Article Text

Download PDFPDF

Evidence that children with special needs all require visual assessment
  1. Meghomala Das1,
  2. Katherine Spowart2,
  3. Stephanie Crossley2,
  4. Gordon N Dutton3
  1. 1Southern General Hospital, Glasgow, UK
  2. 2Royal Hospital for Sick Children, Yorkhill, Glasgow, UK
  3. 3Tennent Institute of Ophthalmology, Gartnavel General Hospital, Glasgow, UK
  1. Correspondence to Meghomala Das, 218/7 Eaglesham Road, East Kilbride, Glasgow G75 8RH, UK; meghomala.das{at}


Design A protocol-based ophthalmological assessment was performed on-site by a skilled investigator.

Patients Children attending schools for special needs in Glasgow were offered eye care within their school. Outcomes for the first 240 participants are reported.

Outcome measures Number of children for whom visual acuity could be measured and the results of refraction.

Results 228/240 (95%) children were able to co-operate in a complete or nearly complete assessment of visual function. Visual acuity could be reliably assessed in 190 children using a range of tests from preferential looking to logMAR charts. 23/190 (12.1%) were found to be visually impaired according to WHO criteria. 105/228 (46.1%) subjects were found to have a refractive error which required correction. 50/105 subjects were wearing an adequate correction (ie, difference of less than 0.75 D sphere or cylinder compared with the retinoscopy result obtained on screening) and 55 (24.1%) children were prescribed a new correction. The prevalence of all types of refractive error was found to be significantly higher in the population with special needs, in particular the learning disabilities subgroup, compared with normal children. High hyperopia and astigmatism were common.

Conclusion With patience, suitably adapted methods, a familiar environment and skill, visual assessment is feasible in the majority of children with special needs. The prevalence of reduced visual acuity is high and unaddressed correctable refractive error is a major cause.

View Full Text

Statistics from


In recent years, awareness has grown that adults with special needs may be at an additional risk of visual abnormality and disability.1 Refractive errors and visual impairment in adults with learning disabilities has been studied extensively2 3 4; van Splunder et al3 4 concluded that the prevalence of visual impairment is higher in individuals with learning disabilities as compared to the general population. Some have gone so far as to suggest that all individuals with severe or profound learning disabilities, and all older adults with Down's syndrome, should be considered visually impaired unless proven otherwise.4 In spite of this evidence base, regular and effective sight tests are rarely performed in this population. Hence, reduced vision may go undetected, leading to further disability, dependence, low quality of life, and in some instances, preventable impairment of vision.

Published epidemiological information concerning visual difficulties in children with special needs is scarce and confined to ophthalmic literature.5 6 7 This is not regularly accessed by paediatricians who are generally responsible for the care of such children. Reliable prevalence data are not available to serve as a basis for national health policies and the provision of specific low-vision services. Unrecognised visual difficulty may adversely affect development, social behaviour and learning, in this vulnerable group. If such visual difficulty is a result of correctable refractive errors, it is potentially a missed therapeutic opportunity.

What is already known on this topic

  • Visual impairment in adults with learning disabilities has been extensively studied; however, epidemiological data concerning visual impairment in children with special needs/learning disabilities is scarce.

  • There is some awareness of the increased incidence of visual abnormality in children with cerebral palsy and/or prematurity (representing a subsection of children with learning disabilities).

What this study adds

  • This study demonstrates that with enough patience, suitably adapted methods and skilled investigators, visual assessment is feasible in most children with learning disabilities.

  • It raises awareness among paediatricians that the population they are responsible for are at increased risk of visual impairment and, importantly, of potentially correctable refractive errors.

Patients and methods


The majority of children with profound intellectual and/or physical disabilities in Glasgow and surrounding areas attend special schools, which address their complex learning and medical needs, and aim to maximise their physical and cognitive potential. Visual testing and optometric examination is now being carried out in these schools as a new clinical service; this has, for the first time, provided the opportunity to determine the frequency of visual and optometric disorders in this group.

