Abstract
Introduction
Effective care coordination is critical to manage unpredictable complications of conditions such as pediatric inflammatory bowel disease (IBD) that have a relapsing and remitting course. Our objective was to explore perspectives of care coordination following emergency department (ED) visits by children with IBD, because these may indicate deficient care coordination.
Methods
Using a multiple case study approach, we sought perspectives through semi-structured interviews of caregivers (parents, primary care providers, and gastroenterologists) for children with IBD who had a recent ED visit in either of two large pediatric referral centers in the southeastern US. We used criterion sampling to identify eligible participants through a medical record report of ED visits, and iterative sampling concurrent with analysis until no new themes were identified. Interviews were transcribed verbatim, and transcripts were coded using directed content analysis to identify emergent themes.
Results
From twenty-six interviews, three major themes emerged: perceptions of appropriate expertise, desire for integration of information and services, and making assumptions instead of engaging. Participants describe distinct roles for primary care and gastroenterology providers and recognize communication and information barriers to better coordination. Some parents and gastroenterologists perceive challenges to engaging primary care providers. Common recommendations include explicit guidance from gastroenterologists to primary care providers and methods for direct communication.
Discussion
Stakeholders describe common barriers and facilitators for effective care coordination, but some express beliefs about provider roles that could hinder improvement efforts. Tools to support asynchronous communication and shared planning may improve coordination and care quality for complications of IBD.
Introduction
Effective management of serious pediatric chronic disease requires care coordination, the purposeful organization of care activities among stakeholders in a patient’s health to facilitate delivery of health services and improve outcomes. 1 Many pediatric chronic diseases, such as asthma, diabetes, and inflammatory bowel disease, have a relapsing and remitting course in which complications are unpredictable. Families and health care teams rely on processes of care coordination to recognize developing complications and to arrange appropriate evaluation and care. Processes such as proactive planning, coaching for parents and caregivers, efficient communication, and timely information exchange support high-quality care that links services provided by community-based primary care providers and hospital-based specialist physicians.2,3
Effective care coordination is associated with better pediatric chronic disease control, higher family satisfaction, and decreased use of the hospital and emergency department (ED).4,5 Lack of care coordination personnel, such as nurses and social workers, limits care coordination in pediatric primary care, 6 but many children with serious chronic disease have access to care coordinators through hospital-based specialty programs. 7
Pediatric inflammatory bowel disease (IBD) is an example of a pediatric chronic disease for which effective care of disease-related complications requires care coordination. IBD affects about 75,000 children and causes substantial morbidity and diminished quality of life.8,9 Care for pediatric IBD includes acute and chronic management of disease-related symptoms, childhood preventive care plus healthcare maintenance services consequent to IBD, and attention to emotional, behavioral, and social consequences of a serious chronic disease. 10 Delivering these services involves a pediatric gastroenterologist (GI), primary care provider (PCP), parents, specialist or generalist nursing staff, and often behavioral health clinicians and school personnel.
Children with IBD experience a high rate of emergency department (ED) visits for reasons such as fever, abdominal pain, and bloody stool 11 ; IBD specialists believe many of these visits may be avoidable through improved care coordination. 12 Avoidable ED visits are expensive to the health care system and expose patients to potential harms such as opiate prescriptions and ionizing radiation. 12 Theory suggests that interventions to support care coordination processes such as parent or caregiver preparedness, health care team communication, and proactive planning may influence use of the ED for complications of chronic disease.13,14 However, prior research has not yet explored the critical components of effective care coordination from the perspectives of stakeholders who direct care for children with IBD.
To improve understanding of barriers to and facilitators of effective care coordination for children with IBD, we conducted a multiple case study through interviews to explore perspectives of parents, PCPs, and GIs engaged in their care. Because ED visits may signal deficient care coordination, we targeted those caregivers for children with IBD who had a recent ED visit to explore opportunities, barriers, and facilitators of effective care coordination.
