Is screening the best approach?
More evidence required before adopting screening for glaucoma
The value of population screening for glaucoma has yet to be determined, and the results of a large, randomised clinical trial will be needed to determine whether the number of new cases it yields will justify its costs, said professor Anja Tuulonen MD, PhD, Tampere University Hospital, Tampere, Finland.
“A major randomised clinical trial is far less costly than unplanned growth of poor quality screen. Well conducted randomised controlled screening trials provide the best evidence,” Dr Tuulonen told the 13th European Glaucoma Society Congress in Florence, Italy.
She noted that the consensus meeting of the World Glaucoma Association in April, 2008, identified three major shortcomings in modern glaucoma care. Namely, unequal access to care with large variations in service, an underdiagnosis rate of 50%, and an equal, 50% rate of overtreatment.
The consensus meeting also concluded that there was insufficient evidence regarding the cost-effectiveness of screening for glaucoma. In particular there is a lack of evidence for most important parameters, based on real-world data and evidence.
Since that time, a longitudinal study published in 2014 showed that screening had no impact on cumulative diagnosis of glaucoma during the years 1981 to 2002, among 856 individuals who were born in 1915 in one city of Sweden. In addition, authors of a UK study concluded that it might not be a cost-effective use of resources to run a screening trial in the UK.
Dr Tuulonen and her associates have initiated the Northern Finland Birth Cohort Eye Study, which is a randomised controlled screening trial in the 1966 Northern Finland Birth Cohort. The original cohort includes 12,058 people born in a northern region of Finland in 1966. In the NFBC Eye Study screening group, 1.1% had definite glaucoma and of those, 95% were undiagnosed previously and 89% were normotensive. Longer follow-up may reveal the best combination of diagnostic tests for detecting glaucoma in an unscreened population, she said.
There are a number of inherent biases that need to be considered when designing a randomised controlled trial of screening, she noted. For example, screening is more likely to pick up cases with a good prognosis, since it is usually the more health-conscious people who attend screening. In addition, it will usually only pick up the slowly progressing disease.
Meanwhile, in the absence of any systematic population screening, opticians are offering screening with direct-to-consumer advertising, inducing patients to undergo unnecessary and expensive tests. That in turn can lead to an overdiagnosis epidemic.
She referred to American Academy of Ophthalmology guidelines refraining from unnecessary testing in patients without visual complaints.
Anja Tuulonen: Anja.firstname.lastname@example.org