Introduction

The Certifications of Visual Impairment (CVI) system was implemented in 2005 and 2007 to quantify and classify the incidence of visual impairment amongst the England and Wales population respectively. The CVI replaced the BD8 form which was used prior this. CVIs can only be completed be Consultant Ophthalmologists.

The CVI requires formal documentation of visual acuity, visual field, and the aetiology. There are guidelines issued by the Royal College of Ophthalmologists, UK. Consent is required from the patient/legal guardian to allow data to be shared/registered with the local authority, GP and certifications office. Completion of a CVI allows the child’s visual impairment to be legally identified as a disability, thus providing formal documentation allowing the local council to offer formal support, such as access to an Educational Health Care Plan. It also allows data collection to allow monitoring of trends and auditing of visual impairment in England and Wales.

Visual impairment is classified as either as ‘Sight Impaired’ (SI) (previously ‘partially sighted’ on the older BD8 form) or ‘Severely Sight Impaired’ (SSI) (previously ‘blind’ on the older BD8 form). The criteria for registration is based on best corrected visual acuity and visual field (Table 1).

Table 1 Criteria for Visual Impairment registration as per the Department of Health and Social Care, United Kingdom.

Following the implementation of the CVI, there has been an increase in the incidence rate for new certifications of visually impaired children across England and Wales. Bunce et al. concluded that 2014/15 certifications were statistically significantly more than new certifications in 1999/2000 and 2007/8 [1]. Hereditary retinal disorders were identified as the most significant singular group of conditions across both SI and SSI groups to account for this increase. Amongst certifications, disorders of the visual cortex, nystagmus and albinism were identified as very significant contributors to paediatric visual impairment. It is important to note prior to use of the CVI, the BD8 system was found to have bias towards permanent central visual impairment rather than conditions that affect peripheral vision [2]. According to existing studies, the number of certifications has risen year on year since inception of the CVI [1]. However, there was a significant drop in the months of April–June 2020 reflecting the impact of the global Covid-19 pandemic on clinical services [3].

To the best of our knowledge, currently there is not a large evidence base assessing the year-on-year change in paediatric sight impairment across the span of 10+ years. Previous publications made spotlight comparisons in certification data rather than assessing longitudinal trends across the entire period of data available [1, 4]. This study investigates the longitudinal trend in the certifications of visual impairment amongst children between the ages of 0–15 in England and Wales, to assess trends over time, including underlying aetiology, with a view to assessing changing treatments and impact of guidelines e.g. ROP screening and treatment.

Methods

We performed a retrospective study reviewing all new visual impairment certifications for children aged 0–15 years old from 1st January 2009 until 31st December 2022. The data was sourced from the Certifications Office for England and Wales. All data received was anonymised and grouped based on the severity of visual impairment (SI vs SSI), primary pathological cause of impairment and year the child was certified. Consent for use in future research was obtained upon registration by patient/legal guardians as part of the CVI process. Table 1 demonstrates the criteria for visual impairment registration [5, 6].

The primary pathological cause was pre-assigned by a Consultant Ophthalmologist on initial registration using Part C of the CVI form with ICD 9/10 coding [7]. Official Census data from the UK Office for National Statistics was used to calculate the size of the 0–15 years old population group [8, 9]. Census surveys in England and Wales are only carried out every 10 years; 2011 and 2021. Therefore when not available, official predictions for population groups were used for any statistical measures.

The data captured was further compartmentalised into 11 broad groups of conditions which were similar in nature and pathology: Neoplasms; Metabolic; Disorders of the Central Nervous System (excluding neoplasms); Disorders of globe; Retinal disorders; Uveitis/Scleritis; Anterior Segment disorders (Glaucoma, Corneal and Lens disorders including cataract); Amblyopia; Disorders of optic nerve; Congenital anomalies of the eye including infections & nystagmus; and Acquired insults/injuries.

In groups with >5 new certifications for every year of analysis, incidence rates were calculated. Any groups with <5 new certifications per year were grouped into an ‘Other’ category in order to preserve anonymity. ‘Acquired injuries’ was grouped into the ‘Other’ category despite having >5 new certifications per year due to the broad nature of conditions making up this group; which included traumatic injuries to the brain and eye in addition to perinatal injuries.

