Do Cost Concerns Limit Screening for Retinopathy of Prematurity?

Do Cost Concerns Limit Screening for Retinopathy of Prematurity?


Retinopathy of prematurity (ROP) screening engenders angst in the ophthalmology community. Timely identification of infants requiring intervention is highly effective in averting a lifetime of blindness.1-9 Conversely, failing to effectively screen an at-risk infant may have disastrous consequences. This has created an environment in which it is unacceptable to have a poor anatomic or functional outcome in the absence of appropriate screening for ROP. The end result is that medicolegal liability often extends into several millions of dollars for cases where lack of appropriate ROP screening is the proximate cause of vision loss.

This concept of appropriate ROP screening has received much attention recently with the advent of telemedicine techniques for ROP screening. A number of converging events have conspired to bring the debate to the forefront of the field. In this article, we will discuss the factors that have created, if not a crisis in ROP screening, then at least a crunch, and we will examine possible solutions. Too often, an ophthalmologist vs neonatal intensive care unit (NICU) mentality is brought to bear. We must remember that our mandate is to prevent blindness, and only by working in close cooperation with our colleagues in the NICU can we seek to eradicate preventable vision loss in this vulnerable population.


Is there an ROP screening crisis in the United States? In 2006, the Joint Statement of the American Academy of Pediatrics, American Academy of Ophthalmology (AAO), and American Association for Pediatric Ophthalmology and Strabismus was significantly revised to update the eligibility, initiation, frequency, and termination ROP screening criteria.10 Significantly, the eligibility criteria was expanded to include infants 30 weeks postmenstrual age (PMA), up from 28 weeks PMA (Table 1).10 Previously, the 1500-g birthweight criteria was the main driver for eligibility.

Looking at final data from the National Vital Statistics Reports (see reference 1, Table 32 in the reference) in 2005 in which 4138349 live births were recorded, when applying the new criteria, the 30-week PMA inclusion would be the main driver, increasing the number of eligible babies from approximately 59000 (birthweight of 1499 g and lower in 2005) to approximately 82000 g (PMA 28 to 31 weeks), an increase of 38% brought about by the stroke of a pen.11 These infants would commence ROP screening earlier (31 weeks PMA or 4 weeks chronological age, whichever is later), have more frequent examinations (Table 2), and undergo a longer duration of screening (Table 3).10

Table 1. Eligibility Criteria for ROP Screening
• Birthweight (BW) <1500 grams, or
• Postmenstrual age (PMA) ≤30 weeks, or
• 1500<BW<2000, or PMA >30 weeks and felt to be high-risk by pediatrics team

Table 3. Termination Criteria for Acute Phase ROP Screening
• Zone 3 retinal vascularization without previous zone 1 or 2 ROP
• Full (mature) retinal vascularization
• Postmenstrual age of 45 weeks and no prethreshold disease (zone 1, any ROP OR zone 2, stage 3 ROP)
• Regression (in the absence of abnormal vascular tissue capable of reactivation or progression)

The AAO commissioned a survey in 2006 of its members who provided ROP screening and learned that, of those presently doing so, only 77% anticipated continuing screening in the future ( The main reasons for discontinuing screening all centered around financial considerations (Table 4). To summarize, the data indicate a 38% increase in eligible patients for ROP screening with an earlier initiation, more frequent examinations, and longer duration of screening at the same time as a 23% reduction in willing and eligible ROP screeners. If not a crisis, this represents an area of concern for the United States as its population of very preterm infants has increased by over 12.15% since 1981 and of very low birthweight infants has increased by 28.4% since 1981.11

Darius M. Moshfeghi, MD, is assistant professor of ophthalmology at Stanford University School of Medicine in Stanford, CA. Antonio Capone, Jr., MD, practices retinal medicine with Associated Retinal Consultants at the William Beaumont Hospital in Royal Oak, MI. Dr. Moshfeghi reports minimal financial interest in Clarify Medical Systems. Dr. Capone reports no financial interests. Dr. Moshfeghi can be reached via e-mail at


There are 2 key elements to the Joint Statement Guidelines: (1) technique, namely binocular indirect ophthalmoscopy (BIO) in conjunction with pupillary dilation; and, (2) qualifications, ie, "…performed by an ophthalmologist who has sufficient knowledge and experience to enable accurate identification of the location and sequential retinal changes of ROP."10 Too often in the current debate, the focus has been on whether telemedicine options are as sufficient as BIO for ROP screening. Rarely is the second — and more important, we would argue — criterion brought up for discussion, ie, that of an ophthalmologist experienced not in ROP screening but rather in the disease progression of ROP, which are 2 different things entirely. In the aforementioned AAO survey, 18% of ROP screeners were general ophthalmologists, the rest either being pediatric ophthalmologists or retina specialists.

