Heterogeneity of beta-cell function in subjects with multiple islet autoantibodies in the TEDDY family prevention study - TEFA

Table 3 The estimates and the 95% confidence intervals of the association between log2-transformed FPIR (mU/L) and autoantibody status, type or count, adjusted for age and sex

Covariates	Model 1 (n = 52)	Model 2 (n = 52)	Model 3 (n = 52)
Covariates	Est (95% CI) p-value	Est (95% CI) p-value	Est (95% CI) p-value
Autoantibody status (positive vs. negative)
IAA	−0.20 (− 0.89, 0.49) 0.564
GADA	0.32 (− 1.02, 1.65) 0.637
IA-2A	−0.40 (− 1.12, 0.31) 0.262
ZnT8(W/Q/R)A	−0.80 (−1.58, − 0.02) 0.046
Number of autoantibodies		−0.39 (− 0.83, 0.05) 0.084
Autoantibody combination group:
A vs. D			−0.45 (−1.39, 0.49) 0.345
B vs. D			−0.68 (−1.64, 0.27) 0.158
C vs. D			−0.17 (−1.18, 0.84) 0.736
Age (per 10 years)	0.15 (−0.16, 0.45) 0.343	0.09 (− 0.23, .040) 0.580	0.17 (− 0.16, 0.49) 0.307
Male vs. female	− 0.14 (− 0.83, 0.56) 0.695	−0.08 (− 0.82, 0.65) 0.818	−0.24 (− 0.98, 0.51) 0.525

The estimates and the 95% confidence intervals of the association between log2-transformed FPIR (mU/L) as the outcome and three measures of autoantibody status as predictors (status, count, combination group), adjusting for age and sex, estimated using linear models. The autoantibody information used as the main predictors was modeled as: the autoantibody status for IAA, GADA, IA-2A and any of ZnT8(W/Q/R) A, with negative status being the reference, the number of autoantibodies detected (possible values were 2, 3, or 4), and the autoantibody status combination group A-D (see Table 2) with group D as the reference

ISSN: 2055-8260