PEDIATRICS Vol. 120 No. 3 September 2007, pp. 527-531 (doi:10.1542/peds.2007-0378)
ARTICLE |
Neonatologists' Practices and Experiences in Arranging Retinopathy of Prematurity Screening Services
a Program on Pediatric Health Services Research
b Departments of Pediatrics
c Ophthalmology, Duke University, Durham, North Carolina
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
BACKGROUND. Neonatologists play a central role in identifying infants who require screening for retinopathy of prematurity and in coordinating screening and treatment when necessary. No previous study has evaluated neonatologists' practices or experiences related to these activities.
METHODS. A national mail survey of a random sample of 300 neonatologists was conducted from September to November 2006.
RESULTS. The response rate was 62%. Nearly all (98%) of the respondents considered gestational age to identify infants for retinopathy of prematurity screening. However, only 19% used the currently recommended gestational age criterion of 30 weeks; instead, 6% used a lower, more-restrictive criterion, and 74% used a higher, more-inclusive criterion. Fewer respondents (77%) considered birth weight to identify infants for retinopathy of prematurity screening, most of whom (86%) used the current criterion of 
1500 g. Although more than one half (67%) of the respondents reported that pediatric ophthalmologists usually screen infants for retinopathy of prematurity in their NICU, nearly one half (46%) reported that retinal specialists provide treatment. Some (36%) reported that they have been unable to transfer a child to a NICU of lower acuity or closer to the child's home because there are no specialists available there for retinopathy of prematurity screening. Some (34%) also reported that they have needed to delay discharge because outpatient follow-up for either screening or treatment of retinopathy of prematurity is not available near the family's home.
CONCLUSIONS. We found variations in how children are identified for retinopathy of prematurity screening and how screening and treatment are provided. Future research is needed to understand how these variations affect visual outcomes and costs of care. Such work must also consider the impact of regional pediatric eye care workforce shortages on retinopathy of prematurity screening and treatment strategies.
Key Words: retinopathy of prematurity • guideline adherence • ophthalmology • neonatology • health manpower • practice patterns
Abbreviations: ROP—retinopathy of prematurity
Indirect ophthalmoscopy to screen for retinopathy of prematurity (ROP) is recommended for infants born in the United States with any of the following: birth weight of <1500 g; gestational age of 30 weeks; or birth weight between 1500 and 2000 g or gestational age of >30 weeks and unstable clinical course.1 Although the birth weight criterion has not changed, the recommendations regarding the gestational age criterion for the screening of infants for ROP changed twice in 2006. In February 2006, the criterion was changed from 28 weeks to 32 weeks2,3; in September 2006, a correction of the recommendations decreased the criterion to 30 weeks.1 The incidence of any degree of ROP among infants with a gestational age of >30 weeks has been estimated to be 
2%.4
On average, neonates require 2 to 9 ROP examinations.5,6 Little is known about the available eye care workforce and how care is organized for the detection and treatment of ROP. A survey conducted by the American Academy of Ophthalmology found that the number of pediatric ophthalmologists and retinal specialists who evaluate and treat ROP is decreasing, primarily because of liability concerns, poor reimbursement, and the complexity of scheduling care for children at risk for developing ROP.7 No national data are available regarding the degree to which infants receive recommended care for ROP or the impact of a possible shortage of specialists able and willing to screen for ROP.
Neonatologists play a central role in ensuring the timely identification of ROP. However, no data are available regarding the ROP-related activities performed by neonatologists. Although the recommendations specify the birth weight and gestational age criteria for ROP screening in the United States, the recommendations also state that "unit-specific criteria with respect to birth weight and gestational age for examination for ROP should be established for each NICU by consultation and agreement between neonatology and ophthalmology services."3 Neonatologists are also responsible for ensuring that infants at risk for ROP are identified and receive timely examinations. The guidelines recommend that neonatologists (1) ensure the availability of any needed follow-up eye care after transfer or discharge, (2) discuss the need for these services with the accepting neonatologist or the family, and (3) arrange the follow-up eye care before transfer or discharge.3 To begin to understand the delivery of care for children at risk for ROP, we surveyed neonatologists about their practices and experiences in arranging for ROP screening. Neonatologists serve a central role in coordinating care for infants at risk for ROP and can provide unique insights into the availability of eye care.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sampling Frame
A national random sample of 300 neonatologists was drawn from the American Medical Association Masterfile, a database of all licensed physicians in the United States. The sampling frame included allopathic and osteopathic physicians providing direct patient care. We excluded physicians in training and those practicing at military or federal facilities. Overall, there were 3590 neonatologists who met our inclusion criteria.
