Table 2.
Demographic Data
| Referable (US) Training, Testing, and Primary Validation Dataset | NonReferable (US) Training, Testing, and Primary Validation Dataset | Referable (Hong Kong) | Non Referable (Hong Kong) | Referable (India) | NonReferable (India) | |
|---|---|---|---|---|---|---|
| Age (in years) | 68.25 (±15.5) | 67.59 (±15.40) | 65.9 (±9.30) | 65.27(±11.30) | 63.84 (±11.72) | 54.763(± 14.95) |
| P value | Reference* | P = 0.5098* Reference** | P = 0.0147* | P = 0.0603** | P < 0.001* | P < 0.001* |
| Asian | 43.5% | 48.60% | 100% | 100% | 100% | 100% |
| P value | Reference* | P = 0.1138 Reference** | P < 0.001* | P < 0.001** | P < 0.001* | P < 0.001** |
| White | 38.6% | 33.40% | 0 | 0 | 0 | 0 |
| P value | Reference* | P = 0.0956 Reference** | P < 0.001* | P < 0.001** | P < 0.001* | P < 0.001** |
| African American | 5.04% | 6.40% | 0 | 0 | 0 | 0 |
| P value | Reference* | P = 0.3592 Reference** | P < 0.001* | P < 0.001** | P < 0.001* | P < 0.001** |
| Hispanic | 9.14% | 7.20% Reference** | 0 | 0 | 0 | 0 |
| P value | Reference* | P = 0.2804* Reference** | P < 0.001* | P < 0.001** | P < 0.001* | P < 0.001** |
| Data of ethnicity unavailable | 3.68% | 4.4% | 0 | 0 | 0 | 0 |
| P value | Reference* | P = 0.5683 Reference** | P < 0.001* | P < 0.001** | P < 0.001* | P < 0.001** |
| Average MD | −7.175 (±7.451) | −1.24 (±2.09) | −8.005(±6.81) | −0.900(±1.30) | −12.74(±9.22) | −1.20(±1.30) |
| P value | Reference* | P < 0.001 * Reference** | P = 0.1209* | P = 0.0363** | P < 0.001* | P = 0.8693** |
| Mean refractive error | −1.67 (±3.2) | −0.4713 ±2.53) | −0.85(± 2.57) | −0.51(±2.15) | −0.483(± 2.25) | −0.440(±2.19) |
| P value | Reference* | P < 0.001* Reference** | P < 0.001* | P = 0.8538** | P < 0.001* | P = 0.8882** |
Demographic data such as age, ethnicity distribution (in %), mean values with SDs for VF parameter in terms of mean field defects and refractive error in terms of spherical equivalent in referable and nonreferable groups from the United States, Hong Kong, and India datasets. The statistical analysis was performed with the Statistical Package for Social Sciences (SPSS) 10.1 (SPSS Inc., Chicago, IL, USA). Results are expressed as mean (±SD) and paired Student's t-test was used to evaluate the level of significance. A P value of 0.001 or less was considered significant. Chi-square test was used for comparisons of categorical demographic data for proportions. US, United States.