Study population

A sample of 240 children and young people, aged 5–19 years, attending six special schools for complex needs and/or physical disability in Glasgow were examined over a period of 24 months (see online figure 1 and online table 1). The children from the schools for complex learning difficulties have severe or profound global learning disabilities, and generally function at or below one third of their chronological age. They are educated in small classes of around six pupils with a high adult:child ratio and progress into adulthood to be dependent. The children in schools for physical impairment comprise a more varied population; some function intellectually within the norm and follow a standard curriculum, but many others have severe learning difficulties, similar to those in the complex needs schools, following a similar individual curriculum, requiring similar support levels and not progressing to independence in adulthood. They may in addition have cerebral palsy, or severe, often end stage, systemic disease, for example, cystic fibrosis or muscular dystrophy, such that they need nursing care and physiotherapy frequently during the school day.

Assessment was offered to all pupils regardless of previous or current eye care. Subjects were not selected for prior visual defects or because of parental/carer concerns regarding vision.

Informed consent was obtained from participants and/or their parent or carers. A protocol-based ophthalmological assessment was performed on-site by a skilled investigator, assisted in most cases by a member of school staff known to the child/young person. Medical records were checked and the main diagnoses recorded, for example, learning disability, cerebral palsy and autistic spectrum disorder (see online table 2). Sampling and assessment were performed between November 2004 and October 2006.

The subjects examined have been divided into two groups, SN (special needs) and LD (learning disabilities). SN refers to the entire population of children with special needs. Out of these 240 children, 183 (76.2%) had predominantly learning difficulties. They have been subgrouped and analysed under the heading LD. The main reason for this subgrouping is to compare our results with those of other studies, all of which deal with individuals having predominantly learning difficulties.


Visual impairment

Visual impairment (includes low vision and blindness) was classified according to WHO criteria.8

Low vision: visual acuity <6/18, but equal to or greater than 3/60 and/or a corresponding visual field loss of <20 degrees in the better eye with best possible correction.

Blindness: visual acuity <3/60 and/or a corresponding visual field loss of <10 degrees in the better eye with best possible correction.

(Snellen criteria were employed in this study.)

Refractive errors

Myopia is defined as spherical equivalent (spherical power+0.5 cylindrical power) of at least −0.50 DS (dioptres spherical) or more.

Hyperopia is defined as spherical equivalent of +2.00 DS or more.

Astigmatism of ≥0.75 DC (dioptres cylinder) was deemed significant.

For statistical analysis of refractive errors, only the results of the eye with better vision were considered, as this equates best with the level of visual disability.

Refraction was determined under cycloplegia (1% cyclopentolate eye drops were used in all patients). The spherical equivalent of the refractive errors was calculated according to the formula: spherical refraction+0.5×cylindrical refraction. The ophthalmological assessment included a structured observation of visual attention and fixation, position of the eyes, measurement of visual acuity, anterior segment examination (external eye structures, cornea, pupils and lens), refraction under cycloplegia (1% cyclopentolate) and fundal examination. Visual acuity measurement, depending on the degree of co-operation, was performed by means of one of the following tests: Snellen chart, uncrowded logMAR, Cardiff Acuity Cards, Keeler gratings and Kay single pictures cards. Near vision was evaluated with the Stycar near-card at 40 cm. Distant visual acuity was assessed with both eyes open and, if possible, monocular. Near vision was only tested binocularly. The presence of pretest corrective prescriptions was established from information given by the parent/carer on the consent form and the use of spectacles at presentation. The correction was measured with a lensometer. Subjects who had spectacles for distance or near vision wore them during the test, with the exception of two subjects who had prescriptions but had broken their glasses or thrown them away. All visual tests were performed by an optometrist with considerable paediatric experience.

Statistical analysis

Patient identifiers, diagnosis, visual acuity, refraction, etc, as above, were recorded and entered into a database on site. Results were then grouped according to the degree and type of refractive error and also by level of visual acuity. A χ2 test was used to compare the results for refractive errors in the LD group with results from a large population study of normal children as well as a study of adults with learning disabilities.



