advertisement

Editors Selection IGR 17-3

Screening and detection: Detecting glaucoma: how early can you go?

Comment by Gustavo de Moraes on:

106428 Estimating the Length of the Preclinical Detectable Phase for Open-Angle Glaucoma, Aspberg J; Aspberg J; Aspberg J; Heijl A; Bengtsson B, JAMA ophthalmology, 2023; 141: 48-54

Find related abstracts

The concept of Preclinical Detectable Phase (PCDP) is one of critical importance in Public Health for any major screenable disease, as it is needed to define the ideal screening frequency, to estimate over- or under-detection for a given screening technique, and to assess the effectiveness of screening. In brief, PCDP is the time period between when a disease can be detected by a screening test to the time of clinical diagnosis, often based upon the symptoms. Because open-angle glaucoma is asymptomatic, clinical diagnosis is often defined based upon a conventional, reference diagnostic method capable of detecting the disease with reasonable confidence. In the present population-based study, Aspberg and colleagues were perhaps the first to provide a more accurate estimate of the PCDP for open-angle glaucoma in a large cohort in Sweden. To achieve this goal, they analyzed the screening and longitudinal data of participants initially screened for eligibility for the Early Manifest Glaucoma Trial (EMGT) in a large population in Malmö, Sweden, from October 1992 to January 1997. Of the 42,497 subjects invited to screening, 32,918 were successfully screened. This is a very high rate of attendance for this type of design (77.5%), which adds to the main strengths of their analyses.

PCDP is the time period between when a disease can be detected by a screening test to the time of clinical diagnosis

A positive screening was defined as (1) an IOP greater than 25 mmHg in one or both eyes; (2) suspected or evident glaucomatous optic disc, retinal nerve fiber defects, or optic disc hemorrhages seen in the fundus photographs; (3) exfoliation syndrome; or (4) manifest glaucoma in one or more first degree relatives. Those screened positive were asked to return for two post-screening visits in which they underwent standard automated perimetry, and diagnosis was defined if two consecutive tests showed an abnormal result based on the Glaucoma Hemifield Test (GHT). A total of 1,388 participants (4.2%) had a positive test result at the screening and were invited to post-screening visits, and 427 had a glaucoma diagnosis. 993 subjects who screened negative had the diagnosis made after screening. As another major strength of the study, the clinical follow-up time after the screening was more than 20 years, allowing ample time to clinical diagnosis.

With a mean PCDP of ten years, their findings suggest that screening for glaucoma based on the definitions employed in this particular study could potentially be performed at five-year intervals with a reasonable cost-benefit ratio

Their main finding was that the mean length of the PCDP for all patients was 10.7 years (95% CI, 8.71-13.0) based upon the prevalence/incidence method and 10.1 years (95% Credible Interval, 8.9-11.2) based upon the Markov Chain Monte Carlo (MCMC) model. Also, the sensitivity of the screening method estimated by the MCMC method was 94% (95% Credible Interval, 93-96). In practical terms, any disease with a short PCDP would imply that the disease develops to the clinical stages rapidly; thus, screening initiatives would be required very often. With a mean PCDP of ten years, their findings suggest that screening for glaucoma based on the definitions employed in this particular study could potentially be performed at five-year intervals with a reasonable cost-benefit ratio. Although the study neither aimed at defining cost-effectiveness nor frequency of screening, the assumptions above are the most reasonable considering the currently available evidence. This was the first attempt to systematically estimate the PCDP for open-angle glaucoma and is based upon powerful data to support its accuracy.

• ← Previous
• Next →