Methods
Study design and setting
We conducted a multiple case study using semi-structured, telephone interviews with parents, PCPs, and GIs caring for a child with IBD in North Carolina or Tennessee. Interviews were designed to gain understanding of parents’ and providers’ experiences of care coordination and perceived barriers to and facilitators of care coordination effectiveness. Recruitment and interviews took place between September 2017 and May 2018. Each academic center provides comprehensive care for IBD, but defers primary care and other health needs to community-based PCPs. The centers fax clinical notes to referring PCPs to share information, but do not have other formal coordination strategies in place. IBD program nurses help families by faciliating authorization of specialty medications, testing orders, and appointments, but do not have formal processes for care coordination with PCPs. The study protocol was reviewed and approved by the Institutional Review Boards at the University of North Carolina (IRB #170491) and at Vanderbilt University (IRB #180027).
Sampling and study participants
Eligible participants were those whose child/patient was between ages 6 and 18 years, had an established diagnosis (>6 months) of Crohn’s disease or ulcerative colitis, and had an ED visit in either academic health system for a problem relevant to underlying IBD during the month prior to recruitment. We first used criterion sampling to identify eligible participants through a weekly medical record report of ED visits. Relevance to underlying IBD was determined by a trained research assistant and by GIs at both sites, who reviewed electronic medical records to determine the diagnosis for each ED visit. For example, visits for fever or abdominal pain were included, but visits for injuries were excluded.
We aimed for data source triangulation by targeting parents and health care providers engaged in care of the same child. We used iterative sampling to conduct small batches of consecutive interviews, and concurrent data analysis informed subsequent sampling from different subject categories (parent, PCP, GI). Interviews continued until we reached thematic saturation, which was defined when data analyses identified no new major themes. We sent an initial letter describing the study and offering participation to eligible parents, PCPs, and GIs identified through children’s medical records. Over the ensuing two weeks, study team members completed up to three follow-up phone calls to offer participation. Stakeholders received a $50 gift card for completing the interview.
Data collection
We developed separate, complementary interview guides for each of the three stakeholder groups, and included overlapping questions about appropriate assessment of symptoms, roles of distinct providers in delivering care, communication and information exchange among providers, and perspectives on collaboration. These topics were derived from the existing literature on care coordination for pediatric chronic disease.1,15 An overview of the topics tailored to each stakeholder group’s interview is presented in Table 1.
Example interview topics by stakeholder groups.
PCP: primary care provider; GI: pediatric gastroenterologist.
Participants completed informed consent by phone, then participated in a 30- to 45-minute telephone interview conducted by an experienced qualitative researcher (MBV). Interviews were audio-recorded with participant permission and transcribed verbatim. We did not collect demographic information about subjects nor about cases. Doing so for subjects would have required a more burdensome consent process that we feared would discourage participation. For cases, obtaining demographic information would have required separate, written authorization to access and use data from the medical record beyond the weekly report of ED visits, for which we received a waiver of authorization for sampling purposes only.
Data analysis
We imported transcripts into ATLAS.ti 7.5.18 (Scientific Software Development GmbH) to facilitate analysis. Existing care coordination literature and the interview questions guided development of the initial codebook, which was revised based on emergent themes. The interviews were coded by two team members (MBV and NdJ) using a directed content analysis approach, with consensus achieved by discussion when initial codes differed. 16 One team member (MBV), who leads a qualitative research methodology unit, led initial coding. She and the second coder (NdJ) trained in qualitative research methods met biweekly to review transcripts and codes and make plans for iterative sampling. Once per month, they met with the team’s IBD content expert (MDK) to clarify questions about participants’ words and code selection.
Extensive prior research on care coordination theoretical frameworks and processes guided initial code selection, including characteristics of setting, purpose, mechanisms, and situations of care coordination.3,17,18 We grouped and condensed codes iteratively until obtaining a limited set that captured chacteristics of care coordination frameworks and the evidence base. Analysis was descriptive and focused on capturing all relevant information provided by stakeholders, allowing for anticipated and emergent themes. Though we anticipated conducting analyses both within- and between-groups, we reached thematic saturation with a smaller number of PCPs, making data source triangulation for each case – and between-groups analysis – impracticable.