Sufficient data were available for 228/240 (95%) children to qualify for inclusion in the final analysis. Eleven of the 183 children in the LD group were unable to complete adequate testing (one due to the acuity being light perception or poorer, two others due to corneal opacities/clouding and eight due to inadequate co-operation with the process). There was one additional child in the whole group (SN) who was uncooperative with testing.

Visual acuity could be reliably assessed in 190 children in the SN group and 134 in the LD group. Thirty-eight children, all in the LD group, could not have their pretest visual acuity recorded due to varying levels of co-operation. However, adequate refraction could be performed and hence they have been included in the final case analyses.

Overall, a high degree of co-operation was possible as the assessments were made in the participants' familiar environment, supported by a known caregiver and 45 min was available for each complete assessment. These circumstances differ greatly from those in general ophthalmological or optometric practices, where co-operation and success in this group are poorer.

Online figure 2 illustrates the presenting distant visual acuity before any intervention (children with pre-existing glasses wore them during the test) in these two groups.

Visual impairment

The following were found to be functioning with vision at a level classifiable as visual impairment according to WHO criteria8: SN, 23/190 (12.1%); LD, 18/134 (13.4%).

Refractive errors

Refraction is best determined objectively by retinoscopy first, followed by subjective correction (with a suitable acuity test) to bring about greater accuracy in the final prescription.

In reality, after undergoing objective refraction under cycloplegia, most of the children could not complete a postrefraction assessment of visual acuity. This was because their pupils were widely dilated and using a pin-hole for subsequent subjective correction required substantial co-operation and concentration, which was beyond their scope. In this study, the results of cycloplegic refraction on the first visit only have been considered and hence we are unable to comment on postrefraction improvement in visual acuity.

Refraction could be adequately performed in 228 children in the SN group and 172 in the LD group. The numbers of children with a refractive error, and the type, in each of the two groups are shown in table 1.

Table 1

Numbers of children with a refractive error, and the type, in the SN and LD groups

Current prescription and new correction

Overall, 65/228 subjects had a pre-existing spectacle prescription. We considered current spectacle prescription as adequate when the prescription differed by 0.75 D or less in sphere or cylinder from the retinoscopy result. Using this criterion, 50 subjects were considered to be wearing an adequate correction. Fifteen subjects wore inadequate distance correction, two of whom had inappropriate near correction (bifocals) as well.

Precise, agreed guidelines for the prescription of spectacles in all individual circumstances do not exist. In general, we have considered a refractive error of less than −0.50 DS or more than +2.00 DS, and/or a cylindrical correction of greater than 0.75 DC, as clinically significant. In all, 40 subjects not currently wearing spectacles were considered likely to benefit from a distance correction. Of these, two required new near correction (bifocals). All 15 subjects with inadequate prescription were given new correction. Hence, in total, abiding by our criteria, 55 subjects were issued with a new prescription (24.1%).

In the LD group, 47/172 participants had a pre-existing spectacle prescription, 45 for distance (nine inadequate) and two for near (one inadequate). Thirty-five subjects not currently wearing spectacles were considered likely to benefit from a new correction. This includes 29 new distance, four bifocals and two near prescriptions. In all, 45 (26.2%) subjects required a new prescription. These are shown in table 2.

Table 2

Current prescription and new correction

We do not know whether any of the subjects wearing spectacle corrections which were ‘inadequate’, had been deliberately undercorrected. Since high-powered spectacle lenses can produce intolerable distortions and magnification, it is feasible that some subjects were intentionally prescribed partial corrections for an initial period. Nevertheless, it is clear that significant numbers of children with special needs are unnecessarily visually impaired by uncorrected or inadequately corrected refractive errors.

Ocular disorders

Ocular disorders were found in many of the children and many subjects had multiple diagnoses. These are shown in table 3.