While data are lacking, it is safe to assume that the vast majority of ROP screening in the United States employs the BIO technique. All considerations of efficacy aside, BIO suffers from a severe lack of leverage: BIO links an ophthalmologist to a single NICU. If he or she is busy screening in NICU #1, then that ophthalmologist cannot be providing screening services 50 miles away at NICU #2. When ROP screeners are plentiful, this is not a problem. However, when financial considerations are driving ROP screeners out of business in droves, the ability to leverage ROP experts for screening purposes is a highly prized commodity.

So what do we know about BIO for ROP screening? We do not know much, it turns out. Data are lacking on the efficacy of BIO as it relates to interobserver and intraobserver variability. BIO technique relies on schematic drawings using the International Classification of ROP standardized fundus drawings that demonstrate the zones of the disease.12-14 When employing BIO, an ophthalmologist's interpretation of fundus features are translated into depictions of the zone, stage, extent, and presence of plus or preplus disease. These drawings have never been evaluated for reliability or accuracy; they have been crowned as the de facto gold standard in the absence of any scientific evaluation. Nowhere else in medicine is a drawing deemed to be a sufficient indicator of the health of an organ system. Indeed, within ophthalmology itself, fluorescein angiography, the gold standard for evaluation of age-related macular degeneration and diabetic retinopathy, has demonstrated significant inter- and intraobserver variability, even among highly trained reading-center specialists.15-17 There is no reason to believe that the situation is different for BIO-derived drawings, and there are many reasons to believe that it is worse.

What about telemedicine for ROP screening? First, we must state that we are specifically talking about employing the RetCam (Clarity Medical Systems, Pleasanton, CA) family of contact wide-angle fundus-viewing systems (Figure 1). No other system is widely deployed or tested or offers a comparable field of view. Ironically, telemedicine has been rigorously evaluated because of the historical bias in favor of BIO. The Photographic Screening for ROP (Photo-ROP) trial employed a side-by-side, prospective, masked comparison of telemedicine vs BIO.18 The Photo-ROP trial found that for ETROP type 1 patients (ie, treatment-warranted ROP), telemedicine had a sensitivity of 42.3%, specificity of 97.8%, positive predictive value of 92%, and negative predictive value of 75%.19 These results suggested that telemedicine was an effective strategy for identifying those patients who would need treatment.

More recently, the results of the Stanford University Network for Retinopathy of Prematurity (SUNDROP) telemedicine initiative for ROP screening have been reported.20 In SUNDROP, telemedicine is used as the sole arbiter for identification of treatment warranted (TW)-ROP. If TW-ROP is identified, a bedside consultation with BIO should be performed. The SUNDROP network has been in existence for almost 3 years, with 24-month data having been presented at the podium. To date, sensitivity and specificity exceed 99%, with a negative predictive value of 100%. These real-world results validate two things: (1) telemedicine is a highly effective method for identification of TW-ROP; and (2) the experience of the examiner in the technique of telemedicine, as well as the ROP disease progression, has an impact on the results (ie, there is a moderately steep learning curve).

The advantages of telemedicine relative to BIO are straightforward and not inconsequential: (1) longitudinal review; (2) hard-copy images for the chart and education; (3) medicolegal protection; (4) leverage of the ophthalmologist experienced in ROP disease progression to screen multiple NICUs in a short period without travel; and (5) ability for consultation with colleagues based upon the photographs.


Historically, NICUs have gotten a free ride on ROP screening, largely thanks to the altruism of the ophthalmology community. By this, we mean that ophthalmologists either agreed to screen for negligible or no charges or they arranged an annual retainer that did not cover their time or their expenses. When successful treatment became available following the Cryotherapy for ROP trial and, subsequently, with the introduction of the diode laser, the risk increased for both parties because of the very success engendered by these therapies. By risk, we are strictly referring to medicolegal liability. This introduction of high, sometimes outlandish, settlements for poor outcomes resulted in an initial culling of willing and experienced ROP screeners. NICUs became more aggressive in their recruitment of physicians — either through the carrot of offering more compensation or the stick of restricting access to the operating rooms — while physicians ran for the nearest exit. This brings us to 2008.

