreitzer, 2009). Abbas and Ghazali (2012) recommended design features to be included in newly built paediatric wards to include natural and therapeutic gardens, easy access to gardens, play areas and garden activities as best practice in supporting better outcomes for young patients. Whitehouse et al. (2001)

| DE S I G N
A narrative inquiry approach comprising of participant observation and participatory visual arts-based activity was used. Narrative inquiry (NI) is the study of experience understood narratively (Clandinin, Caine, Lessard, & Huber, 2016), that is through a representation of events (Clandinin, 2013). Experience in NI is a narratively created phenomena with NI viewed as both methodology and phenomenon (Clandinin, 2013). As our study involved narrative inquiry, the children's experience of the FG was the phenomena we were studying.

| Data collection
The study took place in a paediatric unit in a Northern Thailand regional hospital. The paediatric unit consisted of two medical and surgical wards with 70 inpatient beds.

| Participants
Participants were eligible for inclusion if they were children aged four years and over and deemed well enough to participate in the research activities. We selected the age of four as Irwin and Johnson (2005) have previously advised that in qualitative research with children as young as four years there can be discerning data.
Eligible participants and parents were approached by the nurses on the wards. Each of the 17 children who were approached provided assent to participate in the drawing activities. However, one child declined to draw. This child participated in the interview, so we still included him in the study.

| Procedure
A drawing station was established in the precinct of the paediatric ward with table and chairs. Drawing material was available for each child consisting of white paper and a variety of coloured pencils. The children chose which coloured pencils they would use in the drawing and were asked to draw their experiences of the FG. After their drawing, each child was interviewed about their drawings (Table 1).
Initially, any uncertainty as to the task was managed by friendly talk to place the children at ease and was conducted in Thai by the Thai researcher. Questions of the children included the following: had they done drawing before, did they like to draw and were they happy to make a drawing that told a story about their experience of the FG? Simple language was used to minimize any misunderstandings and to check if the children were comfortable in participating.
The emphasis was on the children telling their story through drawing so that they could present what was important to them and then to talk about their drawing to clarify their experience.
The interviews consisted of the Thai translator (who was part of the research team) and the principal researcher (who was not Thai) were conducted through the Thai translator. Interviews were translated simultaneously at the time of the interview and audio recorded, and each researcher made a running record that became a journal of the process. Recordings and researcher's notes were also later checked for accuracy. The principal researcher had considerable experience in conducting interviews in Thailand with Thai research colleagues. An introductory explanation of the study and the semi-structured questions were prepared in English and then translated to Thai by the Thai researcher. While the children were drawing, the researchers made observation notes using a simple

| Participant characteristics
Most of the children were from rural areas and regional provinces in Northern Thailand. Parents were mostly farmers or small business owners.

| Data analysis
Analysing the data was an iterative process. Our data included the Consisted of three stages: 1. In analysing the data, we drew up a 3. Further to analysing content of the drawings, observations (noted in our grid), conversations with the children along with our notes we used the three metaphorical dimensional space involving temporality sociality and place. (Clandinin, 2013) These three dimensions of inquiry are central to the conceptual framework of narrative inquiry (Clandinin, 2013) and helped shape the design of the study (Clandinin et al., 2016). Clandinin (2013) points out the importance of researchers attending in temporal ways as it takes the researcher to the past, present and future of the person's experience. As narrative inquiry researchers, we also needed to attend to the dimension of sociality involving personal conditions (hopes and dreams) and social conditions (cultural, institutional and familial) (Clandinin, 2013). We also needed to attend to the third dimension, that is place, with a focus on the physical concrete and the topographical boundaries of place where the inquiry and events took place (Clandinin et al., 2016). In the context of the study, place is the FG in a hospital setting between two paediatric wards.

| Ethics approval
This study was approved by the first author's University Ethics Committee and local approvals from the hospital. Informed written consent was provided by parents or grandparents who were the main carers. The study adhered to the Consolidated Criteria for Reporting Qualitative Research (Appendix S1) (Tong, Sainsbury, & Craig, 2007). Family members gave written consent freely, and children gave their verbal assent. Each participant was asked to select their own special name for the study (pseudonym). Names were chosen from a favourite movie (mostly female participants), a sports person (male participants) and for some a favourite alternative name to their own.

