Introduction
Since 2020, the coronavirus disease 2019 (COVID-19) pandemic has had an immense impact on healthcare, leading to a worldwide increase in the use of telemedicine for online consultation [
1]. Telemedicine is the use of information communication technology (ICT) to deliver adequate remote healthcare and is a proper manner to deliver outpatient clinic care at a distance, particularly appropriate during a pandemic [
2]. The Dutch State Institute for Public Health and Environment (RIVM) has noted a decline of 38.0% in hospital visits in 2020; approximately 12.0% of those postponed hospital appointments were transferred to online consultation [
3]. Online consultation, via telemedicine, is an alternative, also in non-COVID-19 times, for paediatric endocrinologists to provide timely and efficient consultation [
4,
5]. Furthermore, earlier studies showed that transitioning outpatient care to the home setting lowers the burden of disease for children [
6]. We anticipate that the integration of telemedicine into paediatric clinical practice will increase, accelerated by the COVID-19 pandemic [
7,
8]. Insight into the facilitators and barriers to the transition to home-based check-ups (HBCU) via video consultation is vital for its successful implementation and adoption. If the hindering factors are known, they can be corrected before the actual implementation and focus can be maintained on the facilitating factors, to encourage all stakeholders involved in the care process.
In the Netherlands, approximately 1/10,000 children annually start growth hormone (GH) treatment. This is in accordance with previous reports (1/4000–1/10,000) [
9‐
11]. Several paediatric diseases, including but not limited to GH deficiency and resistance, result in short stature [
12]. The primary goal of this treatment is to promote height velocity and to improve final height close to the patient’s target height. Reliable height and weight measurements are an essential part of outpatient clinic appointments because estimating the body surface area (BSA) is one of the cornerstones of GH dosing [
13]. GH treatment requires adequate monitoring by a paediatric endocrinologist, including four appointments annually to adapt the GH dosage. In addition, attention is paid to general well-being and compliance, for which consultation at the outpatient clinic is not per se needed. These outpatient clinic visits are time-consuming for parents as well as for healthcare professionals. Whenever the height measurements of these children can be measured accurately in the home setting, at least part of the consultation can be transitioned to the home setting. In our recently published study, we found that parentally performed height measurements (in the hospital setting) strongly correlate to height measurements performed by outpatient clinic nurses, indicating that caregivers are able to accurately measure the height of their child at home [
14]. Furthermore, earlier research in the field of obesity in children also revealed a strong correlation between parentally reported height measurements and measurements performed by observers [
15].
Transitioning outpatient clinic visits for paediatric GH to HBCU is innovative and promising for future healthcare sustainability. Healthcare innovations may offer a solution for the increased workload in hospitals and cost-effectiveness when compared to physical appointments. In addition, transitioning outpatient clinic visits may have a positive impact on patients and families as it saves time, travelling, and money [
16,
17]. This study aims to investigate the facilitators and barriers for relevant stakeholders towards transferring outpatient clinic visits (treatment as usual (TAU)) to HBCU in the future, for children treated with.
Methods
Study design
A mixed-methods study using a sequential design was performed at the Amalia Children’s Hospital, Radboud University Medical Centre (Radboudumc), a teaching hospital in Nijmegen, The Netherlands. The study consisted of a quantitative part (questionnaires) and a qualitative part (semi-structured and focus group interviews) and triangulation of data sources was used to get richer and fuller information about the research topic [
18,
19]. The descriptive nature of this study facilitated its ability to identify barriers and facilitators [
20]. Data was collected by the researcher AR (during the period of data collection a master’s student in Healthcare Policy, Innovation & Management at Maastricht University, and a first-year master’s student in Medicine at Radboud University, with interests in researching the transition to HBCU) and stored in Castor EDC. The researcher had no conflict of interest and had no prior treatment relationship with the participants. We followed The COnsolidated criteria for REporting Qualitative research (COREQ) and Good Reporting of A Mixed Methods Study (GRAMMS) checklists for reporting (Supplementary materials Tables
1 and
2) [
21,
22].
