Childhood Blindness

Paul G. Steinkuller, MD,a Lee Du, MD,a Clare Gilbert, MD, FRCOphth,b Allen Foster, MD,
FRCOphth,b Mary Louise Collins, MD,c and David K. Coats, MDa

Purpose: The objective of this study was to summarize available data regarding pediatric blinding diseases world-
wide and to present the most up-to-date information on childhood blindness in the United States. Methods: We
obtained data from a complete search of the world literature and from direct contact with each of the schools for the
blind in the United States. Results: Five percent of worldwide blindness involves children younger than 15 years of
age; in developing countries 50% of the population is in this age group. By World Health Organization criteria, there
are 1.5 million children worldwide who are blind: 1.0 million in Asia, 0.3 million in Africa, 0.1 million in Latin America,
and 0.1 million in the rest of the world. There are marked differences in the causes of pediatric blindness in different
regions, apparently based on socioeconomic factors. In developing countries, 30% to 72% of such blindness is avoid-
able, 9% to 58% is preventable, and 14% to 31% is treatable. The leading cause is corneal opacification caused by a
combination of measles, xerophthalmia, and the use of traditional eye medicine. There is no national registry of the
blind in the United States, and most of the schools for the blind do not keep data regarding the cause of blindness in
their students. From those schools that do have this information, the top 3 causes are cortical visual impairment,
retinopathy of prematurity, and optic nerve hypoplasia. There has been a significant increase in both cortical vision
loss and retinopathy of prematurity in the past 10 years. Conclusions: There are marked regional differences in the
prevalence and causes of pediatric blindness, apparently based on socioeconomic factors that limit prevention and
treatment schemes. In the United States the 3 leading causes of pediatric blindness are cortical visual impairment,
retinopathy of prematurity, and optic nerve hypoplasia. There is a need for more complete and more uniform data
based on the established World Health Organization reporting format. (J AAPOS 1999;3:26-32)

T he World Health Organization (WHO) lists 5 cate-

gories of visual impairment (Table 1).1 By these cri-
teria in 1992 WHO estimated that there were 1.494
million children in the world who were blind or severely
visually impaired (categories 2 through 5), 1.080 million
(72%) in Asia, 0.264 million (18%) in Africa, 0.078 million
(5%) in Latin America, and 0.072 million (5%) in the rest of
the world combined. Detailed prevalence and incidence
data have been difficult to obtain. Of fully industrialized
nations, only the United Kingdom, the Scandinavian coun-
tries, and Canada have national registries of the blind and
visually impaired,2 and their data are not standardized or
uniformly reported. Other countries have relied on clinic or
hospital-based surveys for this information.3,4 Population-
based surveys are the most satisfactory and accurate way to
obtain prevalence data, but such surveys are extremely
labor-intensive and costly. There have been a few popula-
From the Department of Ophthalmology, Baylor College of Medicine, Houston, Texasa; the
International Centre for Eye Health, Institute of Ophthalmology, University College,
London, United Kingdomb; and Greater Baltimore Medical Center, Baltimore, Maryland.c
Presented at the 24th Annual Meeting of the American Association for Pediatric
Ophthalmology and Strabismus, Palm Springs, California, April 48, 1998.
Submitted April 8, 1998.
Revision accepted August 17, 1998.
Reprint requests: Paul G. Steinkuller, MD, c/o Christian Blind Mission International, 450
E Park Ave, Greenville, SC 29602.
Copyright 1999 by the American Association for Pediatric Ophthalmology and
Strabismus.
1091-8531/99 $8.00 + 0 75/1/95561
tion-based ocular disease prevalence surveys using random
cluster sampling techniques, mostly in developing coun-
tries, and they do not often show an adequate picture of
childhood visual impairment because only 5% of blindness
involves this age group, and most population-based ocular
disease surveys are not large enough to sample enough chil-
dren to generate significant data. A common way to cir-
cumvent this problem has been the tabulation of children
enrolled in schools for the blind. This method is not entire-
ly satisfactory because not all blind children may be at such
schools, because the criteria for enrollment may vary from
country to country and from one time to the another, and
because some children in the schools may not be blind at
all.5 By WHO criteria childhood is 0 to 15 years of age; in
developing countries as much as 50% of the population is 15
years of age or younger, compared with 20% to 22% in
England and the United States.2 WHO estimates that there
are 0.5 million children per year who are born blind or
become blind, of whom 50% to 90% die, most to compli-
cations of severe protein-energy malnutrition. Since 1993 a
standardized form developed by the International Centre
for Eye Health in London and the Programme for the
Prevention of Blindness (WHO) has been used for the
reporting of pediatric vision loss6 (Figure 1). The form has
been used mainly to gather data in developing countries; it
has not yet been used for the assessment of childhood ocu-
lar disease in the United States.