| RE SULTS
In analysing the children's drawings and their interviews, five in-

| Representations of the illness experience
Several drawings presented imagery that fell into more than one thread. The illness experience was presented in several drawings.
Tilly ( Figure 1)  The activity allowed her to concentrate on something she liked doing. Her drawing ( Figure 6) focused on specific objects and things in the FG that she would have actively engaged with during hospitalization. We were not actually aiming in this study to find that drawing would be therapeutic; however, in this instance drawing gave Diamond a voice and may well have had the potential to help her adjust to her illness and in personally focusing on a positive experience.
For Namcha, 12 years of age with diabetes type 1 the FG as place became a distractor where she could forget her illness and the clinical boundaries of the paediatric ward. This is confirmed in her conversation: When I sit on the swing in the FG I get a feeling that I am not being bothered about anything and the movement backwards and forwards helps me not feel bored.
However, we noted that she did not draw herself in her drawing, only those objects of specific interest such as the swings, wishing well, garden walk and cubby house. Namcha also told us: I love to sit near the wishing well and I sit there for 30 min.
In retrospect, we did wonder if she did make a wish (future desire) at the wishing well in the FG.

| Imagination and fantasy
Imagination and fantasy were apparent as a thread in several children's drawings. Through this thread, the dimensions of temporality, sociality and place are evident. Sociality is represented in the expression of feelings and desire to be in the present. Kim reported that she had not yet taken part in any activities in the garden, yet she placed herself in the drawing (see Figure 2)    In her drawing, there is magic in her wishes and desires and this is further confirmed in her interview. Grape reported that she did not play on the items in the garden as most were too small for her and also she felt too tired and preferred to walk in the garden or sit on one of the seats. When asked how she felt while in the garden, she said:

I feel like a child again!
There was a degree of fantasy in Diamond's drawing that was interwoven with a desire for play.
Diamond tells us that she loves art, and we note that she has drawn a sandcastle ( Figure 6) in her drawing. She recounted: I would like to have a sandcastle and to be able to play in the sand.
We wondered if this was a previous experience, she enjoyed and wanted to return to at that time or in the future. However, in Northern Thailand there are no beaches. She saw the FG as a place for physical activity, imaginative play and an opportunity to make and build a sandcastle. The Thai flag was placed in a top and bottom corner and on the cubby house. The garden, contained within a rectangular shape, has been added in an imaginative way to include a sandcastle, digging tools, bucket and toy castle. The colours create a bright environment through her choice of a wide range of colours to emphasize each item.
In her representation of the garden, it suggests a conscious desire to tell a story about the FG as a place of imaginative activity for her. The metaphorical dimensions of time (past, present, future), sociality (an inner desire) and place are captured in this drawing.

| Connections to place; familialfamily and home
This thread seems to be intertwined with imagination and fantasy of the previous thread, and there are strong representations of the dimensions: sociality, place and temporality.
For several of the children, there is a very close desire for connection to place and family, place being home. For Mickey (Figure 7 The dimension of sociality is evident in Mickey's drawing as he spoke of bringing his family into the experience of the FG and connection to family eased that he was in an unfamiliar place, other than home. Identification of place through play becomes an overarching sense of geographical location. We asked ourselves was this a desire for home as Mickey is an inpatient? He used the colour blue a lot in his drawing, and he told us that his grandfather liked blue. There is also a strong sense of cultural patriotism representing the dimension of sociality in several of the children's drawings by including the Thai flag. Importantly, in NI experience is continuous, relational and social (Clandinin, 2013). This is evident in this thread along with the temporality of the past represented in Mickey's drawing, along with place in the present and future for his grandfather to be with him in the FG.  In Minny's drawing, there were a series of items from the garden including cubby house, swing and small ceramic animals. Minny has drawn herself in the garden and next to her is another child. Each is represented with bright colours, expressive faces and hair in patterns to represent a boy and a girl.