These are the facts, and they are indisputable. All NICUs are required to provide ROP screening services to eligible patients; failure to do so may result in loss of accreditation. While ophthalmologists are not similarly held to regulatory statutes, we have a moral obligation ingrained in our profession that mandates we provide these services for this at-risk population. Having said that, ophthalmologists should receive fair compensation for their services. There are three basic screening scenarios: (1) BIO, (2) telemedicine, and (3) no screening. Unfortunately, many small NICUs have interrupted or no coverage for much of the year for reasons enumerated above. Even in large NICUs, there are times during the year when coverage is spotty or suboptimal (eg, an inexperienced screener covering for the main ROP screener). With NICUs being more aggressive in their expansion plans, this problem is likely to be exacerbated.

Michael Trese, MD, performs screening for retinopathy of prematurity.

Regardless of whether BIO or telemedicine is employed, the compensation options are similar. Physicians can bill for their services directly using either Evaluation and Management coding or Eye codes. This direct billing is the least expensive option for the NICU and the most expensive option for the ophthalmologist, as reimbursement for well-baby examinations is low in general and even lower in the Medicaid population. Most importantly, if a physician were to be named in a lawsuit, there would be no malpractice abatement in this scenario. Alternately, the physician may arrange a fee-for-service contract with the hospital that pays the physician a fixed amount per examination and another fixed amount for intervention, independent of the patient's insurance. While this may represent a short-term gain for the physician in terms of higher reimbursement, perhaps offsetting the time lost in clinic or away from family, it still does little to mitigate a multimillion-dollar settlement. In either scenario, the physician may arrange to have hospital-based liability coverage and travel expenses.

The third scenario is the retainer-based model. In the retainer model, the physician negotiates a fixed fee with the NICU for screening services for a given period of time (usually a year). The physician then performs all screening for that fee. Depending on the circumstances, separate malpractice coverage, surgical fees, and transportation fees may apply. The major differences between BIO and telemedicine as relates to these contracts are the following: (1) The NICU is liable for more expenses related to physician travel and time in the BIO scenario; (2) The NICU must pay an up-front fee in telemedicine to purchase a camera, as well as annual maintenance and upgrade expenses; and (3) The ophthalmologist has greater flexibility regarding his or her schedule with the telemedicine option.

Both the NICU and the physician must note that, for telemedicine screening, the camera is inadequate to meet the termination criteria. Therefore, either the NICU and the ophthalmologist must come to an accommodation regarding what to do for an infant who may potentially be discharged, or the NICU must find someone to perform an examination that satisfies this requirement. Additionally, the Ophthalmic Mutual Insurance Company has taken the position that, if telemedicine is to be performed for screening of ROP, then its insured physicians must examine the patient by BIO at least once prior to discharge. Needless to say, this eliminates much of the advantage for the ophthalmologist, as patients will be discharged in a continuous fashion throughout the week.

Courses are offered at subspecialty meetings offering to teach ROP screeners how to negotiate with NICUs. We believe that both sides need education, because most of the vision loss from ROP is preventable. Just because the NICU has a regulatory requirement to provide ROP screening services, and there is an increasingly eligible ROP screening population coinciding with a decreasing ROP screener population, predatory pricing is not the solution. We do not advocate hardball tactics when negotiating with a NICU.

Instead, every effort should be made to educate the NICU about the real costs to the physician: (1) the comparable amount of income that could be earned in clinic for the amount of time devoted to screening; (2) the cost of liability coverage in terms of dollars; (3) the psychological toll of liability that may extend between 18 and 21 years following screening; and (4) the time spent in travel and/or on nights on weekends for screening services. Based on an honest conversation of the real cost of screening, an accommodation can usually be reached that is amenable to all parties. One must keep in mind that, for NICUs, the ROP screening cost is a relatively new expense (and often times a new large expense).