Trustworthiness
To enhance confirmability, we included three members of the research team who have prolonged engagement in the care of children with IBD, and specifically have developed trusting relationships with the parents and PCPs who participated. None of them conducted interviews. We aimed for data source triangulation among three sources for each case (parent, PCP, GI), but did not achieve that adequately. To enhance transferability, we used thick description using participants’ quotes to give context to findings. To enhance dependability and confirmability, the qualitative research expert on the project (MBV) kept an audit trail of all project activities, including process notes, versions of study instruments, and raw data. Monthly meetings with the IBD content expert also served to foster reflexivity.
Results
We completed 26 interviews: 14 with parents, 5 with PCPs, and 7 with GIs (Table 2). About half of invited parents, one-third of invited PCPs, and all invited GIs participated. The vast majority of non-participants were unable to be reached by the study team within the three follow-up phone calls allowed. Three overlapping themes emerged from descriptive analysis: (1) perceptions of expertise; (2) desire for integration; (3) assumptions rather than engagement. Within each theme, participants’ quotes highlight barriers to and facilitators of effective care coordination for pediatric IBD. Participants’ specific recommendations for improving care coordination were grouped separately.
Interview participants (n = 26).
Theme 1: Perceptions of appropriate expertise
Stakeholders in all three groups felt that PCPs were responsible for overall health and “big picture” issues, while GIs were responsible for all health issues related to IBD. However, several parents described a weakened connection to their child’s PCP following an IBD diagnosis. “We don’t really interface much with [name of primary care practice]. I mean we pretty much go right to Dr. [GI Name]. I mean that group is his family physician but it’s not like he regularly sees them or anything.” (Parent) “I don’t go too much to the pediatrician because she has all her shots.” (Parent)
Exceptions occurred when the PCP and GI had a relationship or when families lived far from the location of their child’s GI specialty care. Parents generally expressed confidence in their ability and their child’s ability to distinguish IBD-related symptoms from symptoms related to other childhood illnesses. All GIs wanted their patients to notify them when patients or parents were concerned about symptoms. Some PCPs and GIs wanted their patients to undergo initial assessment in the primary care office, with telephone consultation to the GI practice. Others preferred that patients go directly to the GI practice or to the ED.
Parents looked to their child’s GI for guidance on where and when to take their child for evaluation of symptoms. In the absence of clear guidance, several parents expressed preference for taking their child “straight to the hospital” so that they would not “waste any time” trying to contact their child’s PCP or GI. Conversely, several PCPs and GIs highlighted the idea that primary care should perform an initial “triage” of symptoms to save families trips to the ED or referral center. “I mean they have normal kid things, too, in addition to their specialty care….So any time they have a GI complaint, no, I’m not going to immediately say, oh yeah, this is your IBD, go see GI. We’ll see them and do our best.” (PCP) “So many times, the primary care’s role is really to be our eyes, to evaluate those patients, to see if there are symptoms that may be more related to fear than actual physiology or pathology, as well as whether or not their symptoms have progressed and really warrant further evaluation.” (GI)
Some GIs also felt that limited same-day access in their clinical schedules and an inability to obtain laboratory and imaging studies through the clinic setting made it easier to refer patients directly to the ED for evaluation of symptoms. According to one GI, “It’s a waste to come to our clinic, so then we send them to the [ED].”
PCPs and GIs expressed different ideas about who should be responsible for managing issues such as fatigue and emotional or mental health problems, but both groups wanted their patients to maintain strong connections to a PCP.