26 February 1999 Journal of AAPOS

Journal of AAPOS
Volume 3 Number 1 February 1999 Steinkuller et al 27

FIG 1. Standardized World Health Organization/Prevention of Blindness Committee Eye Examination form for visually
impaired and blind children.

The purpose of this study was to gather and summarize TABLE 1. WHO categories of visual impairment
the available data regarding pediatric blinding disease world- Category Terminology Vision, better eye, best corrected
wide and to collect similar information from schools for the 1 Visually impaired <6/18 to 6/60 (<20/30 to 20/200)
blind in the United States for analysis and comparison. 2 Severely impaired <6/60 to 3/60 (<20/200 to 20/400)
3 Blind <3/60 to 1/60 (<20/400 to 20/2400)
METHODS
4 Blind <1/60 to LP (<20/2400 to LP)
The world literature was surveyed for prevalence, anatom- 5 Blind NLP
ic site, and age of occurrence incidence of childhood (ages LP, Light perception; NLP, no light perception.

28 Steinkuller et al
FIG 1 Continued from previous page.
0 to 15 years) blindness worldwide. Questionnaires were
mailed to each of the schools for the blind in the United
States (Figure 2). The responses are summarized in Tables
2 through 4. To facilitate the development of prevention
and treatment strategies, childhood blindness is catego-
rized (as on the WHO reporting form) according to the
anatomic site of the lesion causing the blindness and
Journal of AAPOS
Volume 3 Number 1 February 1999
according to the age or stage of development at which the
causative insult occurred. Most of the studies performed
since 1993 have used this format.
RESULTS
There are distinct regional differences (Tables 3 and 4)
and important intraregional differences. If data are avail-
Journal of AAPOS
Volume 3 Number 1 February 1999 Steinkuller et al 29

FIG 2 Study form mailed to schools for the blind in the United States.

able for a specific country, city, or segment of population TABLE 2. Worldwide childhood (0 to 15 years) blindness (WHO categories
(eg, urban versus rural), broad generalizations should be 3 to 5)
avoided; individual country reports should be studied. Region Prevalence (per 1000 children) No. blind
All of the schools for the blind and visually impaired in Africa 1.1 264,000
the United States were contacted at least twice by letter Asia 0.9 1,080,000
and at least twice by phone to ascertain whether records Latin America 0.6 78,000
were available regarding the causes of vision loss among North America, Europe, Japan 0.3 72,000
their students. Of the 128 schools contacted, 20 (16%) had World 0.8 1,494,000
data appropriate for this study. A total of 2553 students
were surveyed; the results are summarized in Table 5.
prevalence surveys are relatively difficult and expen-
DISCUSSION sive.2,7,9 Although few countries have efficient national
There have been few population-based ocular disease preva- registries of the blind and visually impaired, the concen-
lence surveys worldwide that involved enough children to tration of such children in schools for the blind has afford-
yield truly adequate data, because pediatric cases account for ed a convenient if not entirely accurate way of assessing
only 5% of all blindness,8 and because population-based the causes of pediatric blindness in given populations; sur-
 Journal of AAPOS
30 Steinkuller et al Volume 3 Number 1 February 1999

TABLE 3. Anatomic site of lesion causing blindness (%)

Region No. in sample Cornea Lens Retina Optic nerve Glaucoma Other
Africa 2086 56 10 11 7 6 11
Asia 2093 21 14 21 7 5 32
Latin America 1735 10 18 31 12 10 18
Middle East 991 10 16 35 10 8 22
Europe/N. America 4228 1 9 27 22 13 29
*Mostly cortical visual impairment.
Canada.