| Happiness and enjoyment
The thread of happiness and enjoyment is probably the strongest of all the threads with acknowledgement to the dimension of sociality and place. All the children told us that the garden made them feel happy and joyful and that it was beautiful place to come and play as there were nice things to see and do. Many of the drawings with figures presented happy, smiling faces, and none of the drawings indicated negative emotional expressions such as frowns, tears or downturned mouths. In Kim's drawing (Figure 2), there was a lovely smiling sun above the gazebo, and Minny ( Figure 9) had a smiling sun looking over the garden. Minny stated that she liked drawing herself in the garden and that she liked to sit on the whale, the swing and the slide. Even Tilly, despite her life-limiting illness of a neuroblastoma, was still full of energy and appeared from our observations happy engaging in the FG.
Happiness is evident in the following comments from participants and is interwoven with the previous thread of play, subsequently signifying the dimension of sociality: The FG gives me beautiful feeling and I am happy playing in the garden with the other children There is a strong personal link here to the place of the FG in her experience and involvement with the garden and a sense of happiness to have such a place for relief and distraction from the clinical environment ( Figure 10).
During the interview, Pippy had an intravenous cannula in her right hand. She still wanted to draw even though we offered her not to do a drawing if this created discomfort for her. We note that she did not draw herself with a cannula in her hand. In her drawing, Pippy sitting on the swing had a happy face with decorative eyes embellished by prominent eye lashes and attractive hair style.
The structure of the swing has a decorative pattern of rectangular shapes, and the figure and objects have been carefully coloured ( Figure 11).

F I G U R E 9
Minny's drawing F I G U R E 1 0 Kitty's drawing healing gardens in hospital settings to enhance health by promoting outdoor play. Ozcan (2006) argues that paediatric patients are more sensitive than adult patients to the environment and, therefore, the impact of healing is more significant when there is engagement with healing environments in the hospital.

| D ISCUSS I ON
Best practice in paediatric wards supports the creation of healing environments that include opportunities for play and accessibility to outdoor therapeutic gardens (Abbas & Ghazali, 2012). The thread of happiness and imagination is consistent with other studies on the FG with the nurses and also student nurses reporting that they observed happiness and there was a sense of imagination and play (van der Riet et al., 2014;Riet, Jitsacorn, Junlapeeya, Thursby, & Thursby, 2017a;Riet, Jitsacorn, Junlapeeya, Thursby, & Thursby, 2017b). In previous studies, play in the garden was seen as a distractor from illness (van der Riet et al., 2014) and this was consistent with what we found. In particular, the FG offered an escape from the boredom of the clinical environment. The children in this study all reported in the interview the FG made them feel happy and gave them joy.
Attention to relationship building was important in this research for several reasons. Firstly, it is important as we were researching with a vulnerable population and several of the children were doubly vulnerable as they had a terminal diagnosis (glioblastoma and neuroblastoma). We were conscious that the children with cancer are a doubly vulnerable cohort due to their very young age and life-limiting and threatening illness (Boles, 2018). Boles and Winsor (2019) have emphasized the importance of building rapport with children and parents. Furthermore, Huang et al. (2014) emphasized the importance of developing child-friendly strategies in building rapport with children to ensure a trustful and equitable relationship exists.
Unbeknown to the children participating in the study, the researchers gathered together a variety of small gifts for the children following completion of their interview. Examples of gifts included a small tea set, comic books, dinosaurs and drawing material. Each child chooses a gift. As researchers, this was a way of honouring and thanking them for their participation. We praised the children on their drawings.

