Table 3.

Associations between patient characteristics and patient-provider discordance in provision of information about KT

CharacteristicDiscordant Provider-Only Report of KTPI (n=326)a
UnadjustedAdjustedParsimoniousb
Age (yr; per continuous per 5-yr increase); spline with knot at 65
 <651.10 (1.01 to 1.21)1.10 (1.00 to 1.21)1.12 (1.02 to 1.24)
 ≥651.22 (1.08 to 1.38)1.23 (1.08 to 1.39)1.26 (1.11 to 1.44)
Female sex1.14 (0.84 to 1.56)1.17 (0.87 to 1.57)
African American race0.71 (0.52 to 0.97)0.80 (0.59 to 1.09)
Highest education level
 Grade school or lessrefref
 High school0.54 (0.33 to 0.91)0.59 (0.36 to 0.98)
 Postsecondary0.78 (0.57 to 1.08)0.88 (0.64 to 1.21)
Number of comorbidities1.09 (0.99 to 1.19)1.07 (0.97 to 1.17)
BMI≥300.74 (0.53 to 1.03)0.91 (0.65 to 1.28)
Current smoking1.49 (1.08 to 2.05)1.67 (1.18 to 2.36)1.66 (1.20 to 2.30)
Time between first seeing nephrologist and dialysis initiation (mo)
 <12refrefref
 ≥120.75 (0.54 to 1.03)0.65 (0.48 to 0.88)0.65 (0.48 to 0.88)
  • Data are presented as the RR (95% CI). RRs and 95% CIs were modeled using modified Poisson regression. Age was modeled as a continuous variable using a linear spline term with a knot at age 65 years, and the risk was modeled per 5-year increase in age.

  • a Limited to participants who had provider-reported KTPI (i.e., excluding participants with patient-only reported KTPI or neither patient- nor provider-reported KTPI).

  • b Parsimonious model derived using Akaike information criteria from the adjusted model.