TABLE 4. Age or stage of development at which blinding insult occurred (%)

Region Hereditary Intrauterine Perinatal Childhood Unclassified
Africa 16 5 2 47 31
Asia 25 3 6 20 46
Latin America 30 8 18 12 32
Middle East 66 2 2 17 14
Europe 39 6 22 12 22

TABLE 5. Causes of vision loss in students in schools for the blind and 90% of blind children are in Asia (72%) or Africa (18%).10
visually impaired in the United States (n = 2553) In economically established countries (such as the United
Diagnosis No. % of total States, Canada, Japan, and western Europe) the prevalence
Cortical blindness 497 19 of childhood blindness is estimated at 0.3 per 1000 chil-
ROP 320 13 dren, ranging from 0.10 to 0.41 per 1000. In Asia the over-
Optic nerve hypoplasia 166 7 all figure is 0.9 per 1000 (range, 0.63 to 1.09), and in Africa
Albinism 140 5 it is 1.1 per 1000.2,5,7,9,11-15 In the 1980s WHO estimated
Optic atrophy 124 5 an annual incidence of blindness (WHO categories 3
Cataract 109 4 through 5) in all age groups combined of 7 million world-
Retinitis pigmentosa 72 3 wide, including 500,000 children (50 to 100 new childhood
Microphthalmia/anophthalmia 65 3 cases per year per million population). The mortality rate
Aniridia 58 2 in this group of newly blind children is estimated at 50% to
Glaucoma 43 2
90%, mostly as a result of the complications of vitamin A
Lebers congenital amaurosis 36 1
deficiency and severe protein-energy malnutrition.2 The
Cone-rod dystrophy 36 1
Coloboma 26 1 number of children (ages 0 to 15) worldwide is expected to
Retinal detachment 21 1 double in 44 years, ranging from 25 years in sub-Saharan
Trauma 18 1 Africa to 233 years in Europe, if current population growth
Diabetes 11 rates remain unchanged16 (Table 6). It therefore is impera-
Congenital infection 7 tive to develop strategies to address this growing problem;
Toxoplasmosis 3 such planning requires knowledge of the causes of and
Rubella 3 trends in childhood blindness. Considerable effort in this
Cytomegalovirus 1 regard has been underway for the past decade.10
Herpes simplex virus 0 The incidence of pediatric blindness in fully industrial-
Syphilis 0
ized countries is much lower; in 1985 the Royal National
Other 0
Institute for the Blind reported 200 to 250 new cases of
Unknown 156 1
Other 648 25 untreatable blindness in English children2 out of a total
Total 2553 100 population of 45 million, of which 20% (9 million), or 1
new case per year per 180,000 to 225,000, only 5% to 7%
of the worldwide rate. If the incidence of childhood blind-
ness in the United States were similar to that in the United
veys of schools for the blind have become accepted epi- Kingdom there would be approximately 1000 new cases
demiologic methods. These studies, in conjunction with per year out of a total population of 270 million. In devel-
demographic data, reveal marked regional and intrare- oping countries 30% to 72% of pediatric blindness is
gional differences in the prevalence and causes of blind- avoidable, with most in fact being preventable; the leading
ness and severe visual impairment worldwide. More than cause of childhood blindness in the world is corneal opaci-
90% of all blind people live in developing countries, and fication and other anterior segment damage caused by a
Journal of AAPOS
Volume 3 Number 1 February 1999
TABLE 6. World population and yearly growth
Region 1998 population (millions) 0 to 15 years
Africa 786 50%
Asia 3663
Latin America 335
United States 270 22%
Europe 708
World 5994 32%
combination of measles, vitamin A deficiency, and the
deleterious effects of the use of traditional eye medicines.
All of these cases are potentially preventable by proven
public health measuresmeasles immunizations, periodic
vitamin A distribution to the at-risk group (children
younger than 6 years of age), and community education.
Other similarly preventable causes of childhood blindness
include congenital rubella, cerebral malaria, and oph-
thalmia neonatorum.2,5,9,10-14,17-21
Of the treatable causes of childhood blindness world-
wide, the leading conditions are cataract, retinopathy of
prematurity (ROP), and glaucoma. ROP, once thought to
be a disease limited to fully industrialized countries, is
emerging as an important blinding disorder in emerging
economies.22-25 Trauma-induced blindness caused by
antipersonnel landmine explosions in regions affected by