Some parents saw the role of the PCP as managing health maintenance concerns and diagnosing and treating minor illnesses. According to one parent, “They’re for basic stuff…like your shots and your basic are you growing, do you have a cold, you know?” PCPs highlighted their expertise with health maintenance issues such as vaccinations and growth. Several were eager to engage with health concerns that might be related or secondary to IBD, such as emotional or mental health problems. “I assess like the social kind of stuff, like with [Child’s Name] even from the beginning, I tried to encourage them to get into counseling and therapy.” (PCP)
Many parents indicated that they would seek advice from the child’s GI for any health concerns because of the likelihood that those could be related to the child’s IBD. Many GIs expressed that they should direct all care for IBD and related health problems, and that a critical part of that role should include guiding the PCP’s care of health concerns secondary to IBD. “In the case of [IBD], we provide comprehensive care, and we work with the pediatrician…to ensure that some healthcare maintenance needs are met.” (GI)
Theme 2: Desire for integration of information and services
Although parents felt that PCPs and GIs had different roles, they wanted all their child’s care to be “interconnected.” Several PCPs also expressed this sentiment, and one framed it using the concept of the medical home, a team-based model of health care delivery that is family-centered, comprehensive, and continuously available (Tschudy et al., 2016).
6
“I see it kind of like…the medical home. So if the family has a question about his symptoms or if they need resources somehow, I would be the one that would be able to contact the GI specialist,…ask his opinion, if it’s something I could handle in my office…or if the GI doc would want to see him and handle it that way.” (PCP)
GIs described connectedness as messages through shared electronic medical record software and through copies of clinical documentation sent to the office of a patient’s PCP. However, several described uncertainty about the strength or effectiveness of the connection. “Well, what I hope is that they [PCPs] will read our letters, and I’m not sure if that happens actually….It would be good to have some reassurance in that regard.” (GI)
Nearly all participants expressed that communication among parents and health care providers was an important, and deficient, part of effective care for a child with IBD. Several parents and PCPs were frustrated with paging systems used by GI practices because of difficulties and delays getting in touch with a GI. GIs expressed frustration with trying to contact PCPs whose offices used computerized telephone systems and who were not available because of their “busy schedule.”
All stakeholder groups described that communication between PCPs and GIs was lacking. Several parents described a sense of having to be information intermediaries between their child’s health care providers. All these perspectives informed a minor theme of “inadequacy of information sharing.” “I don’t think that they have [communicated], but I may be misspoken on that. I think maybe most of the communication that has taken place has been probably through paperwork or my wife’s conversations with [PCP’s name].” (Parent)
PCPs and GIs reported that faxed clinical documentation was the primary mode of communication between them. Several shared that the lack of communication between them negatively affected care for their shared patients. “When patients get frustrated is where they feel like the right hand is not talking to the left hand, and they feel like one doctor doesn’t understand the other doctor’s perspective or they’re getting completely new information from the one doctor who says they’re fine and the other doctor says they’re not.” (GI)
Theme 3: Making assumptions instead of engaging
PCPs and GIs made assumptions about their respective roles in the absence of clear communication. Many PCPs assumed their role was minimal, while GIs described taking on additional aspects of care beyond those related to IBD. “I don’t think there’s anything I would really be able to do, which maybe that’s a lack of education on my part. Maybe if the GI doctor would be able to guide me on something I could do for [his symptoms].” (PCP) “I mean if the primary doctor acts interested and responds, the parents will stay with them and go with them. But most of the time, at some point usually early on, the primary doctor tells them not to call them or when they call them, they tell them to call us. So the parents get the hint pretty quickly.” (GI)
Many parents indicated that lack of communication between their child’s health care providers was a barrier to coordinated care. “No [they don’t communicate], because that’s why I keep a binder, because every time we go, I get asked the same questions again.” (Parent)
Both PCPs and GIs reported that the other’s lack of time and infrastructure (e.g., direct phone lines and electronic messaging portals) were barriers to communication between them. Reliance on faxed information was a particular frustration for PCPs and GIs trying to coordinate care. “It seems like it takes a couple of tries to make sure that the information I’m trying to send them, like lab results for example, not knowing if it would get to the right person.” (PCP)
Even though GIs reported faxing copies of clinical documentation to PCP offices, several PCPs reported that lacking current information about a child’s IBD-related care was a barrier to engaging in efforts to coordinate with GIs.
