Prospective study of factors influencing timely versus delayed presentation of preterm babies for retinopathy of prematurity screening at a tertiary eye hospital in India The Indian Twin Cities ROP Screening (ITCROPS) data base report number 6.

Purpose: To evaluate the factors influencing timely versus delayed presentation of preterm babies for ROP evaluation.
Methods: Preterm babies (≤35 weeks gestational age, ≤2000 g birth weight) were prospectively included in the study. Timely presentation was defined as babies who presented for the first ROP screening within 30 days of birth and Delayed as more than 30 days of birth. An event survey to assess factors influencing timely vs delayed presentation was administered to parents/guardian of babies after obtaining informed consent.
Results: Data of 278 preterm babies (n = 139 timely vs n = 139 delayed presentation) collected in the event surveys were analyzed. The delayed presenters came at a median duration of 6.3 weeks (1st and 3rd quartiles: 5.3 and 9.1) after birth. The odds of any stage of ROP was 2.6 times and the odds of sight threatening ROP was 6.8 times in those presenting delayed compared to those presenting timely. Major Reasons for delayed presentation were not asked to do so/no referral from pediatrician in 64 (46%) participants and unaware of the importance by 46 (33%) participants.
Conclusion: Deviation from screening protocol is an important modifiable risk factor in ROP screening. The study findings suggest the need for creating awareness about timely screening and referral guidelines among the pediatricians involved in "care" of preterm infants at risk of developing ROP.

Retinopathy of prematurity (ROP) is an important avoidable cause of childhood blindness. Currently, middle income countries like India and China are experiencing a third epidemic of ROP blindness due to increased survival of preterm babies with presence of variable neonatal Intensive care and ophthalmic coverage for screening. In 2010, the annual incidence of blindness and visual impairment from ROP was estimated to be 32,200 cases worldwide. India accounted for nearly 10% of all estimated worldwide visual impairment following ROP in 2010, with at least 5,000 developing severe disease and 2,900 children surviving with visual impairment related to ROP.[1]

Timely screening is of paramount importance to avoid blindness due to ROP. The success of any screening program depends on the active participation of child care clinicians in identifying the population at risk and for appropriate and timely referral to the ophthalmologist. This goal can be achieved on a sustainable basis with creation of neonatologists and ophthalmologist networks like NO ROP[2] and ITCROPS (Indian twin cities ROP study).[3] Our ITCROPS prospective database, ongoing since 1998, provided an opportunity to create screening and management guidelines applicable to the Indian population.[345] Timely screening can be achieved only when appropriate screening guidelines are enforced through a planned screening strategy at Neonatal Intensive Care Units (NICUs) in partnership with ROP-trained Ophthalmologists.

While risk factors for ROP occurrence are well known and exhaustively reported in literature.[67891011] There is paucity of information about barriers influencing timely presentation of preterm babies for ROP screening. Hence, the present study was undertaken to find factors influencing Timely vs Delayed presentation of babies for ROP screening to a tertiary eye hospital in South India.

Methods

All preterm babies (New and follow up cases referred by ophthalmologists) who presented for ROP evaluation to a tertiary eye institute in south India were prospectively included in the study. Preterm was defined as gestational age at birth ≤35 weeks and/or birth weight <2000 g as per the Indian national neonatology forum (NNF) guidelines for ROP screening in India.[12] The study was approved by the Institutional Review Board and all babies were also included in our ongoing computerized ITCROP data base.

The study was designed as a prospective cross-sectional cohort study where in the enrolled children were followed after collecting “risk factor” data to see how many developed ROP/sight threatening ROP using an event survey form. Timely presentation was defined as babies who presented for the first ROP screening to any Ophthalmologist within 30 days of birth and Delayed presentation as babies who presented for the first ROP screening at more than 30 days of birth. The primary outcome measure was finding the Demographic/Social/Environmental/parental and medical factors in timely vs delayed presentation. The secondary outcome of the study was to estimate the proportion of sight threatening ROP/ROP blindness in timely versus late presentations. Causes of Timely vs Delayed presentation between both groups were analyzed.

