A third of blindness in two Jammu blind schools was avoidable

Design

cross-sectional observational survey with standardised ophthalmological examination

Population

82 residents of two blind schools in Jammu province, India; mean age 14.3 years, 57.3% male

Primary outcome

aetiology of blindness, categorised as congenital, acquired or undetermined, and as avoidable or unavoidable

Effect

67% congenital, 27% acquired, 6% undetermined; whole globe 26.8%, retina 26.8%, cornea 19.6%; 35% of causes avoidable

Blind school surveys are an imperfect but useful window on childhood blindness, because they capture the children who have already reached the end of the diagnostic pathway. This cross-sectional study examined all 82 residents of two blind schools in Jammu province - one for boys, one for girls - with standardised visual acuity assessment and anterior and posterior segment examination by trained doctors.

Mean age was 14.3 years and 57.3% were male. A family history of blindness was present in 12.2%. Congenital causes accounted for 67% of cases, acquired for 27%, and 6% were undetermined. Anatomically the whole globe and the retina were each affected in 26.8%, and the cornea in 19.6%. Thirty-five per cent of causes were classified as avoidable - that is, preventable or treatable had they been identified in time.

That 35% is the number to carry. The anatomical distribution explains where it comes from: corneal scarring is the classic avoidable cause in this setting, and retinal dystrophies and congenital globe anomalies are not. A blind school population is a biased sample - it over-represents children whose families found and could access such a school - so this cannot be read as a prevalence estimate for the province. What it does provide is local aetiology, which is what a screening or prevention programme has to be built on, and which had not previously been described for Jammu.

• Treat the 35% avoidable proportion as the actionable finding, and corneal scarring as where most of it sits.
• Do not read a blind school survey as a population prevalence estimate - the sample is selected by access to the school.
• A family history in 12.2% argues for genetic counselling being part of the service, not an add-on.
• Congenital causes at 67% shift the emphasis towards antenatal and neonatal detection rather than school-age screening.
• Local aetiology data are what a prevention programme needs; national figures will not substitute for them.

The statistics, in plain English

With 82 participants, every percentage in this study moves by more than a point if one child is reclassified - 35% avoidable means 29 children, and the confidence around that proportion is wide. The design also selects its sample: blind school residents are children whose blindness was severe enough, and whose families were resourced enough, to reach specialist education, so causes associated with early death or with families unable to access services will be under-represented. That does not weaken the finding for its stated purpose, which is describing local aetiology to guide services, but it rules out using these figures as an estimate of childhood blindness in the province.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.