civil war involves children as well as adults and has become
a significant cause of disability in Southeast Asia,26 sub-
Saharan Africa, and Afghanistan.
The United States does not have a national registry of
the blind and visually impaired, and the population-based
ocular disease prevalence surveys that have been done in
this country have concentrated on adults.27 However,
because surveys of schools for the blind in other countries
have become accepted and standardized within the past 10
years,5,11-14,28 a similar study performed here could be at
least partially comparable. Of the 128 schools for the blind
in the United States, 24 (19%) maintain rosters of the
causes of blindness of their students, 22 (17%) responded
to a request for this information, and 20 (16%) had useful
data. Based on the initial information received, a simple 1-
page questionnaire (Figure 2) was developed and sent. The
questionnaire was constructed in such a way that most of
the responding schools could supply numbers for most of
the items. None of the schools used the standard
WHO/PBL reporting form, most had different criteria for
admissions, and many had more than one cause of blind-
ness listed for many students. Therefore the information
obtained is not entirely comparable with that gathered
from recent foreign surveys of schools for the blind in
which standardized WHO methods were used. Moreover,
many blind or severely visually impaired American chil-
dren are not enrolled in schools for the blind but rather
live at home and attend their local schools with assistance
from itinerant professional educators of the disabled.
Children enrolled at a residential school for the blind tend
Steinkuller et al 31
No. 0 to 15 years (millions) Annual growth Doubling (yr)
393 2.8% 25
1.6% 44
1.6% 44
59 0.9% 78
0.3% 233
1918 1.6% 44
to be multiply disabled and not blind only. With an under-
standing of these drawbacks, some of the information thus
gathered may still be useful.
Of the 2553 students in the study, the leading causes of
blindness were cortical visual impairment in 497 (19%);
ROP in 320 (13%); and optic nerve hypoplasia in 166
(7%). After these 3 leading items came albinism, optic
atrophy, cataract (and cataract/cataract surgery complica-
tions), retinitis pigmentosa, microphthalmia/anoph-
thalmia, aniridia, and glaucoma, with each causing 2% to
5% of the total. All of the other conditions caused 1% or
less each: Lebers congenital amaurosis, cone-rod dystro-
phy, coloboma (of the retina/optic nerve), retinal detach-
ment, trauma, diabetes, and the TORCH diseases. In 156
(6%) the cause of blindness was unknown. The leading
anatomic site causing the impairment was the retina (eg,
ROP + retinal detachment + albinism + retinal coloboma +
aniridia) followed by the brain (cortical impairment). The
anterior segment was the cause of blindness in only 6% of
the total (cornea 2%, lens 4%), compared with 35% in
Asia (cornea 21%, lens 14%) and 66% in Africa (cornea
56%, lens 10%).29 Perinatal factors (eg, ROP) (24%) and
hereditary (eg, albinism, some cataracts, retinitis pigmen-
tosa) (18%) factors predominated. In the Middle East and
Sri Lanka, where consanguineous marriages are more
common, hereditary factors cause 66% of childhood
blindness,11,15,30-33 and in Africa the leading factors of the
cause of blindness occurred after the perinatal period, that
is, in childhood (47%).2 The latter finding apparently
reflects the predominance of corneal opacification caused
by the measles/vitamin A deficiency/traditional eye medi-
cine complex in Africa. From those institutions in the
United States that provide sequential data, it appears that
the primacy of cortical blindness began around 1985 and
that the most recent surge in ROP-induced blindness
started in 1990.
In summary, the 3 leading causes of pediatric blind-
ness/severe visual disability in the United States identified
by this study were cortical visual impairment, ROP, and
optic nerve hypoplasia. This may reflect some case selec-
tion to conditions for which there is no treatment and to
conditions associated with other disabilities. Perinatal and
hereditary factors cause most childhood blindness in the
United States and the rest of the industrialized world,
compared with hereditary factors in the Middle East and
Sri Lanka and childhood factors in Africa. Retinal disease