An event survey was administered by one of the study examiners to the parents/guardians of the baby after getting a signed informed Consent. This survey was administered (appendix) while the baby was getting dilated for examination of the fundus. The event survey included details regarding the birth of the child, socio economic status, awareness of ROP, referral pattern, cause of delayed/early presentation, and past treatment received for ROP, if any. Most of the questions were close ended except for the one regarding the late referrals. Details were included only if written material was available as evidence to collaborate the data given. Babies where written details of admission, discharge, birth, and follow-up were not available, were excluded from the study. Sample size of 280 patients (140 in each group of Timely and delayed presentation) was calculated assuming the timely presentation of ROP cases being 80% with 20% effect size and at 5% level of significance.

Statistical analysis

Descriptive statistics included mean and standard deviation for normally distributed continuous variables and median and quartiles for non-normally distributed continuous variables. Students’ t test and Wilcoxon ranksum test was used to compare normal and non-normal continuous variables respectively and chi square test was used for categorical variables. Logistic regression analysis was also performed to evaluate the odds of ROP and sight threatening ROP in those presenting delayed as compared to those presenting timely. A P value of ≤0.05 was considered statistically significant. Statistical analyses were performed using commercial software (Stata ver. 11.1; StataCorp, College Station, TX).

Results

Data of all the 278 preterm babies (n = 139 timely vs n = 139 delayed presentation) collected in the event survey were analyzed. Table 1 summarizes the demographic, social, environmental, parental, and medical factors between timely versus delayed presentation groups. The delayed presenters came at a median duration of 6.3 weeks (1st and 3rd quartiles: 5.3 and 9.1) after birth. There was no statistically significant difference in the distribution of gender between those who presented early versus late. Babies presenting delayed also had lower gestational age and lower birth weight than those presenting timely; however, this difference was not clinically significant. Location of birth was more commonly rural in those who presented delayed. Source of awareness was more commonly the pediatrician in those who presented timely. Sight threatening ROP was seen in significantly more babies who presented late. The number of sight threatening ROP found in babies with late presentation was 25 compared to 3 found in timely group with P value of < 0.001, which was statistically significant [Table 1].

Table 1

Comparison of the demographic, social, environmental, parental and medical factors between timely versus delayed presentation groups

Variables	Timely presentation (n=139)	Delayed presentation (n=139)	P
Gestational age (weeks)	31.1±1.9	30.4±2.1	0.004
Birth weight (grams)*	1700 (1350, 1800)	1500 (1200, 1700)	0.001
Post menstrual age (weeks)*	34 (33, 36)	38 (37, 40)	<0.001
Gender (male: female)	35:104	48:91	0.09
Location of birth
Urban	128 (92.1%)	117 (84.2%)	0.04
Rural	11 (7.9%)	22 (15.8%)
Source of awareness
Pediatrician	1 (0.7%)	9 (6.4%)	0.001
Relative	0 (0%)	12 (8.6%)	0.01
Friend	0 (0%)	1 (0.7%)	<0.001
Retinal status at first presentation
Normal retina	128 (92.1%)	105 (75.5%)	<0.00
Sight- threatening ROP	3 (2.2%)	25 (18.0%)	<0.001

*Median with first and third quartiles in brackets

Table 2 shows the results of the logistic regression models evaluating the effect of delayed presentation on the presence of ROP and sight threatening ROP. Delayed presentation increased the odds of both ROP and sight threatening ROP independent of gestation age and birth weight. The odds of any stage of ROP was 2.6 times and the odds of sight threatening ROP was 6.8 times in those presenting delayed compared to those presenting timely.

