Table 2.

Studies on UM in Australian populations.

Title	Year	Data set	Type of Study	Conclusions	Reference
UV light exposure as a risk factor for ocular melanoma in QLD, Australia	1972 to 1996	QLD cases from the QLD Cancer Registry and from pathology laboratory ocular specimens in QLD 216 choroidal 35 ciliary body 27 iris 35 conjunctival	Determining risk factors using case-control study (125 patients, 375 controls), questionnaire without disclosing study hypotheses OR and 95% CI	No correlation between estimated lifetime cumulative solar radiation exposure and ocular melanoma in QLD from 1972 to 1996 but a protective effect of dark skin, brown eyes, and resistance to sunburn. A family history of ocular melanoma was a strong risk factor.	[33]
Eye colour and cutaneous nevi predict risk of ocular melanoma in Australia	1996 to 1998	NSW, VIC, QLD and Other from “all ophthalmologists and population-based cancer registries in Australia” 222 choroid 22 ciliary body 1 ciliochoroidal 25 iris 19 conjunctival	Determining risk factors using case-control study (290 patients, 914 controls), questionnaire without disclosing study hypotheses OR and 95% CI using STATA and Mantel-Haenszel	Light eyes, cutaneous nevi and inability to tan were found to be risk factors for UM.	[34]
Sun exposure predicts risk of ocular melanoma in Australia	As above	Same dataset as above	Same model as above	Long hours outdoors increase risk of UM.	[37]
Artificial UV light radiation and ocular melanoma in Australia	As above	Same dataset as above	Same model as above, but with 290 patients and 893 controls	Risk of ciliary body and choroidal (but not iris or conjunctival) melanomas increased with exposure to sunlamps or welding independent of personal sun exposure.	[38]
Incidence of ocular melanoma in Australia from 1990 to 1998	1990 to 1998	From Australian population-based cancer registries and all practising ophthalmologists 1990 to 1995 = 768 1996 to 1998 = 539 (choroid or ciliary body 459, iris 42, conjunctiva 37)	Determine the incidence and incidence trends of ocular melanoma based on distribution, latitude and in subpopulations. Incidence trends only used cancer registry-reported cases and excluded QLD	Incidence increased weakly across latitude bands from <30°S to >36°S. Rural residence increased incidence by 50%.	[21]
Survival from UM in WA 1981–2005	1981 to 2005	WA Cancer Registry 229 Choroid 33 Ciliary 46 Iris or Unknown	Determining relative survival estimation and proportional hazards regression models using STATA Eerer II methodology(308 patients)	When life tables were used to account for the baseline death rates in the general population, the relative survival rates at 3, 5 and 10 years were 88% (95% CI: 83–92%), 81% (95% CI: 76–87%) and 71% (95% CI: 63–78%), respectively	[57]
Prevalence of germline BAP1 mutation in a population-based sample of UM cases	(Not specified) 2012 or earlier based on publication date	NSW Patients diagnosed at ≤50 years old (n = 66)	Sanger sequencing to determine BAP1 mutations	2/66 possessed BAP1 mutations	[102]

QLD, Queensland; VIC, Victoria; NSW, New South Wales; WA, Western Australia.