The positive predictive value of photoscreening devices for amblyogenic conditions

Abstract

To determine the positive predictive value (PPV) of four photoscreening devices in detecting amblyogenic risk factors (ARFs). This is a retrospective cohort study of 1,200 failed vision screenings (FVS) referred to a pediatric eye care practice from general pediatric practices over a 5-year period. Comprehensive eye examinations and cycloplegic refractions were performed. Examination findings and FVS parameters were collected through chart review after completion of visits. Inclusion criteria were age 1-8 years, eye examination within 3 months of initial FVS, and availability of FVS and visit documentation. A total of 689 patients met inclusion criteria and were included in final analysis. The PPVs were 64.3% (95% CI, 62.73-65.81) for iScreen (n = 455), 26.8% (95% CI, 26.32-27.25) for Spot (n = 113), 18.9% (95% CI, 17.95-19.92) for Plusoptix (n = 38), and 18.5% (95% CI, 17.86-19.19) for GoCheck Kids (n = 55). For comparison, patients referred for failed Titmus tests were also analyzed. The PPV for Titmus testing was 39.3% (95% CI, 36.65-41.99; n = 28). The only screening device with a PPV >50% was the iScreen. PPV is a useful metric, because it enables referring providers to guide patients on a post-screening plan and to choose devices that minimize over-referral. Device PPV in the general pediatric population can be markedly different from device PPV in already existing pediatric eye care populations due to inherently different prevalence of ARFs. The iScreen also detected conditions such as cataracts, anisocoria, eyelid abnormalities, esotropia, exotropia, hypertropia, and amblyogenic astigmatism.