Survey Instrument Development
We developed an instrument to assess respondent and NICU demographic characteristics and practices, experiences, and attitudes regarding the screening of infants for ROP. The instrument was pilot tested by a convenience sample of neonatologists and was revised to ensure clarity. The final instrument consisted of 21 questions, which primarily were multiple choice or used Likert scales of agreement, and required <10 minutes to complete.
Survey Administration
The first survey mailing, accompanied by a cover letter, an inexpensive incentive gift, and a business reply envelope, was sent in September 2006, after publication of the 30-week gestational age criterion. Two subsequent mailings to nonrespondents were sent at 3-week intervals.
Data Analyses
General frequency responses to all survey items were determined. Pearson 
2 tests of independence were then used to test for associations among the categorical variables. All analyses were performed with Stata 8.2 software (Stata, College Station, TX). We considered P < .05 to be statistically significant. The Duke University Health System institutional review board approved this project.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Response Rate and Demographic Characteristics
Of the 300 physicians in the sample, 13 did not provide direct neonatology care and 19 had undeliverable addresses. The response rate was 62% (166 eligible surveys).
Respondent and NICU demographic characteristics are presented in Table 1. Small NICUs (<26 beds) were less likely to be in urban areas, compared with suburban areas (17% vs 38%; P < .01). Compared with small NICUs, large NICUs (>60 beds) and medium-sized NICUs (26–60 beds) were more likely to be affiliated with residency or fellowship training programs (77% and 63% vs 30%; P < .001).
|
Criteria for ROP Screening
Nearly all respondents (98%; n = 162) reported that gestational age is used to identify whether an infant should be screened for ROP. Of those, 22% did not consider birth weight to be a criterion for identifying whether an infant should be screened for ROP. The median gestational age for ROP screening was
32 weeks, with a range of 28 to 35 weeks. Overall, 19% used the currently recommended gestational age criterion of 30 weeks, 6% used a lower, more-restrictive criterion, and 74% used a higher, more-inclusive criterion.
Most respondents (77%; n = 128) reported that birth weight is used as a criterion to identify whether an infant should be screened for ROP. Only 1 respondent (<1%) considered birth weight but not gestational age as a criterion to identify children for ROP screening. The median birth weight for ROP screening was 1500 g, with a range of 1000 to 2500 g. Overall, 86% used the recommended criterion of 1500 g, 5% used a lower, more-restrictive criterion, and 9% used a higher, more-inclusive criterion.
Many respondents (61%; n = 101) reported using clinical course to identify children at risk of developing ROP. Only 1 respondent reported using clinical course but neither gestational age nor birth weight to identify children requiring screening for ROP.
Many respondents (64%; n = 106) reported that they have unit-specific written guidelines to track which infants need ROP screening. Use of written guidelines was not associated with the likelihood of using the recommended criteria for gestational age or birth weight (P > .18).
ROP Screening and Treatment
More than one half (67%) of the respondents reported that, within their NICUs, pediatric ophthalmologists usually screen children for ROP. However, some reported that retinal specialists, either alone (18%) or with pediatric ophthalmologists (4%), provide screening for ROP. Some (10%) also reported that general ophthalmologists provide screening. One respondent reported that ophthalmology fellows screen infants for ROP.
Nearly one half (46%) of the respondents reported that, in their NICUs, retinal specialists provide treatment for ROP in their main NICU practice setting. Other respondents reported that pediatric ophthalmologists, either alone (38%) or with retinal specialists (9%), provide treatment. Each of the following was reported as providing treatment by only 1 respondent: general ophthalmologist, general ophthalmologist and retinal specialist, and ophthalmology fellow. Few respondents (4%) reported that children are transferred to another setting for ROP treatment. No difference was found for NICUs that train resident or fellow physicians in the likelihood of having a pediatric ophthalmologist provide screening or treatment (P > .36).
