Boxplots: B1 versus B2
Summary Statistics
| Group | Variable | N | Mean | SD | Minimum | Q1 | Median | Q3 | IQR | Maximum |
|---|---|---|---|---|---|---|---|---|---|---|
| B1 | CysCeGFR | 605 | 42.00 | 25.25 | 4.92 | 22.21 | 34.94 | 55.77 | 33.57 | 122.56 |
| B1 | Difference | 605 | 12.57 | 4.28 | 3.65 | 8.97 | 12.94 | 16.11 | 7.14 | 19.97 |
| B1 | ScreGFR | 605 | 54.58 | 25.97 | 9.42 | 34.88 | 48.41 | 69.57 | 34.69 | 138.92 |
| B2 | CysCeGFR | 588 | 50.85 | 20.72 | 11.37 | 34.97 | 48.94 | 64.58 | 29.61 | 118.24 |
| B2 | Difference | 588 | 33.09 | 11.13 | 20.05 | 24.83 | 31.10 | 39.00 | 14.17 | 130.63 |
| B2 | ScreGFR | 588 | 83.94 | 23.17 | 31.62 | 67.02 | 85.93 | 99.44 | 32.42 | 178.39 |
Comparison of Characteristics
| Characteristic |
Group
|
p-value2 | |
|---|---|---|---|
| B1 N = 6051 |
B2 N = 5881 |
||
| Gender | 0.009 | ||
| Female | 62 (10.2%) | 90 (15.3%) | |
| Male | 543 (89.8%) | 498 (84.7%) | |
| Missing / unrecognized | 0 | 0 | |
| Smoking | 0.283 | ||
| Never | 236 (39.1%) | 234 (39.9%) | |
| Former | 290 (48.0%) | 260 (44.4%) | |
| Current | 78 (12.9%) | 92 (15.7%) | |
| Missing / unrecognized | 1 | 2 | |
| BMI, mean (SD) | 33.1 (8.3) | 37.5 (9.1) | <0.001 |
| Missing / unrecognized | 1 | 0 | |
| Steroid | 0.805 | ||
| No | 578 (95.5%) | 560 (95.2%) | |
| Yes | 27 (4.5%) | 28 (4.8%) | |
| Missing / unrecognized | 0 | 0 | |
| Rheum Diseases | 0.265 | ||
| No | 38 (47.5%) | 38 (56.7%) | |
| Yes | 42 (52.5%) | 29 (43.3%) | |
| Missing / unrecognized | 525 | 521 | |
| Thyroid Disease | 0.645 | ||
| No | 500 (87.0%) | 480 (86.0%) | |
| Yes | 75 (13.0%) | 78 (14.0%) | |
| Missing / unrecognized | 30 | 30 | |
| Malignancy | <0.001 | ||
| No | 426 (74.9%) | 462 (84.0%) | |
| Yes | 143 (25.1%) | 88 (16.0%) | |
| Missing / unrecognized | 36 | 38 | |
| Liver Disease | 0.010 | ||
| No | 572 (94.5%) | 533 (90.6%) | |
| Yes | 33 (5.5%) | 55 (9.4%) | |
| Missing / unrecognized | 0 | 0 | |
| HIV | 0.374 | ||
| No | 601 (99.3%) | 587 (99.8%) | |
| Yes | 4 (0.7%) | 1 (0.2%) | |
| Missing / unrecognized | 0 | 0 | |
| Hyperlipidemia | <0.001 | ||
| No | 77 (12.7%) | 119 (20.2%) | |
| Yes | 528 (87.3%) | 469 (79.8%) | |
| Missing / unrecognized | 0 | 0 | |
| Hypertension | <0.001 | ||
| No | 55 (9.1%) | 106 (18.0%) | |
| Yes | 550 (90.9%) | 482 (82.0%) | |
| Missing / unrecognized | 0 | 0 | |
| DM2 | 0.001 | ||
| No | 213 (35.2%) | 260 (44.2%) | |
| Yes | 392 (64.8%) | 328 (55.8%) | |
| Missing / unrecognized | 0 | 0 | |
| 1 n (%); Mean (SD) | |||
| 2 Pearson’s Chi-squared test; Welch Two Sample t-test; Fisher’s exact test | |||
Logistic Regression
| Predictor | Adjusted OR | 95% CI | p-value |
|---|---|---|---|
| genderMale | 1.135 | 0.761 to 1.692 | 0.536 |
| SmokingFormer | 0.898 | 0.685 to 1.178 | 0.438 |
| SmokingCurrent | 0.716 | 0.486 to 1.054 | 0.091 |
| BMI | 0.946 | 0.932 to 0.960 | <0.001 |
| SteroidYes | 0.772 | 0.432 to 1.377 | 0.381 |
| Thyroid DiseaseYes | 0.896 | 0.621 to 1.294 | 0.560 |
| MalignancyYes | 1.467 | 1.071 to 2.011 | 0.017 |
| Liver DiseaseYes | 0.440 | 0.266 to 0.729 | 0.001 |
| HIVYes | 1.957 | 0.213 to 17.970 | 0.553 |
| HyperlipidemiaYes | 1.043 | 0.715 to 1.521 | 0.828 |
| HypertensionYes | 1.647 | 1.091 to 2.485 | 0.017 |
| DM2Yes | 1.155 | 0.872 to 1.529 | 0.316 |
Model N = 1116 (B1: 567; B2: 549). Excluded for missing predictors: 77.
Table notes: Only Group B (CysCeGFR < ScreGFR) is included. Difference = ScreGFR − CysCeGFR, in absolute eGFR units, not percent. B1: 0 < Difference < 20; B2: Difference > 20. Exactly 20 is excluded (0 selected records). Each patient’s latest valid paired laboratory record is selected before restricting to Group B; its date is the later of the two laboratory timestamps. Categorical percentages use nonmissing subgroup denominators. BMI is mean (SD). Comparison p-values use Welch’s t-test for BMI and chi-square or Fisher’s exact tests for categorical characteristics. Logistic outcome: B1 = 1, B2 = 0. Rheum Diseases is excluded from regression and its complete-case check but retained in the comparison table. All other 11 characteristics, including HIV, are candidate predictors. The model uses complete cases for those predictors and omits predictors with no variation in that sample. Reference levels are Female, Never smoking, and No where present. Adjusted odds ratios compare the odds of B1 versus B2; confidence intervals are Wald 95% intervals. Missing/unknown values remain missing. P-values are unadjusted.