A Cross-Sectional Study of Alcohol Consumption Patterns and Biologic Markers of Glycemic Control Among 459 Women CANDYCE H. KROENKE, MPH, SCD 1,2 NAIN-FENG CHU, SCD 3 NADER RIFAI, MD 4 DONNA SPIEGELMAN, SCD 1,5 SUSAN E. HANKINSON, SCD 1,6 JOANN E. MANSON, MD, DRPH 1,2,6 ERIC B. RIMM, SCD 1,2,7 OBJECTIVE — Little research has explored associations of drinking patterns with glycemic control, especially among women. Our objective was to determine the relationship of patterns of alcohol consumption—including average daily consumption, weekly frequency of consump- tion, drinking with meals, and beverage type—with biologic markers of insulin resistance in young women. RESEARCH DESIGN AND METHODS — This study was cross-sectional in design. The subjects consisted of a stratified random subpopulation of 459 U.S. normal-weight and overweight female nurses, 33–50 years of age, drawn from the Nurses’ Health Study II and sampled for distinct drinking patterns. Women provided blood samples and detailed informa- tion on dietary and lifestyle factors between 1995 and 1999. The main outcome measures were fasting insulin, C-peptide, and HbA 1c . RESULTS — Adjusting for age, smoking, physical activity, television watching, BMI, and several dietary factors, average alcohol intake was inversely associated with HbA 1c (units in percentage of HbA 1c ): 0 g/day (reference = 5.36%), 0.1 to 5.0 g/day (-0.04%), 5.0 to 15.0 g/day (-0.09%), 15.0 to 25.0 g/day (-0.10%), and 25.0 g/day (-0.17%) (P value, test for trend 0.001). We found an inverse association of alcohol intake and insulin, but only for women with a BMI 25 kg/m 2 . Specifically, insulin levels were lowest for episodic drinkers consuming 2 drinks per day on 0 –3 days per week. Consumption with meals and type of alcoholic beverage did not further influence these results. CONCLUSIONS — Moderate alcohol consumption of 1–2 drinks per day on a few to several days of the week may have a beneficial glycemic effect, particularly among overweight women. Diabetes Care 26:1971–1978, 2003 M oderate alcohol consumption has been inversely associated with type 2 diabetes in several studies (1–5), with potentially direct and indirect effects on insulin secretion and insulin re- sistance (6 –10). In particular, drinking patterns characterized by frequent, mod- erate consumption may lower the risk of type 2 diabetes and related biologic mark- ers of glycemic control and insulin resis- tance (1,9,10), and, conversely, heavy episodic consumption may have opposite effects (11,12). Meyer et al. (13) found an inverse association between frequency of alcohol consumption and fasting C- peptide levels and a suggestion of an in- verse association with fasting insulin concentrations in a cross-sectional study of middle-aged and older men in the Health Professionals’ Follow-up Study. Davies et al. (9) found that moderate al- cohol consumption reduced fasting insu- lin concentration and improved insulin sensitivity in a randomized controlled trial of 63 postmenopausal women. In this study, we explored similar relation- ships in a cross-sectional study of 459 premenopausal women, ages 33–50 years, in the Nurses’ Health Study II (NHS2). RESEARCH DESIGN AND METHODS Subjects NHS2 is a large prospective cohort study of 116,671 U.S. female nurses, 25– 42 years of age at baseline in 1989. At base- line and during biennial follow-up peri- ods, participants provided detailed lifestyle and medical history information through a mailed questionnaire. All women in the study were also asked to provide a venous blood sample; 29,613 study participants did so between 1996 and 1998. For this analysis, we used blood sam- ples provided by women who were pre- menopausal, were not taking exogenous hormones, were neither pregnant nor ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● From the 1 Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts; the 2 Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts; the 3 Departments of Community Medicine and Public Health, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, Republic of China; the 4 Department of Labora- tory Medicine, Children’s Hospital, and the Department of Pathology, Harvard Medical School, Boston, Massachusetts; the 5 Department of Biostatistics, Harvard School of Public Health, Boston, Massachusetts; the 6 Division of Preventive Medicine, Department of Medicine, Brigham and Women’s Hospital, Harvard Med- ical School, Boston, Massachusetts; and the 7 Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts. Address correspondence and reprint requests to Candyce Kroenke, Channing Laboratory, 181 Longwood Ave., 3rd floor, Boston, MA 02115. E-mail: n2chk@channing.harvard.edu. Received for publication 18 February 2003 and accepted in revised form 9 April 2003. E.B.R. has received honoraria from the Distilled Spirits Council of the United States and the National Beer Wholesalers Association. Abbreviations: NHS2, Nurses’ Health Study II. A table elsewhere in this issue shows conventional and Syste `me International (SI) units and conversion factors for many substances. © 2003 by the American Diabetes Association. Clinical Care/Education/Nutrition O R I G I N A L A R T I C L E DIABETES CARE, VOLUME 26, NUMBER 7, JULY 2003 1971