Adequacy of Services and Barriers to Care
Although some (29%) of the respondents agreed that some children in their state develop ROP-related visual impairment that could have been prevented with timely screening, nearly all (98%) agreed that children within their NICU receive timely ROP screening. Nearly all respondents (99%) reported that ROP screening for children born in their NICU was rarely (42%) or never (57%) delayed by 1 week because the need for screening was not identified. Similarly, nearly all respondents (95%) reported that ROP screening for children transferred into their NICU was rarely (43%) or never (52%) delayed by 1 week because the need was not identified.
Table 2 presents the degree to which respondents considered each of the listed barriers to affect ROP screening, in their experience. The most commonly reported major barrier was the lack of available eye care specialists. The degree to which this was considered to be a barrier did not vary according to urban/suburban/rural setting (P = .20), number of beds (P = .43), or whether the NICU trains resident or fellow physicians (P = .13). The most commonly reported barrier overall was the perception that families do not follow up for scheduled ROP screening examinations after discharge. The degree to which neonatologists endorsed this barrier did not vary according to setting (P = .29), size (P = .54), or whether the NICU trains physicians (P = .89).
|
Some respondents reported that they are usually (10%) or sometimes (26%) unable to transfer a child to a lower-acuity NICU or one closer to the child's home because there are no specialists available there for ROP screening. Some also reported that they usually (7%) or sometimes (27%) delay discharge by >1 day because outpatient follow-up ROP screening is not available near the family's home. The proportions reporting these delays did not vary according to urban/suburban/rural setting (P > .47), size (P > .47), or whether the NICU trains physicians (P > .15).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
We found that many neonatologists use a more-inclusive gestational age criterion for identifying children for ROP screening than is currently recommended. This may reflect the lag time between correction of the guidelines and modification of clinical practice. It may also represent a more-conservative approach to avoid missing infants who may develop sight-threatening ROP. However, our finding that some neonatologists do not consider birth weight as a separate criterion and the variability of the birth weight criterion among those who do consider birth weight suggest that there are important variations in how children are identified for ROP screening. Although underreferral could lead to missed cases of severe ROP and blindness, overreferral could exacerbate any workforce shortage and crowd out children at higher risk for the development of ROP.
Although nearly all neonatologists reported that infants in their NICUs receive timely ROP screening, some reported that some infants in their state developed potentially preventable visual impairment because they did not receive timely screening. We think that this emphasizes the need for quality improvement activities to ensure that all infants receive timely ROP examinations, both before and after (when necessary) discharge. However, quality improvement activities that extend beyond admission are challenging and may be expensive to coordinate. Furthermore, it is unclear who should have primary responsibility for such activities.
One half of the respondents reported that lack of available eye care specialists is at least a minor barrier. It is noteworthy that many eye care specialists other than pediatric ophthalmologists provide screening for ROP. No data are available regarding the training or experience of other providers, including retinal specialists, in the care of ROP or the level of interobserver agreement of examination results with those of pediatric ophthalmologists. Variations in care might lead to important differences in visual outcomes and costs. Gathering evidence related to differences in the threshold for diagnosis or treatment is difficult, however, because it is not feasible to have any individual neonate undergo repeated examinations by multiple different provider types.
Nearly all of the neonatologists in this study were concerned about a lack of follow-up for ROP screening examinations after discharge. Small studies have described how transfer to another NICU and hospital discharge are particularly risky times for missed eye care, most likely because the need for follow-up eye care is not communicated, appointments are not scheduled, or parents do not follow up.8–10 We think that this is a good first step for developing interventions to ensure appropriate follow-up care.
The true impact of differences in NICU screening criteria and the impact of the reported barriers to care are unknown. Although almost all of the respondents thought that patients received timely ROP screening in their NICUs, some thought that children in their state might have developed preventable visual impairment because of delayed care. We did find evidence that eye care workforce shortages affected health care utilization negatively by delaying transfer to other NICUs or discharge to home.
An important limitation of this study is that we surveyed neonatologists shortly after the recommendations were changed from including infants with a gestational age of 32 weeks to including infants with a gestational age of 30 weeks. Practice patterns may change as neonatologists and ophthalmologists become more familiar with the guidelines. We think that future research is needed to understand how care is provided at the child level, as well as the relationship of such care to health outcomes and costs. Such data will be necessary to develop systems and to ensure appropriate resources, including an adequate number of appropriately trained pediatric eye care specialists, to identify and to treat ROP efficiently and effectively.
|
ACKNOWLEDGMENTS |
|---|
Financial support was provided by the National Eye Institute (grant K23-EY14023).
|
FOOTNOTES |
|---|
Accepted Apr 11, 2007.
