Most recorded glaucoma has no subtype, and in China it is probably angle closure

Design

two prospective population cohorts with linked inpatient records and ancestry-specific polygenic risk score analysis

Population

512,504 adults in the China Kadoorie Biobank and 492,329 in UK Biobank, 12 years of follow-up

Primary outcome

incidence of primary open-angle, angle-closure and unspecified glaucoma, and genetic subtype composition of unspecified cases

Effect

1,658 and 7,643 incident cases; over 68% unspecified; angle-closure twofold higher in women; open-angle fivefold higher in Black than White UK participants

Primary open-angle and primary angle-closure glaucoma are different diseases with different treatments, yet most glaucoma in administrative data is recorded without a subtype. This study identified incident primary glaucoma from linked inpatient records in the China Kadoorie Biobank (512,504 participants) and UK Biobank (492,329), then used ancestry-specific polygenic risk scores — including a new East Asian angle-closure meta-analysis — to infer what the unspecified cases actually were.

Over 12 years, 1,658 cases were identified in the Chinese cohort and 7,643 in the UK cohort. More than 68 per cent lacked subtype specification. Where subtypes were recorded, the demographics behaved as expected: incidence rose with age, angle-closure was twice as common in women in both cohorts, and in the UK open-angle incidence was fivefold higher in Black than White participants. The genetic analysis showed the unspecified group aligned closely with angle-closure glaucoma in the Chinese cohort, and was more heterogeneous in the UK.

The methodological point is that unspecified glaucoma is not a random mixture — its composition depends on the population, so any study that treats 'glaucoma' as one exposure is measuring different diseases in different places. The clinical point, for practice in South and East Asia, is the reminder that angle closure carries a large share of the burden and that gonioscopy is the examination that decides which disease you are treating.

• Perform and record gonioscopy in every glaucoma assessment — the subtype determines the treatment
• Record the subtype in the notes and the discharge coding, not just 'glaucoma'
• Maintain a low threshold for angle assessment in women, who carried twice the angle-closure incidence
• Treat published glaucoma epidemiology as population-specific rather than transferable
• In Indian practice, assume a substantial angle-closure share rather than defaulting to open-angle

Why it matters

Two-thirds of the recorded glaucoma burden is a category that means something different depending on where it was recorded.

Don't overread it

Inpatient-record phenotyping with genetic inference — it describes what the coded categories contain, not incidence in the general population.

The statistics, in plain English

Polygenic risk scores here infer group composition, not individual diagnoses: they can say the unspecified cases in China genetically resemble angle-closure on average, and cannot classify any one patient. Case counts also come from inpatient records, so this measures glaucoma severe enough to be admitted, not glaucoma in the community.

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.