Byar’s method was used to calculate incidence rates and confidence intervals across 4 yearly periods, with significant changes in certification rates being assumed if confidence intervals did not cross. The Byar’s method is statistical model recommended by Public Health England when using large population data to calculate incidence rates and confidence intervals [10, 11].

Results

From 2009–2022 there were 18,387 new certifications for visual impairments amongst the paediatric (<16 years old) population. There were 10,915 SI and 7472 SSI total new certifications (Fig. 1 and Table 2). Across the 13-year period of data collected, disorders of the CNS (4145), retinal disorders (2990) and congenital abnormalities of the eye (4491) were assessed as being the most significant causes of visual impairment amongst the paediatric population in England and Wales. The top 3 SI certifications were ICD 362.7 Hereditary Retinal Dystrophies (n = 1305), ICD 270.2 Albinism (n = 1020), and ICD 377.7 Disorders of the visual cortex (n = 739). The top 3 SSI certifications were ICD 362.7 Hereditary Retinal Dystrophies (n = 1042), ICD 377.7 Disorders of the visual cortex (n = 858), and ICD 377.75 Cortical Blindness (n = 542). See Supplementary data for full 4 yearly certification rates. See Fig. 2 for yearly changes in incidence rates.

Fig. 1
figure 1

Pie chart consisting of total number of certifications for both SSI and SI visual impairment groups across 2009–2022.

Fig. 2
figure 2

Line graph for the year-on-year changes in incidence rates of SSI and SI visual impairment certifications between the years 2009–2022 in children aged 0–15 years old in England and Wales.

Table 2 Total number of SI and SSI certifications in children aged < 16 years old between 2009–2022 in England and Wales.

Due to the impact of the Covid-19 pandemic on Ophthalmology services, new visual impairment certifications from 2019/20 to 2020/21 dropped by 39.1% and 40.8% for SI and SSI respectively. However, transitioning into 2021/22, new certifications increased by 75.7% and 54.4% for SI and SSI respectively as would be expected during recovery of clinical services.

Neoplasms

There was a significant decrease in new SSI certifications in 2017–2021/22* (45) compared to 2009–2012 (76). No difference was seen in the incidence rate for SI neoplasm related certifications.

Metabolic

There was a significant increase in new SSI certifications in 2017-2022* (103) compared to data from 2009–2012/13 (45). A significant increase was also seen in the same yearly periods respectively for new SI certifications. There were 97 SI certifications between 2017-2022* compared to 55 between 2009–2012/23.

Disorders of the CNS (excluding neoplasms)

Significant increases in new SI certifications were seen between all three 4 yearly intervals across the length of the study. 462 new certifications were seen in 2009–2012/13 versus 684 (2013-16/17) and 826 (2017–2021/22*). In contrast, there were insignificant changes in the rate of SSI certifications relating to disorders of the CNS.

Notably, there was a substantial increase in the number of SI certifications for cortical blindness in the latter years of the data provided (supplementary information). There were 69 certifications in total from 2009–2016/17. However, from 2017/18–2021/22 there were 496 new certifications for cortical blindness.

Retinal disorders

Amongst SI certifications there was a significant increase in 2013–2016/17 (620) and 2017–2021/22* (561) compared to 2009–2012/13 (415). However, the change between the latter two interval groups were insignificant. Amongst SSI certifications a significant increase was also observed 2013–2016/17 (424) and 2017–2021/22 (493) compared to 2009–2012/13 (328).

Amongst both types of visual impairment certifications, hereditary retinal dystrophies made up an overwhelming majority (>70%) of both SI and SSI certifications within the retina category, in every year assessed.

Disorders of the optic nerve

No significant changes were assessed in the rate of new SSI certifications. There was a significant increase in SI certifications in 2013-2016/17 (302) and 2017–2021/22* (323) compared to 2009–2012/13 (215).

Congenital anomalies of the eye including infection and nystagmus

There was a significant decrease in number of SSI certifications in 2017–2021/22 (316) compared to both 2009–2012-13 (416) and 2013–16/17 (433). In contrast there was a significant increase in the number of new SI certifications in 2013–2016/17 (1163) and 2017–2021/22 (1149) versus 2009–2012/13 (835).