Participants
For the study, two groups of participants were composed. The first group (group A) consisted of healthcare professionals (HCPs)—physicians, outpatient clinic nurses (OCNs), GH instruction nurses, and the departmental management of the Amalia Children’s Hospital, Radboud University Medical Centre; and the second group (group B) consisted of children being treated with GH and their parents or caregivers. Inclusion criteria for both groups were being involved in GH treatment, and the ability to speak and read Dutch. The participants from group A were recruited in person from the outpatient clinic of the hospital or via email. In addition, two paediatric endocrinologists from a nearby hospital (Canisius Wilhemina Ziekenhuis), participated in this study. Participants from group B were all recruited in person during their outpatient clinic visits. Children under the age of 12 participated in the study together with their parents or caregivers, and children over 12 were allowed to choose whether they wanted to participate together with their parents or caregivers or alone (hereafter referred to as children & parents/caregivers). Informed consent was obtained once for the study, for group A either via email, oral agreement, or through a written informed consent form. In addition, written informed consent from the participants from group B was retrieved. Before the start of the study, all participants were asked whether they wanted to participate in the semi-structured interviews. Their decision was recorded through the informed consent form and questionnaire. Data collection took place in May and June 2021.
Definition of HBCU
HBCU are comprised of several elements, to provide adequate home-based care for GH treatment. Children and their parents or caregivers receive instructions from an OCN about accurately measuring height and weight. The results of the height and weight measurements will be uploaded online before the online consultation and will be discussed with the physician during the online consultation. The content of the HBCU was similar to TAU, with the exception of physical examination including pubertal characteristics, blood tests, and skeletal age determination. Even while utilising HBCU, the child will continue to visit the physical hospital at least once a year, for blood tests, control of pubertal characteristics, and determination of the skeletal age. Checking these indicators via online consultation is impossible.
Measurements
Quantitative part (questionnaires)
The evidence-based Measurement Instrument for Determinants of Innovations (MIDI) was used to investigate the facilitators and barriers to the transition of TAU to HBCU [
23,
24]. The MIDI questionnaire is designed to improve the understanding of important aspects that may affect the implementation of healthcare innovations and consists of four domains: the innovation-, the user-, the organisation-, and the socio-political scale. Not all determinants of the questionnaire were suitable for the current phase of the HBCU; therefore, the researchers critically examined which questions were suitable for the current phase appropriate for the stakeholders (Supplementary materials Table
3).
The responses on the MIDI questionnaire were based on the Likert scale and ranged from totally disagree (1) to totally agree (5). Furthermore, two open questions were added to the questionnaire to gain more in-depth information about the personal (dis)advantages (determinant 8 of the MIDI questionnaire) to HBCU. No specific (dis)advantages were tested, to give the responders space to share their perspectives on these matters, and the following open-ended question was added: “If yes, which (dis)advantages?” [
25]. The answers given in the open-ended questions were supplementary to the facilitators and barriers identified from the questionnaire of interviews. In addition, participants were asked to assess an overall grade (range 1–10) for HBCU. After the questionnaires were composed, they were discussed with a representative from the group, and minor adjustments were made. The participants had the option of completing the questionnaire on paper, online, or orally with the researcher.
Qualitative part (semi-structured and focus group interviews)
Semi-structured and focus group interviews were conducted to gain more in-depth information (supplementary to the questionnaires). For the semi-structured interviews, relevant stakeholders (from groups A and B) were identified via purposive sampling, and participants were included until data saturation was reached [
26,
27]. Data saturation was attained when, in at least three consecutive interviews, no new information was revealed [
28]. Topic lists were established before the interviews, based on the determinants of the MIDI questionnaire, to provide structure to the interviews. To represent the opinion of the children and their caregivers, the Children’s Advisory Board (CAB) and the Parents’ Advisory Board (PAB) of Amalia Children’s Hospital were included in the focus groups’ interviews. The CAB includes children undergoing treatment at Amalia Children’s Hospital and the PAB consists of parents of children being treated at the hospital. The participants from these groups are experienced in receiving hospital care and are trained to actively contribute to decisions within the hospital. The content of the interviews was tested with a representative of the focus groups. All interviews were conducted by the female researcher (AR). Field notes were taken by the researcher during the interviews.
