Skip to main content
. Author manuscript; available in PMC: 2023 Jun 1.
Published in final edited form as: Cornea. 2021 Jun 15;41(6):692–698. doi: 10.1097/ICO.0000000000002762

Table 4:

Final Generalized Estimating Equations Model to Predict CDVA

beta 95% CI Variable range Effect over range
Independent Variables Change beta low high p - value minimum maximum
Age 5 year increase 0.026 0.014 0.037 <0.001 *** 35 91 0.29
CD 0–2 mm 3 unit increase 0.022 0.011 0.034 <0.001 *** 17.3 48 0.23
Guttae Severity 1 unit increase 0.024 0.000 0.047 0.046 * 0.5 4 0.08
Cataract Grade 1 unit increase 0.050 0.029 0.070 <0.001 *** 0 4 0.20
Corneal Edema 1 unit increase 0.117 0.063 0.172 <0.001 *** 0 3 0.35
Refraction Group Hyperopia to myopia 0.061 0.016 0.106 0.008 ** hyperopia myopia 0.06
Refraction Group No refractive error to hyperopia −0.026 −0.071 0.019 0.256 N/A – effect not significant
N/A – effect not significant
Refraction Group No refractive error to myopia 0.035 −0.006 0.076 0.097
Intercept 0.172 0.137 0.207 <0.001 ***
Total effect, over the ranges, of all statistically significant variables: 1.21
*

p < 0.05,

**

p < 0.01,

***

p < 0.001,

p < 0.15 (marginally significant).

CCD = central corneal densitometry; CDVA = corrected distance visual acuity; CI = confidence interval; hyperopia = refractive error GT plus 0.5; LogMAR = logarithm of the minimum angle of resolution; mm = millimeters; myopia = refractive error LT minus 0.5; N/A = not applicable; No refractive error = refractive error between −0.5 to +0.5; TOTAL = Total effect from all the “better” categories to all the “worse” (bolded) categories; NOTE: All Refraction Group effects must remain in the model as long as any of the effects are statistically significant.