Polygenic association of glomerular filtration rate decline in world trade center responders

BMC Nephrology

Table 4 Polygenic Associations of Continuous and Ordinal eGFR Outcomes

	Renal Outcome
	eGFR Intercept		eGFR Slope		CKD Stage
	b (95% CI)	p ₁ p ₂	b (95% CI)	p ₁ p ₂	OR (95% CI)	p ₁ p ₂
Zero-Order (not adjusted for covariates)	-0.77 (-1.36, -0.19)	0.010 0.010	-0.02 (-0.06, 0.01)	0.229 0.491	1.07 (1.01, 1.14)	0.021 0.021
Adjusted for age, gender, education, and the first ten genetic principal components	-1.01 (-1.57, -0.45)	< 0.001 0.002	-0.03 (-0.07, 0.00)	0.077 0.306	1.18 (1.06, 1.32)	0.003 0.008
+ adjusted for BMI, diabetes, hypertension, and cardiovascular disease	-1.00 (-1.57, -0.43)	< 0.001 0.002	-0.03 (-0.06, 0.01)	0.163 0.491	1.19 (1.07, 1.34)	0.002 0.008
+ adjusted for exposure severity	-1.00 (-1.59, -0.41)	< 0.001 0.002	-0.02 (-0.06, 0.02)	0.317 0.491	1.18 (1.05, 1.32)	0.007 0.014

Notes. Unstandardized regression coefficients (b) and odds ratios (OR) are reported for interval and ordinal outcomes, respectively, with 95% confidence intervals (CI) shown in parentheses just below each value. Polygenic risk scores (PRSs) were transformed to standardized scores (M = 0, SD = 1) to facilitate the interpretation of estimated coefficients. For example, for every incremental increase in SD in the PRS for rapid eGFR decline, the odds of having a more advanced stage of CKD (i.e., Stage II or III versus Stage I) increases by ~ 18%, holding all other variables constant. P-values indicate the probability of having observed the estimated association if the null hypotheses (b = 0 or OR = 1) were true, before (p₁) and after (p₂) adjustment for family-wise error using the Holm method, whereby each column of point estimates constitutes a “family” (i.e., four null hypothesis significance tests for each renal outcome).

ISSN: 1471-2369