Wang CJ, Little AA, Kamholz K, et al. Improving preterm ophthalmologic care in the era of accountable care organizations. Arch Ophthalmol. Published online July 9, 2012. doi:10.1001/archophthalmol.2012.1890
eAppendix 1. Parent focus group/interview guide
eAppendix 2. Provider interview guide
Wang CJ, Little AA, Kamholz K, Holliman JB, Wise MD, Davis J, Ringer S, Cole C, VanderVeen DK, Christiansen SP, Bauchner H. Improving Preterm Ophthalmologic Care in the Era of Accountable Care Organizations. Arch Ophthalmol. 2012;130(11):1433-1440. doi:10.1001/archophthalmol.2012.1890
SECTION EDITOR: PAUL P. LEE, MD
Author Affiliations: Department of Pediatrics, Stanford University School of Medicine, Stanford, California (Dr Wang); Departments of Pediatrics (Drs Wang, Kamholz, and Cole, and Mss Little and Holliman) and Ophthalmology (Dr Christiansen), Boston University Medical Center, Division of Newborn Medicine, Floating Hospital for Children at Tufts Medical Center (Dr Davis), Department of Newborn Medicine, Brigham and Women's Hospital (Dr Ringer), Department of Ophthalmology, Children's Hospital Boston (Dr VanderVeen), Boston, and University of Massachusetts Medical School, Worcester (Ms Little), Massachusetts; Weill Cornell Medical College, New York, New York (Ms Wise); Journal of the American Medical Association, Chicago, Illinois (Dr Bauchner).
Objectives To understand retinopathy of prematurity (ROP) follow-up care for preterm very low-birth-weight infants (VLBW; <1500 g) in the context of the chronic care model and identify opportunities for improvement under accountable care organizations.
Methods We conducted focus groups and interviews with parents (N = 47) of VLBW infants and interviews with neonatal intensive care unit and ophthalmologic providers (N = 28) at 6 sites in Massachusetts and South Carolina. Themes are reported according to consolidated criteria for reporting qualitative research guidelines.
Results Respondents perceived that legal liability and low reimbursement contributed to shortages of ROP providers. Some neonatal intensive care units offered subsidies to attract ophthalmologic providers or delayed transfers to institutions that could not provide ROP examinations and/or treatment. Sites used variable practices for coordinating ROP care. Even at sites with a tracking database and a dedicated ROP coordinator, significant time was required to ensure that examinations and treatment occurred as scheduled. Parents' ability to manage their children's health care was limited by parental understanding of ROP, feeling overwhelmed by the infant's care, and unmet needs for resources to address social stressors.
Conclusions Under accountable care organizations, hospitals and ophthalmology practices should share responsibility for ensuring coordinated ROP care to mitigate liability concerns. To promote integrated care, reimbursement for ROP care should be bundled to include screening, diagnosis, treatment, and appropriate follow-up. Clinical information systems should be enhanced to increase efficiency and limit lapses in care. Self-management tools and connections to community resources could help promote families' attendance of follow-up appointments.
Very low-birth-weight infants (VLBW; <1500 g) are at risk for retinopathy of prematurity (ROP), a potentially blinding condition,1,2 and other ophthalmologic morbidities such as myopia and strabismus.3 Timely screening and treatment via laser or cryotherapy decreases the risk for blindness due to ROP.1,4- 6 Yet, many infants who need an ROP examination following hospital discharge do not receive one in a timely fashion.7,8 Moreover, we found that only 1 in 5 VLBW children enrolled in Medicaid in 1 state received an ophthalmologic examination between ages 1 and 2 years,9 a quality indicator recommended by an expert panel.10 Known factors associated with inadequate follow-up include Black race,7,9 young maternal age,8 and reliance on parents to schedule appointments.7
There may be opportunities to improve ROP care for VLBW infants by linking reimbursement to quality, efficiency, and patient outcomes11 under accountable care organizations (ACOs)12 or other systems that bundle payments over the full cycle of care.13 Moreover, there is evidence that implementation of the chronic care model can improve quality of care and health outcomes for patients with chronic illness.14 We used this model as a framework to identify opportunities for improvement in the provision of ROP care in the era of ACOs. The model includes 6 components: health care organization, delivery system design, community resources, patient self-management support, clinical decision support, and clinical information systems.15
To apply the chronic care model to ROP care for VLBW infants, we conducted a qualitative study to capture the perspectives of key stakeholders: ophthalmologic providers, neonatal intensive care unit (NICU) providers, and families of VLBW children. We discuss our findings about gaps in care and identify potential applications of concepts from ACOs.