Stakeholder recommendations for coordinated care
Stakeholders in all three groups felt that structures and processes to streamline communication helped to facilitate coordinated care. Parents valued being able to reach clinical staff directly at either practice. For health care providers, access to a phone number or email address to reach clinical staff directly, sharing a common electronic medical record, and having personally talked with each other were all considered beneficial for coordinating care. “So in [Patient Name]’s case, we have exchanged a couple of emails to kind of coordinate some of the things that [PCP name] was doing in relationship to her vomiting.” (GI)
Many participants offered specific recommendations to improve coordination of care for children with IBD. Similar recommendations were offered across stakeholder groups (Table 3). Parents’ recommendations included development of individualized care plans to share across providers, addressing the mental health and emotional consequences of living with IBD, and choosing a PCP and GI doctor in the same health care system.
Stakeholder recommendations for coordinated care of pediatric IBD.
PCP: primary care provider; GI: pediatric gastroenterologist.
“I think [Child Name] thinks it sounds like she has the plague or something. She’s not comfortable with it, and I think that’s because she doesn’t know a lot about it. She doesn’t understand. I mean she probably knows that it’s a chronic illness, but she doesn’t understand it very well, and she thinks she’d be judged because she has this diagnosis. So she has not told anybody about this at all. She lost a boyfriend over this. It’s been kind of ridiculous. I think some education would be a great thing.” (Parent)
Several PCPs and GIs highlighted the importance of explicit guidance from the GI doctor to the PCP on what care activities he or she should deliver. ““If they just sent me a direct [electronic medical record] communication, provider to provider, like I saw [Child Name] today. These are the things that we’d like to – for you to either monitor or follow up on. Something that’s not a note but just something that like here are…five things you should be aware of.” (PCP) “I think we could do a better job communicating with them about what we’re doing in terms of management…and additional things that are needed in terms of routine healthcare maintenance, like vaccines.” (GI)
Some GIs also described a process of intentional care planning that they felt was helpful for guiding families’ decision-making and actions around seeking care. “We actually provide our patients a simplistic symptom log of here are things that warrant contacting the pediatric GI office, so that hopefully it will keep them focused on when they should call.” (GI)
Several parents and PCPs reflected the same sentiments about the benefits of clearly specifying important care activities and responsible parties.
Discussion
In our study, parents and health care providers for children with IBD agreed that PCPs and GIs have different and important roles, that stakeholders should coordinate care through effective communication, and that communication and information sharing are currently inadequate to achieve coordinated care. Several described negative consequences of uncoordinated care, such as frustration for parents, lack of connection to primary care, over-reliance on GIs, and potentially avoidable ED visits.
These problems have been described in prior research examining coordination of care between PCPs and specialist practices. PCPs report that lack of actionable guidance from specialists and poor communication limit their role in the care of children with complex medical problems. 19 Similarly, deficient care coordination prevents specialty practices from being full partners with primary care medical homes. 6 These are critical deficits, because better coordination in the care of children with chronic diseases is associated with higher family satisfaction with health care and with lower use of the ED. 20
Our findings provide guidance to operationalize coordination in the care of children with IBD. This is important because many families of children with complex conditions expect specialist providers to coordinate all aspects of care. 21 However, many health needs do not require the level of care in specialty practices, and specialty practices may not be able to meet all of a patient’s medical needs.
Specifically, this study’s findings (a) demonstrate that parents, PCPs, and GIs are interested in improving coordination of care among them, (b) perceive common barriers – such as inadequate communication and information sharing – to improving coordination, and (c) believe that improving care coordiation could improve quality and outcomes of care for children with IBD. The themes that emerged and recommendations made by stakeholders can inform a process of developing and piloting individualized care plans and tools to enhance communication among parents and health care team members for children with IBD.
Such interventions could take advantage of the growing availability of pediatric IBD-program nurses who function as a directly available, first-line contact for families of children with IBD. Both academic centers involved in this study employ at least two full-time nurses who serve in that capacity. Incorporating nurses into a structured planning and communication process that bridges relationships among families, PCPs, and GIs could be a way to leverage existing resources and make developing effective care coordination models more feasible.
