Table 3 Result of the Cox proportional hazards model for the association between alcohol consumption and CKD development stratified by BMI.

Amount of alcohol consumption	CKD				eGFR decline to < 60 mL/min/1.73 m²				New-onset of proteinuria
	Multivariate model^a				Multivariate model^a				Multivariate model^a
	HR	95% CI	P trend^b	P interaction^c	HR	95% CI	P trend^b	P interaction^c	HR	95% CI	P trend^b	P interaction^c
BMI < 18.5 kg/m²				0.005				0.03				0.03
Infrequent	Ref		0.003		Ref		0.054		Ref		0.019
< 20 g/day	1.26	0.66–2.40			0.79	0.28–2.23			1.77	0.76–4.14
20–39 g/day	3.44	1.60–7.42			1.94	0.48–7.81			4.58	1.75–12.0
≥ 40 g/day	3.21	1.23–8.37			5.86	1.42–24.2			2.66	0.75–9.40
BMI 18.5–24.9 kg/m²
Infrequent	Ref		0.17		Ref		0.15		Ref		0.013
< 20 g/day	0.78	0.65–0.95			0.96	0.72–1.27			0.67	0.52–0.86
20–39 g/day	0.75	0.58–0.98			1.14	0.74–1.78			0.61	0.44–0.83
≥ 40 g/day	0.96	0.71–1.29			1.65	0.99–2.74			0.78	0.55–1.11
BMI ≥ 25 kg/m²
Infrequent	Ref		0.028		Ref		0.018		Ref		0.16
< 20 g/day	0.80	0.57–1.11			0.69	0.40–1.19			0.78	0.52–1.17
20–39 g/day	0.63	0.39–1.04			0.26	0.08–0.86			0.76	0.44–1.32
≥ 40 g/day	0.61	0.34–1.11			0.45	0.13–1.52			0.68	0.36–1.32

Outcomes were CKD (composite outcome of eGFR decline and/or new-onset of proteinuria) and eGFR decline and new-onset of proteinuria, respectively.
CKD chronic kidney disease, eGFR estimated glomerular filtration rate, BMI body mass index, CI confidence interval, HR hazard ratio.
^aMultivariable adjustment included age, sex, eGFR, hypertension, diabetes mellitus, hyper lipidemia, smoking status.
^bP trend was derived from Cox proportional hazards regression models by treating alcohol consumption as a continuous linear term.
^cP interaction was derived by using a likelihood ratio test from models with and without the cross-product term of alcohol category (Infrequent, light, moderate, and heavy drinkers) and risk factor in the multivariable-adjusted model.

Search