Albinism and nystagmus formed the majority of new SI certifications with 2265 certifications. Regarding albinism, there was a 41% increase in new SI certifications following 2015/16–2021/22 (507—excluding 2020/21) from 2009–2014/15 (407). Over the same period there was a 93% increase (778 vs 403) in nystagmus (ICD9 379.5/379.51) SI certifications. This was not replicated amongst SSI certifications, with albinism and nystagmus only making up 356 out of a total 1207 total SSI certifications across the study.

Corneal and lens disorders including cataracts

There were insignificant changes in the incidence rate of both SSI and SI certifications across the whole data set.

Discussions

Summary of results

The aim of this study was to assess the trends in various pathologies leading to new visual impairment certifications across 2009–2022. Previous studies have been conducted assessing the incidence rates of certifications. However, they did not delve into the trends associated with various conditions over a prolonged period of time. Bunce et al. investigated the trends in overall sight impairment across 1999–2015 [1]. However, this study only spotlighted the incidence rate changes across 2008–2015 for total SSI and SI certifications rather than individual pathologies. Mitry et al. similarly investigated CVI data, but analysis compared the years 1999/2000 to 2007/8 and 2009/10 rather than a longitudinal assessment of all certifications spanning data from all these years [12]. The increasing trend in new certifications concluded in previous studies can also be observed in our study, as in 2016/17 we observed a peak incidence rate of 16.2 new certifications per 100,000 children under the age of 16. However, a continuous year on year significant increase was not seen for the duration of the 14 years covered in this study. There were only 4 years where significant increases in the number of certifications could be demonstrated (12/13–14, 14/15–16, 15/16–17, 20/21–22). Across 4-yearly periods, there was a significant increase demonstrated between 2013–2017 compared for 2009–2012.

Impact of Covid

As expected, from 2019–2021 there was a significant decrease in certifications recorded (8.3/100,000 children). This was the lowest certification rate across the entirety of the study— even less than 2009 levels (9.6/100,000). We also observed that registrations decreased from 2018 and did not recover to 2016/17 highs.

A likely reason for this could be the impact of the Covid-19 pandemic. The Covid-19 pandemic had a devastating impact on all non-acutely sight/life threatening services across the UK, including paediatric ophthalmology. Wood et al. analysed that following the 1st peak of the pandemic, outpatient services were reduced by 87.2%, outpatient referrals to ophthalmology by 50.2% and ward reviews by 50% [13]. This would have contributed to the approximately 40% reduction in visual impairment registration observed in our study as a large proportion of NHS services were redirected to inpatient acute medical care in conjunction with the pandemic. Wood et al. also discovered that there was a 12-month delay in children referred via their schools as preschool screening and formal visual assessments were postponed in the height of the pandemic [13]. This could potentially explain the 76% (SI) and 54% (SSI) post-pandemic increase in visual impairment certifications once non-sight/life threatening services resumed. It is unclear whether this was primarily due to an increase in outpatient services following the first peak of the Covid-19 pandemic or secondary to worsening of visual impairment directly caused by delayed ophthalmological appointments. As our data set only covers 1 year post-pandemic, it would be useful for future studies to assess whether the increase in certification rates observed prior to covid would continue once certification data covering the subsequent years thereafter are released.

Neoplasms and optic nerve

Neoplasms in the CNS can present clinically with headaches, neurological symptoms, binocular diplopia amongst other symptoms. However, early stages of the disease children may be asymptomatic and vision preserved. Traditionally patients picked up on routine screening e.g. preschool, or following referral with symptoms, underwent a clinical examination. However, technological developments have allowed more advanced diagnostic and analysis tools to be adopted e.g. optical coherence tomography, which can better allow preclinical detection of pathology [14].

