. Author manuscript; available in PMC: 2013 Dec 17.

Published in final edited form as: J Am Coll Cardiol. 2010 Mar 30;55(13):10.1016/j.jacc.2009.10.056. doi: 10.1016/j.jacc.2009.10.056

Table 2.

Adjusted summary hazard ratios (and 95% confidence intervals) for cardiovascular mortality among NHIS participants in 1988, 1990, 1991, and 1997–2000, according to baseline volume of alcohol consumption.

	Volume of Alcohol Consumption
	Never Drinker	Lifelong Infrequent Drinker	Former Drinker	Light	Moderate	Heavy
CVD Deaths¹	2754	1538	1700	1131	697	354
Person-Years	305966	242849	174714	520562	238498	94815
Age-Sex-Race Adj.	1.00	0.98 (0.91–1.05)	1.15 (1.06–1.24)	0.66 (0.56–0.77)	0.58 (0.47–0.72)	1.07 (0.94–1.22)
Multivariable Adj.²	1.00	0.95 (0.88–1.02)	1.02 (0.94–1.11)	0.69 (0.59–0.82)	0.62 (0.50–0.77)	0.95 (0.82–1.10)
MV+Activity³	1.00	0.96 (0.89–1.03)	1.02 (0.94–1.11)	0.71 (0.61–0.83)	0.65 (0.53–0.79)	0.96 (0.83–1.11)
Pooled MV HR⁴	1.00	0.96 (0.89–1.03)	1.02 (0.94–1.10)	0.71 (0.65–0.78)	0.70 (0.63–0.78)	0.95 (0.82–1.10)
Two-Year Lag⁵	1.00	0.96 (0.89–1.04)	1.03 (0.94–1.13)	0.72 (0.62–0.84)	0.74 (0.63–0.86)	1.04 (0.90–1.21)

Hazard ratios derive from random-effects meta-analyses of NHIS years 1988, 1990, 1991, and 1997–2000; 1987 and 1992 did not include information with which to identify former drinkers.

Multivariable-adjusted hazard ratios adjusted for age, sex, race, smoking, marital status, education, region, urbanization, smoking, BMI, and general health status.

Analyses adjusted for covariates in multivariable models and physical activity assessed in 1990, 1991, and 1997–2000.

⁴

Hazard ratios derive from pooled (rather than meta-analyzed) analyses of individual NHIS administrations.

⁵

Lagged analyses exclude individuals who died within the first two years after administration of the respective NHIS.