32 Steinkuller et al
causes most pediatric blindness in the United States and
other fully developed countries, compared with cataract in
Latin America and the Middle East and corneal opacities
in Asia and Africa. More detailed and accurate data from
schools for the blind in the United States could be
obtained by a unified survey using the standardized
WHO/PBL format.
References
1. World Health Organization. Manual of the International Statistical
Classification of Disease, Injuries, and Causes of Death, Vol. 1.
Geneva: WHO, 1977.
2. Foster F, Gilbert C. Epidemiology of visual impairment in children.
In: Taylor D, editor. Paediatric ophthalmology. Oxford: Blackwell
Science Ltd; 1997. p. 3-7.
3. Goggin M, OKeefe M. Childhood blindness in the Republic of
Ireland: a national survey. Br J Ophthalmol 1991;75:425-9.
4. Backhouse O. Causes of decreased vision and blindness in
Madagascar. Community Eye Health 1996;17:14-6.
5. Gilbert C, Wood M, Waddel K, Foster A. Causes of childhood
blindness in East Africa: results in 491 pupils attending 17 schools
for the blind in Malawi, Kenya, and Uganda. Ophthalmic Epidemiol
1995;2:77-84.
6. Gilbert C, Foster A, Negrel D, Thylefors B. Childhood blindness: a
new form for recording causes of visual loss in children. Bull World
Health Organ 1993;71:485-9.
7. Prevention of childhood blindness. Geneva: World Health
Organization, 1992.
8. Whitfield R, Schwab L, Ross-Degnan D, Steinkuller P, Swartwood
J. Blindness and eye disease in Kenya: ocular status survey results
from the Kenya Rural Blindness Prevention Project. Br J
Ophthalmol 1990;74:333-40.
9. Foster A, Gilbert C, Rahi J. Epidemiology of cataract in childhood:
a global perspective. J Cataract Refract Surg 1997;23:601-4.
10. Thylefors B. A global initiative for the elimination of avoidable
blindness [editorial]. Am J Ophthalmol 1998;125:90-3.
11. Eckstein MB, Foster A, Gilbert CE. Causes of childhood blindness
in Sri Lanka: results from children attending six schools for the
blind. Br J Ophthalmol 1995:79:633-6.
12. Gilbert CE, Canovas R, Hagan M, Rae S, Foster A. Causes of child-
hood blindness: results from West Africa, South India, and Chile.
Eye 1993;7:184-8.
13. Gilbert C, Foster A. Causes of blindness in children attending four
schools for the blind in Thailand and the Philippines. Int
Ophthalmol 1993;17:229-34.
14. Gilbert CE, Canovas R, de Canovas RK, Foster A. Causes of blind-
ness and severe visual impairment in children in Chile. Dev Med
Child Neurol 1994;36:326-33.
15. Gilbert C, Rahi J, Eckstein M, Foster A. Hereditary disease as a
cause of childhood blindness: regional variation. Ophthalmic Genet
1995;16:1-10.
16. The World Almanac 1998. World Almanac Books. K-III Reference
Corp. 1998.
17. Loewenthal R, Peer J. A prevalence survey of ophthalmic diseases
among the Turkana tribe in northwest Kenya. Br J Ophthalmol
1990;74:84-8.
18. Rahi JS, Sripathi S, Gilbert CE, Foster A. Childhood blindness in
India: causes in 1318 blind school students in nine states. Eye
1995;9:545-50.
19. Zerihun N, Mabey D. Blindness and low vision in Jimma Zone,
Ethiopia: results of a population-based survey. Ophthalmic
Epidemiol 1997;4:19-26.
Journal of AAPOS
Volume 3 Number 1 February 1999
20. Prevention of blindness, prevalence and causes of blindness and low