Following the Honey Rose trial in 2016, where papilloedema in a child was missed in the community due to a series of systematic errors (the ‘Swiss cheese’ effect), there has been an uptick in referrals to acute paediatric ophthalmology, particularly for headaches and/or ‘suspicious’ optic discs [15]. This has been formally captured by the Neuro-Ophthalmological department in Bristol Hospital Trust, which saw a 500–600% increase in optic disc oedema related referrals from 2015–2019 with an increase in false positive referral rates from 33.3% in 2015 to 60.9% in 2016 [16]. In Nottingham Trust there was also a formally documented increase scan requests and referrals referencing papilloedema following the Honey Rose trial [17]. Even though there has been a large false positive rate, we speculate that the sheer increase in children being investigated for optic disc related pathologies has had a knock-on effect in the number of registrations due to earlier diagnosis, demonstrated by a decrease in the number of SSI certifications as a whole but an increase in the number of SI certifications following 2016.

Disorders of CNS

Cortical blindness was one of the leading causes for new SI and SSI certifications. Cortical blindness is determined as visual loss secondary to occipital cortex damage[18]. In children (particularly those born premature) it often is secondary to hypoxic-ischemic encephalopathy [19]. Chang et al. surmised that cortical blindness is the most frequent cause of paediatric visual impairment in developing countries, and is increasing in prevalence [19].

When looking at individual ICD coding it would seem there was a 618% increase in new SI certifications for cortical blindness (ICD 9 377.75) following 2017 compared to the years prior. However, on further investigation of the diagnostic codes there was a change in diagnostic criteria/coding so that the majority of cases previously being labelled as ‘Disorders of the Visual Cortex (ICD 9 377.7)’ are coded as ‘Cortical Blindness.’ This presents a limitation regarding the variability of ophthalmologist led registrations as the apparent significant changes in certifications rates of specific conditions are in reality due to numerous codes. This makes longitudinal analysis of the data set challenging with the risk of false conclusions being made.

Following the line of assumption that cases are being registered under a new code, when combined together the vast increase initially calculated in cortical blindness certifications is not replicated. Pre 2017 totals for these 2 diagnostic codings were 826 (SSI) and 700 (SI) and post-2017 were 574 (SSI) and 604 (SI). This however does not represent a decrease in cases as post-2017 in our dataset there is 1 less year of analysis. Interestingly, when assessing incidence rates and confidence intervals a significant increase is only observed in SI certifications from 2013-17 compared to 2009-13.

Metabolic and retinal

Genetic disorders such as leukodystrophies (ICD 330.0) and hereditary retinal dystrophies (ICD 362.7) were coded for a large majority of new SI and SSI certifications across the ‘Metabolic and Retinal Disorders’ sub-groups. This might represent a real increase in incidence due to increases in immigration of people from communities where consanguinity is more common. Or it may just relate to better diagnosis and more aggressive testing due to improved availability of genetic testing driven by possibility of gene therapies in certain cases due to advancements in Whole Genome Sequencing. From a systematic review in 2021, a total of 271 genes have already been identified to impact photoreceptors and retinal pigment epithelium [20]. In the early stages of many retinal disorders, the visual impairment may be mild/variable, or difficult to assess in the infant/younger child, with a ‘normal’ retinal appearance. Genetic sequencing allows for earlier molecular diagnosis of retinal dystrophies such as Leber Congenital Amaurosis [21]. Furthermore, with increasing interest and work into gene therapy, earlier molecular diagnosis allows access to research trials and treatment. In a meta-analysis on complete gene augmentation trials, there was an increase in retinal sensitivity amongst RP65-mutated populations found in 2 trials [20, 22, 23]. With advancements and improvements in gene therapy, a more widespread replicable improvement in visual acuity may come to fruition, which will have a lifetime impact on the child.

Leukodystrophies alone were responsible for 87.3% of metabolic-related visual impairment certifications. Leukodystrophies present with varying symptoms throughout infancy into adulthood. Traditionally MRI and electrodiagnostics were central to diagnosis. However, since 2015 due to the development of next generation sequencing, genomic testing was adopted into diagnostic guidelines by the Global Leukodystrophy Initiative [24, 25]. Whole exome sequencing has reported a yield of 25% for the diagnosis of Leukodystrophies which helps overcome limitations associated with analysis of MRI images [26]. Currently newborn screening for Leukodystrophy is not recommended by the National Screening Service which makes it difficult to make a preclinical diagnosis before the development of visual motor or cognitive dysfunction. Interestingly in the US, screening for X-linked adrenoleukodystrophy (X-ALD) is recommended by the Health and Human Services Advisory Committee on Heritable Disorders in Newborns and Children (ACHDNC) and included as part of routine screening in more than 15 different states [27, 28]. A future study assessing visual impairment registrations of Leukodystrophy in both countries would be highly informative in assessing whether screening protocols are associated with an earlier age of diagnosis and subsequently better clinical and social support.