We recruited parents and health care providers (ie, neonatologists/NICU nurses, ophthalmologists, ROP coordinators, discharge planners, and follow-up clinic staff) of VLBW children at 6 hospitals in Massachusetts and South Carolina between July 2008 and April 2010. Sites were selected to provide sociodemographic (Northern and Southern states), geographic (urban and rural), and socioeconomic diversity. Each site's institutional review board approved the study.
Parents were eligible to participate if their children were born VLBW and were discharged from the NICU during the previous year. One parent from each family was eligible to participate. Of these, we excluded parents not fluent in English or Spanish owing to staff language constraints. Each participant attended a 2-hour focus group at the site from which his or her child was discharged. Subjects received $30 compensation for their time plus reimbursement for transportation and child care. Focus groups were conducted in English by staff trained in qualitative research techniques. Because few Spanish-speaking parents were eligible and available, 30-minute individual telephone interviews were conducted in Spanish. Subjects gave verbal consent at the time of participation. Focus groups and interviews were conducted using a semistructured guide, which addressed reasons for not receiving ophthalmologic follow-up care, experiences with accessing care, and suggestions for improvement (eAppendix 1.
We recruited health care providers involved in vision care of VLBW infants, including those who performed ROP examinations, treatment, and/or ongoing follow-up after the retina is mature (eg, strabismus screening), and those who coordinated ROP care and/or educated parents. Eligible health care providers were e-mailed an invitation to participate in a 30-minute interview conducted by trained research staff. Subjects verbally agreed to participate. The semistructured interview guide (eAppendix 2) included questions about connecting patients to ophthalmologic follow-up care, problems encountered when arranging care, and strategies to address these issues.
Focus groups and interviews were recorded, transcribed, and analyzed according to grounded theory.16 Transcripts were coded independently by 2 research team members and reviewed by a third. Discrepancies were reconciled to generate a final code book. Each transcript was recoded, and codes were organized into major themes. Theme saturation occurred when no new themes emerged and recurrent themes continued to be identified. Findings are reported according to consolidated criteria for reporting qualitative research guidelines.17
The following strategies were used to ensure data validity18,19: (1) investigator triangulation in which investigators from different disciplines read the transcripts independently before developing the code book together; (2) expert triangulation in which primary investigators experienced in health services research, neonatal intensive care, and pediatric ophthalmology crosschecked the methods and findings; and (3) parent-provider triangulation in which findings from parents and health care providers were compared to identify co-occurring themes. A software package was not used; however, such software is a tool for data management, not analysis.18
We conducted 10 focus groups at 5 sites (2 groups per site). Groups had 3 to 7 participants. Three individual Spanish interviews were conducted at 1 site. A total of 47 parents participated. Most were women (96%) and identified as black/African American (45%) or white (40%). Most (87%) reported that their children had received vision services since hospital discharge (Table 1).
We interviewed 28 health care providers across the 6 sites, including ophthalmologists and retina specialists (N = 6), ROP coordinators (N = 3), NICU physicians and nurses (N = 12), discharge planning and social work staff (N = 4), and follow-up providers (N = 3) (Table 1). Of the ophthalmologists and retina specialists, 3 performed ROP screening and treatment, 2 provided screening but not treatment for ROP, and 1 conducted ongoing follow-up after the retina is mature.
Here we discuss our findings using the 6 components of the chronic care model.15Table 2 summarizes gaps in the current system, and Table 3 lists opportunities for improvement under ACOs.