Similar interventions to develop individualized care plans and implement those through nurse team members have helped reduce hospitalization and morbidity for children with a wide range of complex medical problems.22,23 Positive outcomes associated with individualized planning are most well documented for children with asthma, 24 but heterogeneity of interventions and lack of reproducibility in the study of other, individual childhood chronic diseases have hampered assessment of the relationship between individualized plans and outcomes. 25 This study’s findings can help advance the field of care coordination for childhood chronic diseases by informing development of interventions grounded in the deficiencies and opportunities that stakeholders describe.
This study has several limitations, which can be organized by the Standards for Reporting Qualitative Research. 26 The first relates to transferability. We reached theme saturation quickly with PCPs because many had inadequate information about the patient in question. As a result, we had less rich information from PCPs than from parents and GIs. We also did not have a potential sample of nurses to recruit for interviews. We believe this limitation reinforces the importance of the topic, though; even as GIs consistently describe a need for more meaningful collaboration with PCPs, those PCPs lack information needed to participate more actively in coordinated care; and even as the demand for nurses and other frontline staff to deliver care coordination grows, those staff are sometimes not involved in initiating or planning care coordination activities. The second limitation also relates to transferability. Recruitment of subjects was completed through two large, regional, pediatric referral centers. But all care coordination is contextually specific, and so the experiences and perspectives of stakeholders near other referral centers in the United States and in other countries may differ in ways we are unable to capture. However, focusing recruitment on two referral centers allowed us to purposefully sample parents and health care providers of patients with recent ED visits. We believe this strategy helped to enrich our sample with participants who had valuable information to contribute about care coordination. The third is an issue of data collection methods that could limit credibility. In an ideal multiple case study, we would have triangulated not just data sources but also data types. An even richer study, for example, could approach the issue of care coordination not just through caregiver interviews but also through review of relevant medical record data and through surveys. The resources available for this study precluded such thorough triangulation, but those would be excellent subjects for future research on this topic. Finally, a richer characterization of interview participants and of cases themselves would provide more thorough context for understanding this study’s findings. Collection and reporting of those data were not feasible for this study, which could affect both transferability and dependability.
Parents of children with IBD and their health care providers share similar goals for care coordination and describe common barriers to and facilitators of effectiveness. However, some parents and physicians express beliefs and assumptions about provider roles and engagement that could hinder improvement efforts. Efforts to improve care coordination for children with IBD should leverage the widely shared perception that mitigating deficits in communication and information sharing is critical to manage pediatric IBD more effectively across care teams.
This study’s findings have particular relevance for patients and families because their lived experience can inform development of more effective care coordination tools and models. Specifically, parents and health care providers agreed that care plans to guide families’ actions and explicit guidance to engage PCPs may be effective strategies to improve care coordination and delivery for children with IBD. Implementing those strategies into care teams that may be fragmented by different electronic medical record systems and may lack dedicated care coordination staff will require development and testing of new tools and models that leverage existing resources and staff. Translation of those models into templates sharable across medical record platforms may make it feasible for diverse stakeholders to collaborate on care coordination processes that improve care delivery and outcomes for children with IBD and other serious pediatric chronic diseases.
Footnotes
Authors' note
Dedrick E Moulton is currently affiliated with Louisiana State University Health Sciences Center, USA.
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Neal deJong reports funding from the National Center for Advancing Translational Sciences.
Maihan Vu, Jiawei Cui, Michael Dole, and Dedrick Moulton have no disclosures to report.
Michael Kappelman reports research funding from the Patient Centered Outcomes Research Institute and the Crohn’s and Colitis Foundation of America.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by a UNC Department of Pediatrics PROMISE Research Grant. The funder played no role in the design or conduct of the study and did not influence interpretation of the findings.
Dr. deJong’s time on this project was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant KL2TR002490. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