Congenital abnormalities of the eye

Albinism and nystagmus made up the majority of visual impairment certifications in the SI ‘Congenital abnormalities of the eye’ subgroup. Our study observed a continued upward trend in certification rates for both Albinism and nystagmus which we hypothesise is due to a variety of reasons. Firstly, there has been a change in mindset and guidelines for the diagnosis of infantile nystagmus which has allowed for earlier diagnosis and more certifications to support access during schooling years [29]. Ocular investigations are more appropriately performed as first line unless there are ‘red flags’ or neurological symptoms or signs [30, 31]. In addition, the Nystagmus Care Pathway, and promotion of guidelines for best practice has resulted in increased awareness for the benefits and support children receive with SI registration [29, 32].

Secondly, the Royal College of Ophthalmologists (RCO) has not only encouraged, but also recommended, official registrations via the CVI and referrals to long term support programs for children with congenital abnormalities of the eye [31]. Institutions and support groups such as the Eye Clinic Liaison Officer (ECLO), The Albinism Fellowship and the Nystagmus Network have been signposted to clinicians in order help improve awareness to both clinician and patient families to help register children as visually impaired promptly in order to help give children with nystagmus the support they require in the community, particularly educational support. This increased awareness of registration pathways may help explain the increase seen in incidence rates of SI certifications associated with Albinism and nystagmus without a corresponding change in prevalence rates [29].

Limitations

Due to the design of our dataset, sight impairment certifications were not separated into age groups. As a result, our study was unable review trends within age-based population groups so any conclusions drawn were based on the entire population. Similarly, we did not have specific data for different geographic regions which could potentially highlight differences in access to ophthalmic services and subsequent registration rates secondary to socio-economic deprivation.

This study used crude significance based on the assumption that if confidence intervals did not cross then there were significant changes in incidence rates between our pre-defined periods. SI/SSI certifications cannot be assumed to correlate with the prevalence of each condition, so conclusions cannot be drawn regarding visual function/SI/SSI in different conditions based on this study.

To ensure that patients could not be identified, we excluded any data amongst conditions which had <5 new certifications per year, in every year of the study. This resulted in 4 broad clinical groups being omitted from any analysis and assigned into an ‘other’ category.

Conclusion

Visual impairment certifications have shown an increasing trend from 2013–2022 compared to 2009-2013. Similarly to previous studies, we observed congenital abnormalities of the eye, disorders of the CNS and retinal disorders as the 3 pathological groups responsible for the highest number of VI certifications [1]. We have discussed possible reasons for these changes, which may include advances in genomic testing, earlier screening and diagnoses, as well as pressure from external groups such as the Royal College of Ophthalmologist to register earlier in order for children to access extra support. The impact of Covid-19 led to significant decreases in certification rates across both SI and SSI datasets. Future studies may wish to obtain data on different paediatric age groups and geographical regions to provide a better understanding of challenges to registration.

Supplemental material is available at Eye’s website.

Summary

What was known before

  • Formalisation of Visual Impairment in children involves completion of the Certificate of Visual Impairment form which allows consultant Ophthalmologists to register children. The Committee of Visual Impairment office then certifies the visual impairment which allows children to access greater clinical and social support.

What this study adds

  • Longitudinal analysis across 13 years (2009–2022) in the trends of visual impairment (VI) which revealed increasing trends of registration and certification of VI in children.

  • Similarly to previous studies, we observed congenital abnormalities of the eye, disorders of the CNS and retinal disorders as the 3 pathological groups responsible for the highest number of VI certifications.

  • The impact of the Covid-19 pandemic led to a significant decrease in certification rates across both SI and SSI datasets.