Several participants perceived a limited availability of ophthalmologic providers who perform ROP screening and treatment. One ophthalmologist stated, “There are a lot of hospitals . . . scrambling to get their ROP services covered. ” Interviewees acknowledged that access to training in ROP may influence provider availability. However, they also perceived that concerns about financial compensation and legal liability may influence decisions about whether to perform ROP screening and treatment. Another ophthalmologist said, “I do ROP because it's the right thing to do, but . . . you don't get paid very much, it's very time and labor intensive, and it's a minefield medically and legally. You really have to be nuts to do this.” Provider shortages can affect ROP care before and after hospital discharge. Some NICUs offered subsidies or supplemental malpractice insurance to ROP providers, delayed transfers to institutions that could not provide ROP examinations or treatment, or transported infants between hospitals for ROP consultations. After discharge, families sometimes must travel long distances for ROP follow-up appointments, especially in rural areas.
Some participants felt that telemedicine through digital retinal imaging could help alleviate perceived ophthalmologic provider shortages. One said, “Probably a lot of rural places . . . could really benefit from it.” However, another ophthalmologist thought it could contribute to legal liability: “It documents what you see, which is both an advantage and a disadvantage . . . I’ve seen [images] . . . used against doctors in a way that was totally inappropriate.”
Some sites had dedicated ROP coordinators who identified infants at risk for ROP, arranged examinations and treatment, and monitored whether they occurred. Many interviewees felt that coordinators were critical for preventing missed or delayed examinations. One ophthalmologist said, “Were we to lose [our ROP coordinator], I would stop doing my exams . . . because it's sort of like pilots who are off flying airplanes without a checklist . . . Sooner or later there's going to be a bad mistake.” At some sites, the coordinator also provided education about ROP and maintained relationships with families. One coordinator felt that this promoted attendance at ROP follow-up appointments: “I talk to every parent once a week. If you say . . . ‘how's things, did you get my letter . . . have any questions?’ Well, then they're much more likely to, they have that rapport with me, it's not like [I’m saying] ‘come here on Monday, you have to be there!’”
The first outpatient ROP examination usually was scheduled prior to discharge by the ROP coordinator or a NICU nurse. At a few sites, the unit secretary made the appointment. Several participants stated that NICU or clinic staff who were not trained in ROP sometimes made inappropriate follow-up appointments. Some also reported gaps in follow-up owing to weekend or holiday discharges or when regular staff were not available to schedule appointments (Table 2). While NICU staff were sometimes able to coordinate outpatient ROP appointments with other specialty visits, this usually required special effort.
Unmet needs for resources can make it difficult for families to attend outpatient ROP follow-up appointments. Lack of transportation was a common barrier for both rural families in South Carolina and the urban poor in Massachusetts. One parent said, “I didn't have a ride [to the ROP appointment] so I just didn't come. Other clinics provide transportation . . . if that's the only reason why you can't make it, they’ll make it happen.” Moreover, providers reported that transportation available through Medicaid sometimes is not reliable. Some parents also need help navigating insurance policies or accessing benefits to cover medical expenses. In South Carolina, some parents had to request insurance authorization for ophthalmology visits, which delayed care: “[The pediatrician and ophthalmologist] wouldn't accept the same kind of Medicaid program . . . [it] took me forever to get special permission for her to go back to the special eye doctor.” Others described delays in receiving benefits such as Medicaid and Supplemental Security Income: “We didn't know that we were eligible for any financial aid until we . . . ran into [an acquaintance] who happened to have a preemie . . . It turns out we were eligible for [Medicaid] and a little bit of money every month.” These delays can create barriers to attending ROP visits: “I haven't even been approved [for Supplemental Security Income] yet . . . That's 1 deterrent of going to the eye doctor. Because I’m paying $97 co-pay . . . with a preemie, 2 or 3 appointments, eye doctor's appointments . . . it's getting excessively expensive.”
We defined effective self-management as timely attendance of all recommended outpatient appointments for ROP care and long-term ophthalmologic follow-up.10 It is critical for parents to understand the significance of ROP screening and make every appointment a priority. While most parents stated they were aware of the need for ROP follow-up, health care providers felt that parents often do not truly understand the importance and time-sensitivity of ROP examinations. One ophthalmologist said, “Eye doctors are not generally considered to be critical by the public . . . [ROP] is probably 1 exception where time course is of critical importance to an eye doctor. And mom doesn't always get that.” Some sites required parents to acknowledge the risks of missing ROP appointments in writing.
