Skip to Main Content
Table 1

Development of retinopathy or microalbuminuria outcomes as a function of eTIR and/or HbA1c

Per 10% lower eTIR*Per 0.5% higher HbA1c+
Cohort and statisticsUnadjustedAdjusted for HbA1c aloneAdjusted for HbA1c and other covariates#UnadjustedAdjusted for eTIR aloneAdjusted for eTIR and other covariates#
Retinopathy: three or more steps sustained progression       
 All, n = 271 cases of 1,441       
  HR (95% CI) 1.58 (1.46, 1.71) 1.11 (1, 1.24) 1.12 (1.00, 1.26) 1.32 (1.27, 1.37) 1.27 (1.21, 1.34) 1.31 (1.24, 1.39) 
  χ2 (P139.8 (<0.001) 3.7 (0.059) 4.3 (0.042) 229.2 (<0.001) 77.4 (<0.001) 89.3 (<0.001) 
 Primary, n = 107 cases of 726       
  HR (95% CI) 1.71 (1.5, 1.94) 1.12 (0.94, 1.33) 1.10 (0.92, 1.31) 1.38 (1.30, 1.46) 1.33 (1.22, 1.45) 1.36 (1.25, 1.49) 
  χ2 (P68.8 (<0.001) 1.7 (0.206) 1.2 (0.283) 112.5 (<0.001) 43.2 (<0.001) 49.9 (<0.001) 
 Secondary, n = 164 cases of 715       
  HR (95% CI) 1.53 (1.39, 1.69) 1.12 (0.97, 1.29) 1.14 (0.99, 1.32) 1.30 (1.24, 1.37) 1.25 (1.16, 1.34) 1.27 (1.18, 1.37) 
  χ2 (P74.2 (<0.001) 2.6 (0.113) 3.4 (0.072) 119.8 (<0.001) 37.6 (<0.001) 41.7 (<0.001) 
Nephropathy: sustained microalbuminuria       
 All, n = 116 cases of 1,441       
  HR (95% CI) 1.37 (1.23, 1.53) 1.11 (0.95, 1.30) 1.11 (0.95, 1.30) 1.22 (1.15, 1.28) 1.17 (1.07, 1.27) 1.19 (1.09, 1.29) 
  χ2 (P33.7 (<0.001) 1.7 (0.202) 1.8 (0.191) 49.3 (<0.001) 13.5 (0.003) 15.7 (<0.001) 
 Primary, n = 36 cases of 726       
  HR (95% CI) 1.61 (1.31, 1.97) 1.40 (1.04, 1.89) 1.39 (1.03, 1.88) 1.25 (1.13, 1.37) 1.10 (0.95, 1.28) 1.11 (0.95, 1.28) 
  χ2 (P21.3 (<0.001) 5.0 (0.027) 4.7 (0.032) 21.3 (<0.001) 1.6 (0.217) 1.7 (0.193) 
 Secondary, n = 80 cases of 715       
  HR (95% CI) 1.31 (1.15, 1.50) 0.99 (0.81, 1.20) 1.01 (0.83, 1.22) 1.24 (1.15, 1.33) 1.24 (1.12, 1.38) 1.22 (1.10, 1.36) 
  χ2 (P16.4 (<0.001) 0.1 (0.901) 0.1 (0.926) 34.7 (<0.001) 16.6 (<0.001) 14.2 (<0.001) 
Per 10% lower eTIR*Per 0.5% higher HbA1c+
Cohort and statisticsUnadjustedAdjusted for HbA1c aloneAdjusted for HbA1c and other covariates#UnadjustedAdjusted for eTIR aloneAdjusted for eTIR and other covariates#
Retinopathy: three or more steps sustained progression       
 All, n = 271 cases of 1,441       
  HR (95% CI) 1.58 (1.46, 1.71) 1.11 (1, 1.24) 1.12 (1.00, 1.26) 1.32 (1.27, 1.37) 1.27 (1.21, 1.34) 1.31 (1.24, 1.39) 
  χ2 (P139.8 (<0.001) 3.7 (0.059) 4.3 (0.042) 229.2 (<0.001) 77.4 (<0.001) 89.3 (<0.001) 
 Primary, n = 107 cases of 726       
  HR (95% CI) 1.71 (1.5, 1.94) 1.12 (0.94, 1.33) 1.10 (0.92, 1.31) 1.38 (1.30, 1.46) 1.33 (1.22, 1.45) 1.36 (1.25, 1.49) 
  χ2 (P68.8 (<0.001) 1.7 (0.206) 1.2 (0.283) 112.5 (<0.001) 43.2 (<0.001) 49.9 (<0.001) 
 Secondary, n = 164 cases of 715       
  HR (95% CI) 1.53 (1.39, 1.69) 1.12 (0.97, 1.29) 1.14 (0.99, 1.32) 1.30 (1.24, 1.37) 1.25 (1.16, 1.34) 1.27 (1.18, 1.37) 
  χ2 (P74.2 (<0.001) 2.6 (0.113) 3.4 (0.072) 119.8 (<0.001) 37.6 (<0.001) 41.7 (<0.001) 
Nephropathy: sustained microalbuminuria       
 All, n = 116 cases of 1,441       
  HR (95% CI) 1.37 (1.23, 1.53) 1.11 (0.95, 1.30) 1.11 (0.95, 1.30) 1.22 (1.15, 1.28) 1.17 (1.07, 1.27) 1.19 (1.09, 1.29) 
  χ2 (P33.7 (<0.001) 1.7 (0.202) 1.8 (0.191) 49.3 (<0.001) 13.5 (0.003) 15.7 (<0.001) 
 Primary, n = 36 cases of 726       
  HR (95% CI) 1.61 (1.31, 1.97) 1.40 (1.04, 1.89) 1.39 (1.03, 1.88) 1.25 (1.13, 1.37) 1.10 (0.95, 1.28) 1.11 (0.95, 1.28) 
  χ2 (P21.3 (<0.001) 5.0 (0.027) 4.7 (0.032) 21.3 (<0.001) 1.6 (0.217) 1.7 (0.193) 
 Secondary, n = 80 cases of 715       
  HR (95% CI) 1.31 (1.15, 1.50) 0.99 (0.81, 1.20) 1.01 (0.83, 1.22) 1.24 (1.15, 1.33) 1.24 (1.12, 1.38) 1.22 (1.10, 1.36) 
  χ2 (P16.4 (<0.001) 0.1 (0.901) 0.1 (0.926) 34.7 (<0.001) 16.6 (<0.001) 14.2 (<0.001) 