Address correspondence to Alex R. Kemper, MD, MPH, MS, North Pavilion, 2400 Pratt St, Room 0311, Terrace Level, Durham, NC 27705. E-mail: alex.kemper{at}duke.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- American Academy of Pediatrics, Section on Ophthalmology, American Academy of Ophthalmology, American Association for Pediatric Ophthalmology and Strabismus. Errata: screening examination of premature infants for retinopathy of prematurity.
Pediatrics. 2006;118
:1324
[Free Full Text] - American Academy of Pediatrics, Section on Ophthalmology. Screening examination of premature infants for retinopathy of prematurity.
Pediatrics. 2001;108
:809
–811
[Abstract/Free Full Text] - American Academy of Pediatrics, Section on Ophthalmology, American Academy of Ophthalmology, American Association for Pediatric Ophthalmology and Strabismus. Screening examination of premature infants for retinopathy of prematurity.
Pediatrics. 2006;117
:572
–576
[Abstract/Free Full Text] - Ahmed MA, Duncan M, Kent A, NICUS Group. Incidence of retinopathy of prematurity requiring treatment in infants born greater than 30 weeks' gestation and with a birth weight greater than 1250 g from 1998 to 2002: a regional study. J Paediatr Child Health. 2006;42 :337 –340[CrossRef][Web of Science][Medline]
- Castillo-Riquelme M, Lord J, Moseley MJ, Fielder AR, Haimes L. Cost-effectiveness of digital photographic screening for retinopathy of prematurity in the United Kingdom. Int J Technol Assess Health Care. 2004;20 :201 –213[CrossRef][Web of Science][Medline]
- Ho SF, Mathew MR, Wykes W, Lavy T, Marshall T. Retinopathy of prematurity: an optimum screening strategy. J AAPOS. 2005;9 :584 –588[Medline]
- Altersitz K, Piechocki M. Survey: physicians being driven away from ROP treatment. Ocular Surgery News US Edition. August 15, 2006. Available at: www.osnsupersite.com/view.asp?rID=18018. Accessed March 29, 2007
- Aprahamian AD, Coats DK, Paysse EA, Brady-McCreery K. Compliance with outpatient follow-up recommendations for infants at risk for retinopathy of prematurity. J AAPOS. 2000;4 :282 –286[Medline]
- Attar MA, Gates MR, Iatrow AM, Lang SW, Bratton SL. Barriers to screening infants for retinopathy of prematurity after discharge or transfer from a neonatal intensive care unit. J Perinatol. 2005;25 :36 –40[CrossRef][Medline]
- Wittchow K. Shared liability for ROP screening. Ophthalmic Risk Manage Dig. 2003;13 :3 . Available at: www.omic.com/new/digest/DigestFall_03.pdf. Accessed March 29, 2007
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Day, A. M. Menke, and R. L. Abbott Retinopathy of Prematurity Malpractice Claims: The Ophthalmic Mutual Insurance Company Experience Arch Ophthalmol, June 1, 2009; 127(6): 794 - 798. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. Mills Retinopathy of Prematurity Malpractice Claims Arch Ophthalmol, June 1, 2009; 127(6): 803 - 804. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Hellstrom, A.-L. Hard, E. Engstrom, A. Niklasson, E. Andersson, L. Smith, and C. Lofqvist Early Weight Gain Predicts Retinopathy in Preterm Infants: New, Simple, Efficient Approach to Screening Pediatrics, April 1, 2009; 123(4): e638 - e645. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Dhaliwal, E Wright, C Graham, N McIntosh, and B W Fleck Wide-field digital retinal imaging versus binocular indirect ophthalmoscopy for retinopathy of prematurity screening: a two-observer prospective, randomised comparison Br J Ophthalmol, March 1, 2009; 93(3): 355 - 359. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Kemper, D. K. Wallace, and G. E. Quinn Systematic Review of Digital Imaging Screening Strategies for Retinopathy of Prematurity Pediatrics, October 1, 2008; 122(4): 825 - 830. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