Self-management also requires that parents feel confident in their ability to manage their children's medical needs. Parents reported feeling overwhelmed by the amount of information received during hospital discharge and the stress of caring for a VLBW child. The number of specialty follow-up appointments can be difficult for parents to handle. One provider explained, “If the mother is confronted with me [ophthalmologist] wanting to see the baby, and [3 other doctors] wanting to see the baby . . . she feels like it's too much to go to every single visit.” Indeed, some parents had difficulty keeping track of multiple appointments: “She has either 1 [ophthalmology appointment] coming up or I just missed it. Because . . . so many appointments that she had throughout that time, and me dealing with my other kids, taking them to doctors . . . ” Parents also must be able to manage social stressors, which can make ROP care seem a less important priority. Another provider explained, “The parents will have limited resources . . . they're balancing all these conflicting needs, and, you know . . . who cares about an eye appointment? So they postpone it or they forget to come in.”
While clinical guidelines are available for ROP management,20 oversights by ophthalmologists, NICU staff, and follow-up health care providers still lead to gaps in care (Table 2). One NICU provider explained, “The residents are . . . supposed to e-mail the ophthalmologist . . . upon discharge. But it goes through about 3 cycles [of residents] and it doesn't get passed on.” Another provider stated, “The patient was scheduled for outpatient retinopathy follow-up [but] got admitted to the hospital for diarrhea. Nobody recognized . . . that we should be following the eyes, and the eyes didn't get examined for another month . . . the baby had a retinal detachment.” Reminders about recommendations for ongoing follow-up after the retina is mature10 are also important, especially for general pediatricians who may see few VLBW patients. One parent said, “[In the NICU,] they say that in 1 year they schedule [follow-up ophthalmology appointments], but . . . their pediatrician told me that it's not necessary to take them to eye care doctors.”
The primary challenges for clinical information systems were identifying infants at risk for ROP and monitoring whether examinations and treatment occurred as scheduled. Some sites had a formal database maintained by an ROP coordinator, while others used ad hoc methods (eg, Google calendar) because of low numbers of at-risk infants or lack of institutional resources. Even at sites with formal systems, significant effort was required to monitor and update the database. One ROP coordinator said, “It relies on my being super conscientious . . . in making sure I don't miss a baby and making sure babies that go home or are transferred . . . attended their appointment.” After discharge, an additional challenge was maintaining current contact information for families. Because addresses and phone numbers change frequently, making appointment reminders and monitoring attendance of follow-up appointments can be time consuming. One provider said, “It is like being kind of a detective yourself, you're getting some emergency numbers, you're finding friends [of the family].” Staff sometimes worked with pediatricians to reach families or asked parents about ROP during NICU follow-up clinic visits.
Addressing gaps in ROP follow-up for VLBW children7- 9 is important to prevent adverse visual outcomes and promote efficient, high-quality care. We used the chronic care model15 as a framework to analyze our qualitative findings about gaps in ROP follow-up and identified areas for improvement in the era of ACOs.
Ensuring reliable ROP coverage in the NICU and appropriate follow-up after discharge is in the best interest of the hospital. While infrequent, ROP malpractice claims often target hospitals, neonatologists, and ophthalmologists for poor care coordination and loss to follow-up; settlements can range into the millions.21,22 Inadequate ROP coverage also creates inefficiencies in the NICU, such as delaying transfers to sites that lack ROP providers. To achieve reliable coverage, ACOs may need to mitigate potential concerns of ophthalmologists about reimbursement23 and liability.24 One approach may be for hospitals to share responsibility for ensuring appropriate ROP follow-up with ophthalmology practices and subsidize or offer supplemental malpractice insurance to ROP providers.