Coefficient β-estimate, SE of the estimate, 95% with limits βL and βU, and P value for each factor (eTIR or HbA1c) from a Cox proportional hazards model for sustained progression of retinopathy (three or more steps of progression on the Early Treatment of Diabetic Retinopathy Study [ETDRS] scale) and for sustained onset of nephropathy (microalbuminuria). For each factor (eTIR or HbA1c), the corresponding HR and its 95% confidence limits per the designated change in the factor (C), where C = 10 for eTIR and 0.5 for HbA1c, were obtained as exp(Cβ), exp(CβL), and exp(CβU), with βL and βU being the 95% confidence limits on the coefficient estimate. For each factor, models were fit with no adjustment for other covariates, adjusted for the other factor alone, and adjusted for the other factor plus other covariates. The β-coefficient for a factor is obtained as ln(HR)/ln(C). P values are not corrected for multiple tests.

*

HR computed per 10% lower eTIR from discrete-time Cox proportional hazards regression models as in Beck et al. (3).

+

HR per 0.5% higher HbA1c as in Beck et al. (3).

#

Cox proportional hazards regression models stratified by the ETDRS level of retinopathy at baseline (no retinopathy, 10, 20, or 30 in worse eye) and adjusted for the preexisting duration of diabetes, retinopathy cohort, and duration by cohort interaction, as performed previously (3).

Close Modal

or Create an Account

Close Modal
Close Modal