Provider perspectives of barriers and facilitators to the transition from pediatric to adult care: a qualitative descriptive study using the COM-B model of behaviour

Research objectives

This study aimed to explore current practices and experiences of transition from pediatric to adult care from the perspective of pediatric and adult health care providers. We addressed the following research objective: Identify barriers and facilitators to supporting youth and families for the transition from pediatric to adult care.

Design

We used a qualitative descriptive design [17] involving semi-structured interviews to develop a comprehensive understanding of health care providers’ perspectives on the process and barriers/facilitators to supporting youth and families for transition from pediatric to adult care. This study is part of a larger program of research aimed at understanding youth (aged 13+) and caregivers experiences with transition from pediatric to adult care with three conditions of focus: Inflammatory Bowel Disease, Diabetes, and Juvenile Idiopathic Arthritis. We selected these three conditions based on the existence of locally developed clinical databases which allowed us to identify youth with specialist-reported diagnoses. This aligns with a separate quantitative study, which is examining health system use for pediatric patients with these conditions using clinical databases and/or validated pediatric case definitions. In addition, these diseases are prevalent, chronic conditions in the local pediatric population, and have been reported by clinicians to have more linear transfer pathways than some other pediatric areas, such as oncology where children stay in pediatric care longer, or for mental health and addictions as there may be earlier transition of care. This study serves as a first step to understanding transfer processes in these specific disease groups, which will set the stage for future research focusing on patients with more complex care transitions.

Setting

This study was conducted in Nova Scotia, Canada, home to one tertiary pediatric health centre (herein after referred to as the ‘pediatric health institution’) that serves youth from three Maritime provinces. In Nova Scotia, most adult services are under a separate health authority that serves the rest of the province (herein after referred to as the ‘adult health institution’). We received ethics from the pediatric and adult health institutions Research Ethics Boards.

Integrated knowledge translation approach

This research project used an integrated knowledge translation (IKT) approach by engaging knowledge users as team members on the research project, which includes decision-makers, and health care provider representatives from the pediatric health institution, adult health institution, the provincial health department, as well as two patient partners. These members were involved in different phases of the research process based on interest and capacity. This included development of the research approach and interview guide, assistance with recruitment, data collection, data analysis, interpretation of the findings, knowledge translation (KT) planning, and general project management. Frequent check-in and update meetings were conducted with the core research team throughout the research project to ensure all members involved were well-informed and had a consistent opportunity to share feedback and perspectives related to different phases of the process.

Further, the co-lead of the research team is the coordinator for the Transition of Care Committee (TOCC) based at the pediatric health institution. This committee includes youth, caregivers, health care providers and community partners who have experience or interest in the transition and transfer of care process. Knowledge exchange meetings were held with the TOCC to facilitate two-way dialogue and feedback on different parts of the research project including interview guide development, study findings, recruitment ideas and KT planning. These meetings were co-facilitated by patient partners on the research team.

Theoretical framework

The implementation science literature, including the Medical Research Council [18], propose that interventions are more likely to be effective if theoretical models are used in intervention development. The Behaviour Change Wheel (BCW) is one such approach [19]. It is a systematic, theory-based guide to intervention design based on the principles of the COM-B model, which suggests that for any behaviour to occur there must be a change in one or more of the following: capability, opportunity and/or motivation (Table 1)The BCW has been used extensively to guide design of behaviour change interventions in health services [16, 19]. As such, we used this framework to guide the barriers and facilitators assessment.

Participants

We used a stratified purposive sampling strategy [20] approach with convenience sampling techniques [21] to recruit health care providers with experience caring for youth in transition in the past three years. We recruited through existing email lists and used snowball sampling techniques by which participants could disseminate information on the study to their networks. Based on participant preference, we conducted interviews on Zoom through video or audio. Interview participants included the following five health care provider groups who have treated youth with at least one of the three conditions of focus (Inflammatory Bowel Disease, Diabetes, and Juvenile Idiopathic Arthritis) in the past three years:

1. Pediatric physicians

2. Pediatric nurses and allied health professionals (e.g., psychologists, occupational therapists, social workers)

3. Adult care physicians

4. Adult care nurses and allied health professionals (e.g., psychologists, occupational therapists, social workers)

5. Primary care providers

We applied Guest et al.’s [22] qualitative sampling criteria (i.e., clarity of data, alignment with research question, variations and contraindications of data) to determine if additional interviews were needed following data analysis.