Table 2

Multivariate analysis of Risk factors for the presence of any ROP and sight threatening ROP in preterm babies as compared to No ROP

Variable	Any ROP			Sight threatening ROP
	OR	95% CI	P	OR	95% CI	P
Gestational Age	0.78	0.64-0.96	0.02	0.76	0.58-0.99	0.04
Birth weight	0.99	0.99-1.00	0.001	0.99	0.99-1.00	0.001
Location of birth (urban/rural)	1.90	0.75-4.82	0.18	2.59	0.87-7.72	0.09
Timely vs Delayed Presentation	2.64	1.21-5.78	0.02	6.81	1.88-24.71	0.004

Based on the logistic regression models, Fig. 1 shows the predicted probability of sight threatening ROP in babies presenting early (timely) and delayed. The probability of sight threatening ROP which was 1.2% (95% CI: 0.3 to 4.2) in those presenting timely increased to 8.0% (95% CI: 4.1 to 15.3) in those presenting delayed. Fig. 2 shows predicted probability of presenting with sight threatening ROP based on post conceptional age (Gestational age + chronological age) at presentation after adjusting for the effect of birth weight.

Figure 1

Predictive probability of presenting with Sight threatening ROP in timely Vs Delayed presentation (adjusted for effect of gestational age and birth weight)

Figure 2

Predictive probability of presenting with Sight threatening ROP based on postconceptional age (adjusted for effect of gestational age and birth weight)

It shows that with increased postconceptional age, the severity of ROP significantly increases. Predicted probability of severe ROP increased from 2.1% (95% CI: 0.8 to 5.1) at a post conceptional age of 29 weeks to 5.2% (95% CI: 2.7 to 9.7) at a post conceptional age of 39 weeks and 12.4% (95% CI: 5.8 to 24.7) at a post conceptional age of 49 weeks.

The major reasons for delayed presentation of babies for ROP screening was not asked to do so/no referal from paediatrician was told by 64 (46%) participants and unaware of the importance by 46 (33%) participants. Other reasons mentioned for delayed screening included baby was sick by 16 (12%), busy in a social event by 11 (9%) and due to superstition by 1 partcipant.

Discussion

Our study findings also indicate that the major modifiable factor in a ROP screening program is non-adherence to timely screening protocol whereas other factors like preterm delivery, gestational age, birth weight, and of preterm babies cannot be modulated to have favorable outcome.

When we looked at the causes of delayed presentation of babies “not asked to do so” (ascertained from the written notes of the child care clinicians) was the reason in a majority (65%). ‘were told/was written in the discharge summary but were not aware of the importance of the timing was the next important cause (46%).” These data show the importance of creating awareness about timely screening and referral guidelines among all stakeholders. The pediatricians should effectively communicate the need for timely screening and referral guidelines with the parents. The stakeholders include pediatricians, parents, neonatal nurses, ophthalmologists, and also family physicians and the public at large. Recently Indian national neonatology forum guidelines were developed and distributed to promote the same among neonatologists.[12] The pediatric society of India is also developing similar guidelines.

Indian twin cities Retinopathy of Prematurity Screening (ITCROPS) database review clearly indicate the need for 20-day screening strategy for babies with gestational age less than 30 weeks and 30 day screening strategy for babies with gestational age more than 30 weeks.[13] This strategy is important to detect sight threatening ROP, especially Aggressive posterior ROP (APROP), on time.

Our present study found that the number of sight threatening ROP found in babies with late presentation was 25 (with P value of <0.001 – Table 1), which was statistically significant. The odds of any stage of ROP was 2.6 times and the odds of sight threatening ROP was 6.8 times in those presenting delayed compared to those presenting timely [Table 2]. These findings suggest the important need for timely screening and appropriate referral to prevent needless blindness due to ROP. Our study data based predictable risk models based on the time of presentation and post conceptional age [Figs. 1 and 2] clearly shows the importance of timely screening to prevent blindness from ROP. Azad et al. also reported lack of screening as major cause of stage 5 ROP blindness in a tertiary eye center setup from India. They found that out of 115 babies presented with stage V ROP, 103 (89.6%) babies never had eyes examined for ROP screening before.[14]

Though ROP care is offered by ophthalmologists of different training backgrounds there needs to be a uniform approach to screening and referral.[3] There are variations in how children are identified for ROP screening and how screening and treatment is provided. These suggest that the workforce should be adequately trained to screen, refer or treat if indicated.[4]

The awareness of ROP screening among pediatricians is better in metro cities but is poor in small towns as detected by 2 studies from South India.[1516] Lack of awareness of ROP and lack of proper referral guidelines was observed in a study conducted in Pakistan.[17]

A coordinated strategy by establishing and monitoring the implementation of national guidelines for screening of ROP and to identify causes for failure of compliance has been successfully done by Ziakas NG, et al.[2] Their audit of adherence to ROP screening guidelines revealed that compliance for adherence improved from 47% (262 out of 558 babies) in the first cycle to 73% (264 out of 360 babies) in the second cycle.