Retinopathy of prematurity screening is entering a transitional period on several fronts. Now that effective therapies exist, screening has assumed the primary role in prevention of blindness from ROP. Telemedicine offers a potential solution for the present crisis in ROP screening, leveraging the few remaining willing screeners experienced in the progression of ROP. For our goal of preventing all vision loss from ROP to be realized, ophthalmologists and NICUs must learn to cooperate as a team. Toward this end, we advocate education over hardball techniques when negotiating a screening contract. We have to remember what is in the best interests of these helpless infants. RP


  1. Coats DK, Miller AM, Brady McCreery KM, et al. Involution of threshold retinopathy of prematurity after diode laser photocoagulation. Ophthalmology. 2004;111:1894-1898.
  2. Coats DK, Miller AM, Hussein MA, et al. Involution of retinopathy of prematurity after laser treatment: factors associated with development of retinal detachment. Am J Ophthalmol. 2005;140:214-222.
  3. Rezai KA, Eliott D, Ferrone PJ, Kim RW. Near confluent laser photocoagulation for the treatment of threshold retinopathy of prematurity. Arch Ophthalmol. 2005;123:621-626.
  4. Connolly BP, Ng EY, McNamara JA, et al. A comparison of laser photocoagulation with cryotherapy for threshold retinopathy of prematurity at 10 years: part 2. Refractive outcome. Ophthalmology. 2002;109:936-941.
  5. Ng EY, Connolly BP, McNamara JA, et al. A comparison of laser photocoagulation with cryotherapy for threshold retinopathy of prematurity at 10 years: part 1. Visual function and structural outcome. Ophthalmology. 2002;109:928-934; discussion 35.
  6. Connolly BP, McNamara JA, Sharma S, et al. A comparison of laser photocoagulation with trans-scleral cryotherapy in the treatment of threshold retinopathy of prematurity. Ophthalmology. 1998;105:1628-1631.
  7. Axer-Siegel R, Snir M, Cotlear D, et al. Diode laser treatment of posterior retinopathy of prematurity. Br J Ophthalmol. 2000;84:1383-1386.
  8. Banach MJ, Ferrone PJ, Trese MT. A comparison of dense versus less dense diode laser photocoagulation patterns for threshold retinopathy of prematurity. Ophthalmology. 2000;107:324-327; discussion 8.
  9. Multicenter Trial of Cryotherapy for Retinopathy of Prematurity: ophthalmological outcomes at 10 years. Arch Ophthalmol. 2001;119:1110-1118.
  10. Section on Ophthalmology American Academy of Pediatrics; 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. Erratum in: Pediatrics. 2006;118:1324.
  11. Martin JA, Hamilton BE, Sutton PD, et al; Centers for Disease Control and Prevention National Center for Health Statistics National Vital Statistics System. Births: final data for 2005. Natl Vital Stat Rep. 2007;56:1-103.
  12. Committee for the Classification of Retinopathy of Prematurity. An International Classification of Retinopathy of Prematurity. Arch Ophthalmol. 1984;102: 1130-1134
  13. ICROP Committee for Classification of Late Stages ROP. An international classification of retinopathy of prematurity, II: the classification of retinal detachment. Arch Ophthalmol. 1987;105:906-912.
  14. International Committee for the Classification of Retinopathy of Prematurity. The International Classification of Retinopathy of Prematurity revisited. Arch Ophthalmol. 2005;123:991-999.
  15. Friedman SM, Margo CE. Choroidal neovascular membranes: reproducibility of angiographic interpretation. Am J Ophthalmol. 2000;130:839-841.
  16. Holz FG, Jorzik J, Schutt F, et al. Agreement among ophthalmologists in evaluating fluorescein angiograms in patients with neovascular age-related macular degeneration for photodynamic therapy eligibility (FLAP-study). Ophthalmology. 2003;110:400-405.
  17. Kaiser RS, Berger JW, Williams GA, et al. Variability in fluorescein angiography interpretation for photodynamic therapy in age-related macular degeneration. Retina. 2002;22:683-690.
  18. Balasubramanian M, Capone A, Jr, Hartnett ME, et al. The Photographic Screening for Retinopathy of Prematurity Study (Photo-ROP): study design and baseline characteristics of enrolled patients. Retina. 2006;26(Suppl): S4-S10.
  19. Photographic screening for retinopathy of prematurity (Photo-ROP) cooperative group. The photographic screening for retinopathy of prematurity study (Photo-ROP): Primary Outcomes. Retina. 2008;28(Suppl):S47-S54.
  20. Murakami Y, Jain A, Silva R, Lad E, Moshfeghi D. Stanford University Network for Diagnosis of Retinopathy of Prematurity (SUNDROP): 12-month experience with telemedicine screening. Br J Ophthalmol. 2008 Aug 14. [Epub ahead of print].