To promote coordinated follow-up after discharge, ACOs could create a dedicated service line that bundles reimbursement for the full cycle13 of ophthalmologic care for VLBW infants within the first year of life (ie, screening, diagnosis, treatment, and follow-up). The payment bundle should include an ROP coordinator to monitor examinations and educate families about follow-up visits. Ophthalmologic follow-up services also could be offered in a multispecialty clinic for VLBW children to decrease the burden on families traveling to multiple appointments and to promote communication between health care providers.
Further enhancements to delivery system design, clinical information systems, and decision support could decrease the likelihood of staff oversight and increase efficiency. In the NICU, guidelines for ROP screening and treatment20 should be integrated into a codified decision tree to help staff monitor each child's care. Maintaining a list of local ophthalmologic providers with expertise in ROP care could help NICU staff schedule appropriate follow-up appointments during weekends, holidays, or staff transitions. Electronic health records also could include automatic reminders about ROP follow-up during discharge and recommendations for long-term ophthalmologic follow-up.10 Moreover, a common clinical database would allow health care providers to share updates on the status of ROP examinations or changes to families' contact information.
Better parental self-management skills may promote attendance of scheduled ophthalmologic appointments and offer additional backup against provider oversights because confident, informed parents will be better equipped to advocate for their children's care. To promote effective self-management, institutions could provide parents with VLBW-specific tools to track medical information and appointments based on quality indicators. Prior to discharge, ACOs also could formally assess social stressors that may undermine ROP follow-up and connect families with helpful community resources.
There are limitations to our study. First, our qualitative findings are derived from participants at 6 hospitals in 2 states and may not be generalizable to other populations. However, we included urban, suburban, and rural sites, and participating parents were diverse in age, race/ethnicity, and educational status. Future research with families and ophthalmologic providers in other regions could identify additional barriers to ROP care. Second, owing to the qualitative design, we cannot assess the frequency and importance of the barriers identified. Third, the responses of participating providers reflect their own perceptions. The true risk of liability in ROP care may be lower than they perceived. Yet, in other specialties, perceptions about liability and compensation have been linked to job satisfaction25 and decisions about providing care.25- 27 Fourth, our sample did not include hospital or health plan administrators, who may have a stake in the development of an ROP follow-up system under ACOs. Future studies of the effectiveness of our proposed strategies should include administrators' perspectives on quality, costs, and liability. Fifth, our focus group participants had telephone access and were willing to participate in a discussion about improving care; nearly 90% reported that their children received vision services after hospital discharge. Therefore, our findings about self-management and unmet needs for community resources likely underestimate the challenges faced by many families.
Despite these limitations, we have identified opportunities for improvement of preterm ophthalmologic care in the era of ACOs. While some recommendations may seem intuitive, none of the institutions in our qualitative study had implemented all of them. Our recommendations can inform future quality improvement and comparative effectiveness research to assess the impact of these and other potential strategies on quality and efficiency of care as well as long-term visual outcomes for VLBW children. To optimize effectiveness, ACOs should incorporate strategies from multiple components of the chronic care model14 according to institutional priorities for improvement.
Correspondence: C. Jason Wang, MD, PhD, Department of Pediatrics, Stanford University School of Medicine, CHP/PCOR, 117 Encina Commons, Stanford, CA 94305 (firstname.lastname@example.org).
Submitted for Publication: February 7, 2012; final revision received April 2, 2012; accepted May 2, 2012.
Published Online: July 9, 2012. doi:10.1001 /archophthalmol.2012.1890
Author Contributions: Dr Wang had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Financial Disclosure: None reported.
Funding/Support: Dr Wang's work is funded by K23 Career Development Award 1K23EY018668 from the National Eye Institute.
Online-Only Material: Listen to an author interview about this article, and others, at http://bit.ly/Mcb8Qq.
Additional Contributions: The authors thank all of the families and health care providers who participated in the study. We are indebted to the study staff from all participating sites, including Linda Biedul-Kornilowicz, RN, BSN; Melissa Brown, MSN, RNC, NNP; Katherine Hamblett, MSW; Brenda MacKinnon, RNC; Ann McGill, LMSW; Amy Ruddy, BA; and Richard Saunders, MD.