Table 2.

Associations of provider and predialysis care characteristics with estimated GFR at dialysis initiation

CharacteristicsModel
Logistic Regression, AOR (95% CI)a,b,cLinear Regression, Estimated GFR, Parameter Estimateb
Provider sex
 Men1Reference
 Women0.97 (0.93 to 1.01)−0.11
Years in practice
 0–81Reference
 9–210.92 (0.89 to 0.95)d−0.20d
 ≥220.86 (0.83 to 0.90)d−0.33d
Foreign medical graduate1.16 (1.12 to 1.19)d0.33d
Nephrologists per 100 ESRD patients, ratio
 01.00 (0.92 to 1.08)0.02
 >0 to ≤0.60.91 (0.86 to 0.97)d−0.28e
 >0.6 to ≤1.00.96 (0.93 to 1.00)−0.08
 >1.0 to ≤1.51Reference
 >1.50.91 (0.88 to 0.95)d−0.22d
Predialysis nephrology care
 None1Reference
 <12 mo1.17 (1.13 to 1.21)d0.35d
 ≥12 mo1.05 (1.01 to 1.10)e0.02
Access at initiation
 Catheter1Reference
 Fistula1.07 (1.02 to 1.12)e0.29e
 Arteriovenous graft1.18 (1.10 to 1.26)d0.08
  • AOR, adjusted odds ratio; CI, confidence interval.

  • a Estimated GFR>10 versus≤10 ml/min per 1.73 m2.

  • b Adjusted for patient age, race, ethnicity, primary cause of ESRD, socioeconomic status, eligibility for Medicaid, urban status, comorbid conditions (coronary artery disease, congestive heart failure, cerebrovascular accident/transient ischemic attack, other cardiovascular disease, peripheral vascular disease, diabetes, chronic obstructive pulmonary disease, cancer, inability to ambulate, and inability to transfer), hemoglobin, albumin, body mass index, and variables in the table.

  • c The C statistic for the model that tested associations between patient and provider characteristics and estimated GFR at dialysis initiation was 0.668.

  • d P<0.001.

  • e P<0.05.