Data analysis
Quantitative part (questionnaires)
The Statistical Package for the Social Sciences (SPSS) 25.0 was utilised for the statistical analysis. Descriptive analyses (mean, standard deviation, and percentages) were computed to evaluate the facilitators and barriers to the transition from TAU to HBCU. We tested the internal consistency (Cronbach’s alpha) of our questionnaire to assess the reliability of the questionnaires [
29]. Determinants for which ≥ 20% of the total number of responses answered “strongly disagree” or “disagree” were considered barriers and determinants for which ≥ 80% of the total number of responses answered “strongly agree” or “agree” were considered facilitators [
30]. For this study, facilitators noted in both groups and barriers reported in one of the groups were discussed as facilitators and barriers to the transition from TAU to HBCU. The definitions of facilitators and barriers were based on the study of Bach-Mortensen and Verboom [
31]. Factors were considered as facilitators if these promote the implementation or adoption of innovations and/or transitions in healthcare. Factors were considered as barriers if these factors impede the implementation of innovations and/or transitions in healthcare. The mean of the total grade (1–10) was calculated for the groups. Moreover, an inductive approach was used for analysing open-ended questions about personal advantages and disadvantages. Topics (personal advantages and disadvantages) noted at least two times by the children/parents or HCPs were determined as facilitators and barriers for HBCU.
Qualitative part (semi-structured interviews and focus group interviews)
The steps of the reflective thematic approach were used for the analysis [
32,
33]. The first step of the reflexive thematic analysis was familiarisation with the data. The interviews were conducted by the female researcher (AR). In addition, field notes were taken by the researcher during the interviews to get familiarised with the data. The interviews were manually transcribed by the researcher (AR) in MS Word. After member-checked approval, the transcripts of the interviews were coded by the same researcher via Atlas.Ti.20. The researcher reviewed the acquired data critically, subsequently followed by the second step, coding of the data. A combined approach was used for the coding process [
34]. The data derived from the interviews was supplementary to the data of the questionnaire and therefore was the MIDI questionnaire a starting point for the data analysis. Deductive codes were assigned based on the MIDI questionnaire [
24]. In addition, inductive coding was applied to explore the data on important domains not identified in the deductive phase. The codes contained more than one word, so it is clearly understood what they mean as well. Each coded segment was iteratively read and coded. The third step of the reflexive thematic analysis was generating initial themes. This was done by critically assessing the retrieved codes, central concepts were sought under which the various codes could fall. This was a first step in identifying themes. The next step of the reflexive thematic analysis was reviewing and developing themes (fourth step). The themes were further adapted and developed by looking again at the coding and the data. This was also discussed with PvS. If necessary, certain themes that corresponded were merged, new themes were added, and duplications in the codes were removed. The fifth step of the reflexive thematic analysis was refining, defining, and naming the themes. The themes were divided into final themes and given descriptions and names. Based on this, a code tree was created (see Supplementary materials Fig.
1). The last (sixth) step of the reflexive thematic analysis was producing the report. Based on these themes (along with the previously collected data from the questionnaires), the data was represented in story form, and quotes that were supportive of the story were added to the text. The quotations derived from the analysis were translated directly from Dutch to English.
Ethical considerations
Children and parents were informed about the study’s goals and procedures, after they were asked for written informed consent to collect and analyse data. The procedures followed were in accordance with the World Medical Association Declaration of Helsinki. The Medical Research Ethics Committee of Nijmegen determined that this study did not fall within the remit of the Dutch “Medical Research Involving Human Subjects Act” (No. 2021–7506). In addition, the study was approved by the FHML-REC-commission (file no. FHML-REC/2021/026/HPIM.077) of Maastricht University.
Discussion
This is the first mixed-methods study to examine the facilitators and barriers to the transition from TAU to HBCU for children being treated with GH. Both HCPs and children (including parents/caregivers) reported predominantly facilitators. Many of the facilitators were mentioned in the user scale: self-efficacy, convenience, potentially increased accuracy in height measurements, social support, client/patient satisfaction and cooperation, patient-centred care, the flexibility of HBCU, and a physical start-up period. In addition, non-complex cases were perceived as being more suitable for HBCU than complex cases and therefore HBCU must be tailored to the child’s needs. The remaining facilitators (in the other scales) were related to the procedure and a potential decrease in healthcare costs. Although several barriers were recognised: compatibility with current practice, increased workload for the staff, insufficient ICT network, and the unsettledness of the organisation. In addition, two tested outcome expectations may be perceived as barriers; however, in the context of this study, these factors may even be facilitators. The non-complexity of the condition of many of the children ensures that HBCU suit the (in general non-complex) target group. Overall, this study showed that all stakeholders were highly positive about HBCU, and several relevant comments were made in relation to the implementation of HBCU.
