Skip to main content
. 2022 Dec 22;141(2):140–147. doi: 10.1001/jamaophthalmol.2022.5567

Table 2. Association of Baseline and Year-16 Visit Characteristics With AMD: Unadjusted Model and Adjusted Model for Age, Sex, Race and Ethnicity, and Treatment Arms.

Characteristic No. AMD, mean (SD) P value
Without (n = 1108) With (n = 479) Unadjusted Adjusteda
Age at randomization, y 1587 48.0 (8.6) 51.4 (9.6) <.001 NA
Sex, No. (%)
Female 1125 801 (71) 324 (29) .06 NA
Male 462 307 (66) 155 (34)
Race and ethnicity, No. (%)
African American 333 241 (72) 92 (28) .50 NA
American Indian 126 90 (71) 36 (29)
Asian 81 52 (64) 29 (36)
Hispanic 264 177 (67) 87 (33)
Non-Hispanic White 783 548 (70) 235 (30)
Education level, y 1587 10.9 (3.0) 10.6 (3.4) .19 .80
Baseline BMI 1587 33.7 (6.1) 33.6 (6.7) .25 .71
Baseline fasting plasma glucose, mg/dL 1587 106 (8.1) 106 (8.8) .49 .34
Baseline 2-h plasma glucose, mg/dL 1587 164 (16.5) 165 (18.0) .74 .68
Baseline HbA1c, % 1583 5.9 (0.5) 5.9 (0.5) .01 .07
Baseline dietary protein intake, median (IQR), g 1553 81.0 (60.8-108.4) 78.2 (59.5-105.2) .22 .91
Baseline weight, kg 1587 92.8 (18.3) 92.2 (19.2) .43 .24
DPPOS variables at AMD assessment
Follow-up weight, kg 1587 91.3 (17.8) 89.9 (19.2) .14 .31
Follow-up HbA1c, % 1587 6.1 (0.8) 6.1 (0.7) .50 .72
Diabetes at assessment, No. (%)
No 669 456 (68) 213 (32) Reference Reference
Yes 918 652 (71) 266 (29) .22 .47
Diabetic retinopathy, No. (%)
No 1319 910 (69) 409 (31) Reference Reference
Yes 268 198 (74) 70 (26) .11 .07
Smoking status, No. (%)
Never 992 719 (72) 273 (28) .001 Reference
Ever 506 321 (63) 185 (37) .02
Current 89 68 (76) 21 (24) .23
Years with any metformin 1587 8.5 (9.3) 7.9 (9.3) .33 .69
Years with study metformin 1587 2.9 (7.0) 2.90 (7.0) .95 .37
Years with out-of-study metformin 1587 2.9 (4.8) 2.78 (4.8) .23 .71

Abbreviations: AMD, age-related macular degeneration; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); DPPOS, Diabetes Prevention Program Outcomes Study; HbA1c, hemoglobin A1c, NA, not applicable.

SI conversion factors: To convert glucose to millimoles per liter, multiply by 0.0555; hemoglobin to proportion of total hemoglobin, multiply by 0.01.

a

For categorical variables, the unadjusted P values are calculated from Pearson χ2 tests, while the adjusted P values are calculated from logistic regressions comparing each level with the reference level. For continuous variables, the unadjusted P values are calculated from 2-sample t test or Wilcoxon rank sum test (baseline dietary protein intake), while the adjusted P values are calculated from their coefficients in the logistic regression using AMD as the outcome.