Materials

We developed a semi-structured interview guide based on previous guides sourced from the literature that focused on examining transitions from pediatric to adult care (see Appendix A, B, C) [23], [24], [25]. The interview guide was organized into components of the COM-B model of behaviour [19] with additional prompts added if clarification of questions was needed. A draft interview guide was reviewed by members of the TOCC, which included 7 youth, 2 caregivers, and 5 health care providers, to ensure questions were relevant and terminology was clear. The guide was refined and edited based on feedback received.

Procedure

Following initial contact by the prospective participant, the Research Assistant sent the consent form via email. The consent information was reviewed at the beginning of the interview and verbal consent to participate was obtained. The principal investigator (CEC), who has been trained in conducting interviews using the COM-B model [19] and designing interventions, facilitated the first three interviews using the theory-based interview guide. The Knowledge Translation Coordinator (JCK), also trained in the COM-B model [19], participated in these interviews and led subsequent interviews. Research team members (JP, AH, RW, LR) also participated in the interviews along with either CEC or JCK. At the end of the interview, the researcher collected contact information and demographic information related to gender, interdisciplinary role, and years of experience. All interviews were conducted via Zoom and audio-recorded using Zoom recording features. Interviews lasted approximately 30–45 minutes.

Data analysis

Audio-recordings for all interviews were transcribed verbatim and all information was de-identified. First, four reviewers (CEC, JCK, LR, AH) reviewed six transcripts (two reviewers each) using Microsoft Word. The primary reviewer used directed content analysis [26] and a color coding system to code similar statements into the three COM-B categories [19], while the second reviewer used the comment box to point out any discrepancies. A meeting was held with all four reviewers to go through each of the six interview transcripts and come to a consensus on any discrepancies identified. Further, preliminary sub-themes of participants’ specific beliefs within the initial coding scheme of the COM-B(19) were identified using inductive thematic analysis [16, 27]. A specific belief is a group of similar responses that suggest the belief may influence the target behaviour [16]. A codebook was developed based on the COM-B categories and affiliated sub-themes consisting of: code name, brief definition, full definition, when to use, when not to use, and an example [28]. All research team members reviewed the codebook, provided feedback, and the codebook was adapted accordingly.

The initial six transcripts along with the remaining twenty transcribed interviews were imported and coded in Nvivo Qualitative Software [29] by two reviewers (CEC, JCK) using the finalized codebook. The coded data were then inductively examined to further identify barriers and facilitators to transition from pediatric to adult care within each sub-theme. A final list of themes and affiliated summaries were developed and reviewed by the research team. Following this process, the initial findings were presented to the TOCC at the pediatric health institution and refined based on their feedback.

Capability

Our study participants described the need for specific knowledge and skills to support youth and their families in the transition process. Despite this, our study highlighted the lack of formal training that health care providers receive related to transition of care. This finding is consistent with previous research that has found physicians feel unprepared to care for youth transitioning to adult care [30, 31]. Limited training interventions or programs exist to support health care providers’ development of this unique skillset [32]. Our findings support improved training efforts, and in particular, interprofessional training with a range of health care providers to fully understand other roles and responsibilities in the transition of care process. Implementation science research has shown that increased knowledge does not always lead to behaviour change [19]. As such, solely relying on improved training of health care providers will not necessarily improve the transition process. Future training efforts need to be included in multi-component interventions that tackle other known barriers and facilitators.

