Accuracy of diagnosis of pterygium by optometrists and general practitioners in Australia

Lawrence W. Hirst, Jane W Smith

Research output: Contribution to journalArticleResearchpeer-review

1 Citation (Scopus)
239 Downloads (Pure)


Background: It was the author’s (LWH) observation that pterygium was frequently misdiagnosed
by general practitioners that led to this study. The aim was to identify the rate of
misdiagnosis of pterygium by optometrists and general practitioners based upon assessing
referral accuracy to a single ophthalmologist (LWH).
Methods: This study involved a prospective case series from 25 March 2015 to 18 December
2018 in a tertiary referral practice specialising in pterygium. The accuracy of diagnosis of
pterygium, based upon the content of the clinical referral, was undertaken for optometrists
and general practitioners. The benchmark for diagnostic accuracy was the diagnosis made
by the author (LWH) during a consultation in person by the author (LWH) using a hand-light
examination and confirmed by slitlamp examination.
Results: A total of 1,511 consecutive patients were included in the study with 90/549 incorrectly
diagnosed (16 per cent) by general practitioners and 14/962 (1.4 per cent) by optometrists.
General practitioners were 13.28 times more likely to incorrectly diagnose a
pterygium than optometrists (95% CI 7.48–23.57). Almost exclusively, the incorrect diagnosis
made by general practitioners was naming a pinguecula, a pterygium. The same misdiagnosis
was made by optometrists but far less frequently.
Conclusion: General practitioners misdiagnosed pterygium far more often than optometrists
which may reflect a reduction in training in eye health.
Original languageEnglish
Pages (from-to)197-200
Number of pages4
JournalThe Australian journal of optometry
Issue number2
Early online date6 May 2019
Publication statusPublished - 1 Mar 2020


Dive into the research topics of 'Accuracy of diagnosis of pterygium by optometrists and general practitioners in Australia'. Together they form a unique fingerprint.

Cite this