Table 3

Ocular disorders of the best eye


Refractive errors are not uncommon in the child population, however comparison of our data with a large child population study emphasises the significant difference between the two groups, demonstrating the need for children with special needs to be regarded separately. Comparison with the adult population with learning disabilities, who are already recognised as needing specialist eye services, demonstrates that children with learning disabilities are more similar to these adults than they are to other children.

Direct comparison of our findings with other studies of normal child populations and of people with learning difficulties is difficult, as definitions and/or classifications vary greatly. However, the present study is best compared with the report by Woodhouse et al9 regarding refractive errors associated with adult learning disability and the study by Junghans et al10 in normal children. The definitions of refractive errors differ in these papers; hence our figures have been recalculated to exactly match the definitions used in each paper to allow direct comparison.

The study by Junghans et al presents data concerning a large population of Australian school children.10 The age range is a little narrower than in our study (3–12 years), but does cover the age during which myopia is expected to occur.11 Due to the inclusion of 3- and 4-year-old children, a higher incidence of hyperopia would be expected in this study. However, despite this, the incidence of hyperopia >1.50 DS and astigmatism >1.00 DC was found to be significantly higher in our population as compared to normal children (table 4). It is to be noted that there is considerable overlap in the numbers of children presenting with hypermetropia >1.50 and >2.0 DS, myopia ≥ −0.50 and ≥ −2.0 D as well as astigmatism ≤0.25 DC and between −0.25 and −1.00 DC.

Table 4

Comparison of refractive errors among SN/LD children and normal children

Woodhouse et al looked at a Welsh population of 148 young adult clients at day-care centres.9 She included data from the right eye only, regardless of whether this was the eye with best vision, whereas we have considered the refractive error of the eye with better vision. Refractive errors in our cohort are less common than in the study population of adults with learning disabilities, especially with respect to astigmatism and myopia (table 5). However, myopia is known to increase with age in the general population and the differences do not reach statistical significance other than for minor degrees of astigmatism.

Table 5

Comparison of refractive errors among SN/LD children and adults with LD


This study presents the initial results of a new clinical service for vision assessment of children with severe intellectual and physical disabilities. It has demonstrated that with enough patience, on-site assessment, suitably adapted methods and skill, visual assessment is feasible in the majority of this population. Also, a considerable number of children with special needs have undetected and untreated reduced visual acuity, which might be remedied by simply prescribing accurate spectacle correction.

The manner in which an eye test is conducted is most important for these children. Visits to the optometrist are infrequent, making the environment less familiar, and this can be confusing or frightening. The situation is further complicated by the fact that these children usually cannot communicate effectively. Optometrists have noted the value carers bring to eye tests in preparing adults with learning difficulties for the examination, bringing briefing information regarding each individual's capabilities.12 In this study, a high percentage of co-operation was possible as the assessments were made in the participant's own environment, supported by a known caregiver.

Many adults with learning difficulties cannot read letters and few children with learning disabilities can. Hence, optometrists attempting to assess these children need to have access to and be trained in the use of alternative testing methods. However, specific training for optometrists related to learning difficulties has been shown to be rare.13 The optometrist involved in visual assessment in this study, however, had considerable paediatric experience.

Studies have claimed that post-test support should be enhanced, for example, to improve tolerance to glasses and use of other visual aids.12 However, most studies have been completed across a single period, in adults and without a return visit to assess follow-up. A longitudinal study is required to assess the effectiveness of eye testing and sight optimisation aids in children. This clinical service, of which these are the first results, will provide such an opportunity.

The type, quantity and quality of services for children with special needs and learning difficulties vary greatly, depending on geographical location. This study has examined a small sample of such children; the true number across Scotland is not known. Limited national data suggest that there are nearly 1564 children with the most severe (multiple/complex) disabilities.14 If these results could be extrapolated to include the entire vulnerable population with severe/profound learning and/or physical disabilities, they would reveal a much larger number of children who are living with uncorrected refractive error. Therefore, services are required to be developed universally to address these issues and prevent further disability in this already handicapped group.


The authors are grateful to all the subjects who took part in the screening and to the school staff and carers who helped in the smooth running of this service.


View Abstract


  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.