Opportunity

Most of the identified barriers and facilitators fell within the opportunity component, which highlights that transition work is highly relational, as well as socially and systemically influenced. First, participants described the barriers and facilitators to effective communication with youth during the transition process. Providers described the value of a person-centred and non-judgemental approach to care; however, as described above, providers recognize the unique communication skills needed to support the youth in an emerging adulthood stage of life. From a qualitative study exploring patient-provider relationships across the transition from pediatric to adult diabetes care, Ritholz and colleagues highlight the need for pediatric providers to actively promote autonomy among youth to prepare them for the adult care setting [33]. The focus on autonomy is an important transition intervention consideration in addition to facilitators identified in our study, including person-centred care.

Second, health care provider participants also described their relationship with caregivers that occurs alongside interactions with youth. Other researchers have examined this role from the caregiver perspective and found that caregivers are important facilitators in transition planning, and need to change their behaviour to encourage this shift of responsibilities to their children [34, 35]. Our study builds on these findings as health care provider participants described their dual role in building trust to support youth and caregiver transition; care providers help youth gain independence, while simultaneously preparing caregivers to shift the responsibility of care.

Third, our study findings highlight the lack of intersection between pediatric and adult care providers during the transition process. The lack of communication among care providers is a significant gap that is known to lead to adverse events across the care continuum [36]. A significant evidence base exists for improving provider communication and youth transitions to ensure safe handoffs of patient information [36]. In the pediatric to adult transition of care context specifically, our study illustrates the value of transfer clinics, described as point of care meetings between adult and pediatric team that includes the youth. These transfer clinics are common in the diabetes care context, and have shown to improve youth and provider satisfaction with the transfer process [37]. Despite their value, transfer clinics focus solely on the transfer handoff and do not encompass the full transition process. There is an opportunity to build upon transfer clinics and redesign these interventions to focus on the entire transition process and ensure individualized planning, coordinated transfer, and secure attachment.

Our findings illustrate the value of designating a provider(s) to enhance care across the transition continuum. This type of role would be responsible for coordination across team members, facilitating the transition of responsibilities from caregiver to youth, and acting as a liaison across pediatric and adult health care systems. Some participants indicated the value of having someone on the health care team to take on these navigational responsibilities. Previous research evaluating patient navigator programs has shown an impact on system-level outcomes including reduced costs (hospital resource use and family out-of-pocket expenses), length of stay in hospital, and fewer delays in accessing adult care [38]. At an individual level, these interventions can improve knowledge and self-management skills, appointment attendance, and medication adherence [39]. Our research highlights an opportunity to explore the utility of a navigator program and the type of designated provider in these roles to help mitigate common challenges for the transition in care process. Our findings suggest an interdisciplinary approach may be more appropriate to leverage the skills, knowledge, and expertise of allied health professionals. A team from Alberta, Canada are currently conducting a pragmatic trial of a social worker in a patient navigator role to decrease emergency room utilisation in transition age youth with chronic conditions [40]. Results from this trial will be helpful to inform future development and implementation of navigator programs to support transitions in care.

Lastly, although our research did not specifically focus on the impact of the COVID-19 pandemic, participants described how the pandemic influences the transition in care process. Most commonly, providers shared how the pivot to virtual platforms catalyzed the development of innovative ways to communicate with youth and colleagues that could be sustained beyond the pandemic. Virtual platforms facilitated communication with youth regardless of geographic location, which helped to mitigate common barriers to in-person visits including travel costs and appointment times. This is consistent with the overwhelming adoption of new technologies during the pandemic to support patient care [41]. Specific to transition of care, Chandra et al. [42] evaluated the impact of a transition intervention for youth transitioning to adult transplant care and found increased participation related to multiple modes of interaction. Further, Agarwal et al. [43] introduced telemedicine during the COVID-19 pandemic for pediatric patients transitioning to adult diabetes care and found that a multidisciplinary team, including dieticians, educators, and psychologists, were able to offer improved care coordination by leveraging virtual resources. Moving forward, we can build on these lessons learned from the pandemic and incorporate multiple modes of communication to enable dialogue with youth.