| Limitations and strengths
A limitation of this study could well be that that we really did not look in detail at the development levels of the children's drawings, that is the mental schemas that show representation of images. We have not reported in detail any gender differences in the children's drawings, and this could be seen as another limitation of the study.
Our strengths, however, are that we were flexible in our approach to our research and acknowledge data collection methods need to be flexible in researching with children (Boyles, 2018). For example, we were accommodating in enabling each child to take time when drawing. We allowed Tilly to freely come and go during the draw and talk processes, and we were encouraging in enabling the children's voices to be heard.
Another important strength was our method of using drawing to communicate the children's lived experience of the FG.
Drawings provided possibilities of how children perceive and communicate their experiences of their world (Carter & Ford, 2013).
In drawing, children are able to engage and express their views and imaginations (Driessnack, 2006). This point is further explained from The United Nations Convention on the Rights of the Child 1989/1990 UNCRC in article 12-13 which points out that children have a right to express their experience through forms of representation such as text, spoken words and visual methods (Khoja, 2016). There are also studies that have concluded that drawings give children access to a voice in telling their stories that F I G U R E 11 Pippy's drawing they might otherwise have difficulty in presenting (Brady, 2009;Pipe, Salmon, & Priestley, 2002;Weinle, 2002;Wesson & Salmon, 2001).
It has also been asserted that drawing may promote children's agency (Greene & Hogan, 2005;Mannion, 2007). The opportunity for young children to present their stories places the emphasis on the child as agent of their experience and have their experience and be honoured. There is also recent literature that points out the importance of having a child's perspective rather than taking the child's perspective (Bryan, Bluebond-Langner, Kelly, Kumpunen, & Oulton, 2018;Coyne & Harder, 2011;Nilsson et al., 2015). In the co-construction process of research, Lim and Lim (2013, p. 65) remind us that "both adult and child are equal players and the resulting dialogical process plays a major role in the constitution of the phenomena." Boles (2018) argues for a culturally appropriate and child-centred approach to research that captures both verbal and non-verbal data.
More recent work by Boles & Windsor (2019) support data collection that is child-centred and directed by the child. Furthermore, in our study we chose draw, observation and talk as a method to capture both verbal and non-verbal data. In essence, the method of drawing a child's experience can be seen as data directed by the child.
Another strength was that we engaged in a high level of reflexivity. For example, straight after the interviews with the children both researchers met to discuss and clarify any tensions or concerns. We were very mindful in our approach to our observations, the interviews and analysis of the data. Having raised the issue of reflexivity, we now address this as a means of ensuring trustworthiness and rigour in our qualitative study.

| Rigour and trustworthiness
We would argue that in establishing trustworthiness and rigour in our study our strong focus on reflexivity ensured our research demonstrated rigour and trustworthiness. Reflexivity has certainly become a critical topic in qualitative research (Shaw, 2016) and an elusive term (van der Riet, 2012) that requires the researcher to operate on multiple levels (Etherington, 2004) through self-awareness and ensuring there is methodological cohesion. For example, ensuring the research question and aim match the methodology (van der Riet, 2012). In our study, the research aim involved experiences of a FG and this was clearly cognizant with a NI approach that focuses on the phenomena of experience. We were mindful that the stories that we told through the children's drawing were both our stories from the participants and the researchers. This is consistent with narrative inquiry as it is relational methodology.
Trustworthiness of the data is also linked to credibility, dependability, confirmability and transferability (Guba & Lincoln, 1994). In relation to dependability and credibility, we kept a very careful audit trail in keeping journal notes, meeting with each other straight after the interview to discuss any differences and confirm our understandings. This procedure was important, especially as the main author was not from the Thai culture. Credibility is achieved through triangulation of the data, and we achieved this with the transcripts, drawings, field notes and our reflective journals. Our triangulation of data assisted in ensuring data saturation.
Generalizability has become a controversial topic (Carminati, 2018). We would argue that generalizability is at odds with narrative inquiry methodology and fits more with a positivistic paradigm of quantitative research. Narrative inquiry researchers attend to participants' stories as they are composed over time and in relation to people and situations at various different places and social settings so there is less likelihood of generalizations and certainties (Clandinin, 2013) in NI studies such as ours.

| CON CLUS ION
Through the physical representation of drawings and conversations with the children, we saw in this study a different world, one where there is an escape from the medical representation and boundaries of hospitalization. The FG brought a community/play environment that counters the biomedical and clinical environment. We would argue that healing environments and green space such as the FG have had a positive experience for the children in this study and provide a more humanized approach to the care of hospitalized children.
Globally, a top priority for healthcare services now in the area of paediatric services is to build more natural landscapes for hospitalized children to engage in play, lessen stress and escape the boredom of clinical unwelcoming spaces.

| RELE VAN CE TO CLINI C AL PR AC TI CE
Creating therapeutic environments for sick children benefits their hospital experience and provides more humanized care in helping to provide an escape from the clinical environment and assuage the stress of medical procedures.

ACK N OWLED G EM ENTS
We would like to thank the participants for their participation in the study.

CO N FLI C T O F I NTE R E S T
We have no conflicts of interest to declare.

AUTH O R CO NTR I B UTI O N S
PV, PT: Data design. PV, CJ: Data collection. PV, PT, CJ: Data analysis.