The reported facilitators and barriers are in line with existing evidence regarding the facilitators and barriers to the transition from TAU to HBCU. The study of van Wijngaart et al. [
32] notes several facilitators and barriers to the implementation of eHealth innovations in daily practice. The convenience, compatibility, physicians feeling less in control, and a malfunctioning ICT network are the main corresponding facilitators and barriers between our study and the study of van Wijngaart et al. [
32], although the political context was requested more predominantly in this study and therefore divergent from our study. Furthermore, the convenience of HBCU (reduced waiting time and being user-friendly), along with its flexibility and being patient-centred, is reported facilitators in other studies and is in accord with our findings [
33‐
35]. In addition, earlier studies have suggested that parents can perform measurements and online consultations at home [
33,
36]. These factors align with the facilitators of self-efficacy and the possibility of performing accurate measurements at home recognised in our study. Insufficient planning, the possibility of missing medical aspects (concerns about lack of physical examination), and inadequate ICT facilities were the perceived barriers of HBCU which are in accordance with previous research [
37]. The unsettledness of the organisation was reported in other studies to be a barrier to the implementation of innovations (as HBCU); however, as suggested, the unsettledness may be in line with HBCU [
38].
Insight into the facilitators and barriers is of fundamental basis for implementing healthcare innovations (as HBCU). If these factors, especially barriers, are considered before the implementation and adoption, they may positively influence the outcome of the implementation. Flexibility and adaptation to the patient’s stage of life, treatment, and their needs and wishes are crucial factors in determining whether to implement the transition from TAU to HBCU. Pubertal characteristics for example cannot be investigated via online consultation. In addition, children with more complex issues (for example children being treated with GH because of panhypopituitarism) may not be the most suitable target group for HBCU. Accordingly, HBCU should not become the standard for all children being treated with GH. However, HBCU may work perfectly for children being treated primarily with GH (for example children with an isolated GH deficiency or SGA). Regarding the barriers, investing in good ICT facilities and efficient staff planning will advance the implementation of HBCU. Furthermore, it is important to take into consideration that HBCU differ substantially from TAU during the implementation and transition.
Our study has five strengths. First, data triangulation was used, via multiple (both quantitative and qualitative) sources, which improved the credibility of our findings [
39]. Second, a prospective research design was chosen to investigate the research topic, which provided more information about our research topic [
40]. Especially since COVID-19 played a pivotal role in the implementation and adoption of telemedicine innovations in healthcare. Third, a validated questionnaire was the foundation of the study. The MIDI questionnaire is widely investigated and used for identifying factors that may affect healthcare innovations and its implementation and therefore this questionnaire was chosen as the foundation of our study [
24]. Fourth, the diversity of the different stakeholders participating in the study represented the various groups involved towards the transition to TAU and resulted in a multidisciplinary approach of the research topic. Fifth, the participants in this study covered a wide range of diverse backgrounds. Especially, the various medical backgrounds of the children participating demonstrated the difference between the non-complex and complex conditions of GH-treated children. As such, the non-complex group may be more suitable for HBCU. Despite these various strengths, five limitations were recognised in our study. First, several determinants of the questionnaire approached the cut-off values set for facilitators and barriers. These items did not reach the classification threshold for barriers or facilitators, and this may influence the implementation and adoption of HBCU. However, the qualitative approach of our study addresses this limitation. Second, presumably because several items of the questionnaire were deleted, the reliability score for one group was lower. Third, approximately 10% of the total population of children treated with GH at Radboud University Medical Centre participated in this study; this could be a potential threat to the generalisability of our study. We tried to circumvent this by including GH-treated children with various causative medical conditions. Fourth, purposive sampling was used to select participants for the semi-structured interviews, which may have resulted in a selection bias. However, we assume that the likelihood of selection bias is limited because the data gathered in the interviews was in line with the data from the questionnaires. Fifth, we did not ask our participants about their previous experiences with telemedicine. This could be a potential threat to the validity of our study.
Insight into the facilitators and barriers to the successful implementation and adoption of HBCU is crucial. HBCU might yield considerable benefits for both patients and HCPs in terms of convenience, delivering care tailored to patient needs, possibly more accurate height measurements, providing even more information to HCPs in comparison with TAU, and cost savings. HBCU should be flexible to the different stages of treatment, stages of life, and the wishes and needs of both children and parents. Whenever height and weight measurements at home prove reliable, we do think that our findings are transferable to other paediatric patient groups, at least partially (for example children with celiac disease, metabolic diseases). Notwithstanding the benefits of HBCU, several barriers are noted, which should be considered and actively monitored while implementing HBCU. It might be valuable for further research to investigate the first experiences with HBCU and to fine-tune the process during the implementation phase of HBCU. In conclusion, this study revealed the potential of HBCU for future healthcare.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.