Motivation & opportunity

The COM-B model suggests that opportunity and motivation are closely connected, with one influencing the other and vice versa [19]. This is evident with the social opportunity and motivation-related findings of our study: pediatric providers have a close relationship with the youth they treat and as a result, are motivated to ensure a successful transition. Pediatric provider participants described feelings of worry and concern for the youth they treat during the transition process and wanting to ensure youth avoid “falling through the cracks” in the adult care system. It is evident this level of concern motivates them to ensure a successful transition, including a secure attachment to the adult system. Contrarily, adult care providers do not yet have a relationship with the youth transitioning and may not be as motivated by feelings of concern and a sense of responsibility to facilitate the transition process. This is consistent with previous research with adult care providers and their experience taking on youth from pediatric care, demonstrating discrepancy among practices they think are important (e.g., reviewing patient pediatric records) and low frequency of occurrence in practice [44]. These discrepancies can cause deficiencies in transition care coordination [44]. Therefore, efforts are needed to leverage different motivators in the adult care system. Our study findings highlight that both pediatric and adult care providers are motivated by youth health outcomes. Fair et al. identified 10 transition outcomes through a Delphi process with an interdisciplinary group of participants, including individual outcomes (quality of life, understanding the characteristics of conditions and complications, knowledge of medication, self-management, adherence to medication, and understanding health insurance), health services outcomes (attending medical appointments, having a medical home, and avoidance of unnecessary hospitalization), and social outcomes (having a social network [45]. Audit and feedback interventions may be effective in this context, to gather local evidence on transition outcomes and provide feedback to providers as a motivator to support youth in the transition process. Integration of these transition outcomes data with patient experience data is critical to show the layers of impact of effective transition of care.

Implications

This study has important implications for intervention design, implementation, evaluation, and sustainability. The implementation science literature has shown that interventions are more effective when theory is used to identify determinants and tailor interventions to address these factors [15]. The COM-B model offered a clear framework for understanding the range of factors influencing transition of care. By describing barriers and facilitators in behavioural terms, this approach helps to design tailored interventions to address the known behavioural determinants of supporting transition of care. Future research efforts should move beyond a barriers and facilitators assessment and begin refining or designing new interventions to support transition of care that tap into the capability, opportunity, and motivational determinants of health care providers. Further, these findings can be used for a retrospective intervention analysis; we recommend using our findings to examine existing interventions and identify barriers that may be impeding their success.

Previous transition research has outlined three key foci for transitions of care: individualized planning, coordinated transfer, and secure attachment [2]. Our findings describe barriers and facilitators to realizing each of these important transition foci but extend the conversation to behavioural determinants that are influencing successful transition. The emphasis on an effective intervention design will be valuable for organizations, groups, and individual practitioners aimed at improving transition of care. In our local context, the Transition of Care Committee was a close research partner in our IKT research approach, and the goals and objectives of the research project align well with the committee’s mandate, as they are responsible for developing and implementing programs to support the transition from pediatric to adult care. The Transition of Care Committee can use these findings during their program development and evaluation to ensure the transition interventions address the perceived barriers and leverage the facilitators to transition of care. Other transition groups or health care providers may find similar utility in the study findings to design programs that address barriers and leverage facilitators in their local context.

Limitations

Our study findings should be considered with the following limitations in mind. Firstly, despite significant recruitment efforts, we struggled to recruit more primary care providers. We recognize the important role primary care providers play in helping youth to navigate their health care experience. As such, we recommend future research further examine the primary care provider perspective to identify similar or unique barriers and facilitators for this clinician group. Secondly, we recruited more pediatric care providers than adult care providers. This may be because transition of care is often viewed as a pediatric care concern; with the pediatric providers supporting the transition process. Although our adult care provider participants recognized this topic as a key concern, this study could be strengthened by incorporating more adult care provider perspectives to understand adult care-specific barriers and facilitators to supporting transition. Lastly, our research highlights the importance of building trusting relationships among health care providers, youth and caregivers. These findings need to be compared with youth and caregiver perspectives to ensure interventions are designed to address the needs of all parties involved.