Babies transferred between units, discharged home before screening, or who failed to qualify for screening on one of the two defined criteria (gestational age and birth weight), were more likely to be missed.[2]

Bain, et al.[13] reported that in California state (United states of America) higher gestational age (OR = 1.25 for increase of 1 week, 95% CI, 1.21-1.29), higher birth weight (OR = 1.13; 95% CI, 1.10-1.15), and singleton birth (OR = 1.2; 95% CI, 1.07-1.34) were associated with higher probability of missing ROP screening. Level II neonatal intensive care units and neonatal intensive care units with lower volume were more likely to miss ROP screenings. When we look at our study from developing country, lower gestational age, low birth weight seems to be the factors associated with delayed presentation for ROP screening. These high-risk babies with possible prolonged hospital stay might be missing the important ROP screening due to possible lack of NICU based ROP screening program. Vinekar et al. reported improved compliance to ROP screening with early screening before baby discharge from NICU. This early screening protocol of babies before discharge instead of conventional screening protocol of 3-4 weeks allowed less communication gaps and better compliance for follow-up screening visits.[18]

There is a need for stressing the importance of timely screening to prevent blindness by the treating pediatrician to the parents. As the baby may be healthy the parents can overlook the blinding condition. Some parents don’t give the required importance as they are busy in some social event as evident from our study. This attitude needs to be changed by distributing pamphlets, attaching posters on the walls and employing counselors at the place of birth so that the message is passed on to the parents in their own language. Vinekar et al. found that the reasons related to travel logistics and awareness, knowledge, and attitude were the two main categories of reasons for dropout for ROP screening.

They also reported in gender bias (female baby so did not bring) for ROP screening. In present study no gender difference noted between babies with timely/late presentation. They have incorporated several important measures such as making parents aware of using government scheme vehicles to transport babies for screening, providing financial reimbursement especially those babies who can not afford treatment, obtaining two alternative contact mobile numbers of parents, providing them a discharge card with scheduled ROP screening date clearly, providing the information through a project manager who acts as a liaison between the neonatologist and parents etc., By implementing these measure they found that, there was a reduction in attrition from 20.8% to 3.8% (p < 0.001) in two time periods of their tele imaging based rural ROP screening model. Pediatricians’ role in ensuring follow-up as a reason for dropout remained an important persisting reason in both time periods of their study and our present study as well.[19]

The limitations of the study were that some of the late referrals had advanced disease and could have led to selection bias. The case control study design would help to overcome some of this bias. One of the study examiners (aware of the study hypotheses) administered the event survey form. This might have led to information bias. The ascertainment of referral was based a written document, either a discharge card or a reference letter and so the authenticity of other referrals who had none could not be proved. Another important limitation of the study was that we evaluated the barriers for presentation of infants for ROP screening at a tertiary exclusive ophthalmic institute setup. Hence, the study results cannot be extrapolated to outcomes of a NICU-based ROP screening programs where in ophthalmologists will be examining infants at the NICU timely instead of expecting them “to come to them” in their eye clinics for ROP screening. In comparison to an institute-based ROP screening facility, the NICU based (urban) and outreach based (rural) teleimaging ROP models have several advantages such as better yield of screening, detection of sight threatening ROP and preventing needless blindness due to ROP.[2021] There is a definite role for tertiary eye centers to actively promote, implement, and expand NICU and outreach based ROP screening models to reduce ROP blindness.[2022]

Conclusion

To conclude, awareness creation about timely ROP screening protocol and referral guidelines among all stakeholders involved in preterm baby care is essential to prevent needless blindness due to ROP.

Financial support and sponsorship

Funded by Unrestricted grants from Hyderabad Eye Institute.

Conflicts of interest

There are no conflicts of interest.