Abstract
Purpose:
To compare the accuracy of the Plusoptix S09 vision screening device vs. the LEA /Random Dot E screening system, for detection of referral-warranted ocular disease in children ages 3-5.
Methods:
In a multi-year study from October 2010 to December 2012, 348 Head Start children between 3 to 5 years old were enrolled in a vision screening study. Each child was screened using the Plusoptix S09 and the LEA Distance VA/Random Dot E screening system. The Plusoptix referral criteria is: Hyperopia, SE ≥ 2.50 D; Astigmatism, Cylinder ≥ 1.50 D;Myopia, SE ≤ -1.50 D; Anisometropia, ≥ 1.00 D (difference in SE). Referral by the LEA Distance VA/ Random Dot E system is caused by an acuity below the critical line of 20/40 in one or both eyes or less than 4 of 5 correct stereopsis attempts. Eye exams including cycloplegic retinoscopy, visual acuity testing, a cover test, and examination of the anterior and posterior segments were performed on each child to detect any ocular abnormality as a reference standard. These exams were performed by volunteer pediatric ophthalmologists and optometrists who were masked from the screening results. Ocular abnormality by examination is defined by the following referral criteria: Hyperopia, SE ≥ 2.50 D; Astigmatism, Cylinder ≥ 1.50 D; Myopia, SE ≤-1.50 D; Anisometropia,≥ 1.00 D (difference in SE); Amblyopia Suspect, 2 line acuity difference, acuity, anisometropia, moderate astigmatism, moderate hyperopia; abnormal anterior posterior segment exam, ptosis, or strabismus.
Results:
Based on the 308 dilated eye exams on children who completed both screening methodologies, the sensitivity of the LEA method was 51.5% (95% CI: 38.9-64.0) and the Plusoptix device was 80.3% (95%CI: 68.7-89.1). The specificity of the LEA method was 81.4% (95% CI: 75.9-86.1) and the specificity of the Plusoptix S09 device was 72.7% (95%CI: 66.7-78.2). A comparison of the receiver operating characteristic area under the curve (AUC) suggests that the Plusoptix S09 is significantly more reliable overall in detecting eye disease than the LEA methodology (p=0.017). Overall, the study found the LEA system to be accurate 66% of the time and the Plusoptix device to be accurate 77% of the time.
Conclusions:
The Plusoptix S09 is significantly more accurate in detecting referral-warranted eye disease in children ages 3-5 than the LEA/Random Dot E screening system.
Keywords: 417 amblyopia •
460 clinical (human) or epidemiologic studies: health care delivery/economics/manpower •
754 visual acuity