vision, Benin. Wkly Epidemiol Rec 1991;46:337-40.
21. Chirambo MC, Tielsch JM, West KP, et al. Blindness and visual
impairment in Southern Malawi. Bull World Health Organ
1986;64:567-72.
22. Gilbert C, Rahi J, Echstein M, OSullivan J, Foster A. Retinopathy
of prematurity in middle-income countries. Lancet 1997;350:12-4.
23. World Health Organization. Blindness and visual disability. Part IV:
Socioeconomic aspects. Retinopathy of Prematurity (ROP). WHO
Fact Sheets 1997;145:3.
24. World Health Organization. Blindness and visual disability. Part V:
seeing aheadprojections into the next century. WHO Fact Sheets
1997;146:1.
25. Blindness and visual disability. Part V of VII: seeing aheadprojec-
tions into the next century. WHO Fact Sheet No. 146, Feb 1997.
26. Jackson H. Bilateral blindness due to trauma in Cambodia. Eye
1996;10:517-20.
27. Tielsch JM, Sommer A, Witt K, Katz J, Royall RM, Baltimore Eye
Survey Research Group. Blindness and visual impairment in an
American urban population. Arch Ophthalmol 1990;108:286-90.
28. Kasmann-Kellner B, Hille K, Pfau B, Ruprecht KW. Augen und
Allgemeinerkrankungen in der Landesschule fr Blinde und
Sehbehinderte des Saarlands. Ophthalmologe 1998;95:51-4.
29. Steinkuller PG. Pediatric ocular disease in Africaor how I met
Gunter K. von Noorden. Am Orthopt J 1997;47:72-7.
30. Rahi JS, Sripathi S, Gilbert CE, Foster A. The importance of prena-
tal factors in childhood blindness in India. Dev Med Child Neurol
1997;39:449-55.
31. Merin S, Lapithis AG, Horovitz D, Michaelson IC. Childhood
blindness in Cyprus. Am J Ophthalmol 1972;74:538-42.
32. Baghdassarian SA, Tabbara KT. Childhood blindness in Lebanon.
Am J Ophthalmol 1975;79:827-30.
33. Elder MJ, De Cook R. Childhood blindness in the West Bank and
Gaza Strip: prevalence, etiology, and hereditary factors. Eye
1993;7:580-3.
Other Useful References
Gurland JE, Wheeler MB, Wilson ME, Liegner JT. Children with low
vision: evaluation and treatment in the 1990s. In: Burde RM,
Slamovits TL, editors. Advances in clinical ophthalmology. Vol 1. St
Louis: Mosby Inc; 1994. p. 275-94.
Hatfield EM. Blindness in infants and young children. Sight Saving
Review 1972;42:69-89.
Hatfield EM. Causes of blindness in school children. Sight Saving
Review 1963;33:218-33.
Phillips CI, Levy AM, Newton M, et al. Blindness in school children:
importance of heredity, congenital cataract and prematurity. Br J
Ophthalmol 1987;71:564-78.
Riise R, Flage T, Hansen E, et al. Visual impairment in Nordic children.
I. Nordic registers of prevalence data. Acta Ophthalmol Scand
1992;70:145-54.
Robinson GC. Causes, ocular disorders, associated handicaps and inci-
dence and prevalence of blindness in childhood. In: Jan JE, Freeman
RD, Scott EP, editors. Visual impairment in children and adoles-
cents. New York: Grune & Stratton; 1977. p. 27-47.
Robinson GC, Jan JE. Acquired ocular visual impairment in children.
Arch Pediatr Adolesc Med 1993;147:325-8.
Robinson GC, Jan JE, Kinnis C. Congenital ocular blindness in children
1945 to 1984. Arch Pediatr Adolesc Med 1987;141:1321-4.
Rosenberg T, Flage T, Hansen E, et al. Visual impairment in Nordic chil-
dren. II. Aetiological factors. Acta Ophthalmol Scand 1992;70:155-64.
Ver Der Pol BAE. Causes of visual impairment in children. Doc
Ophthalmol 1986;61:223-8.