JACN Did you know that you can get alerts when a new issue is online?
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Meyer, K. A.
Right arrow Articles by Rimm, E. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Meyer, K. A.
Right arrow Articles by Rimm, E. B.
Journal of the American College of Nutrition, Vol. 22, No. 3, 185-194 (2003)
Published by the American College of Nutrition

Original Research

Alcohol Consumption Patterns and HbA1c, C-Peptide and Insulin Concentrations in Men

Katie A. Meyer, MPH, Katherine M. Conigrave, MD, Nain-Feng Chu, MD, Nader Rifai, PhD, Donna Spiegelman, ScD, Meir J. Stampfer, MD and Eric B. Rimm, ScD

Department of Epidemiology (K.A.M., D.S., M.J.S., E.B.R.), Boston, Massachusetts
Department of Biostatistics (D.S.), Boston, Massachusetts
Department of Nutrition (M.J.S., E.B.R.), Boston, Massachusetts
Harvard School of Public Health, Department of Pathology (N.R.), Boston, Massachusetts
Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital (M.J.S., E.B.R.), Boston, Massachusetts
Harvard Medical School, Department of Laboratory Medicine, Children’s Hospital (N.R.), Boston, Massachusetts
Drug and Alcohol Department, Royal Prince Alfred Hospital and Departments of Medicine, Psychological Medicine and Public Health and Community Medicine, University of Sydney, AUSTRALIA (K.M.C.)
Department of Public Health, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, REPUBLIC OF CHINA (N.-F.C.)

Address correspondence to: Eric B. Rimm, ScD, Department of Nutrition, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115. E-mail: eric.rimm{at}channing.harvard.edu


   ABSTRACT
TOP
 FOOTNOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 REFERENCES
 
Objective: Observational studies support inverse associations between moderate alcohol consumption and fasting insulin concentrations, but the importance of drinking pattern on the effect of alcohol on insulin sensitivity has not been fully explored. We examined the relations of alcohol consumption patterns—including average daily consumption, frequency of consumption and drinking with meals—to fasting insulin, fasting c-peptide and hemoglobin A1c (HbA1c).

Methods: A cross-sectional study of 462 disease-free men selected from the Health Professionals’ Follow-up Study to provide information on a range of drinking patterns. Study participants were 48 to 82 years of age who provided a blood sample and detailed information on diet, life-style and alcohol consumption patterns in 1994. Among the study participants, 267 men provided a fasting blood sample and contributed to the analyses of insulin and c-peptide.

Results: Biologic markers were not strongly related to average alcohol consumption. Compared to abstainers, differences in insulin concentrations—all statistically non-significant—were 0.06, 1.25, 1.02, and 0.12 µU/mL for consumers of <1, 1–1.9, 2–2.9, 3+ drinks per day, respectively. The frequency of alcohol consumption was inversely related to fasting c-peptide and insulin concentrations after controlling for average alcohol consumption and other potential confounding variables. Compared to men who reported consuming alcohol one to three days per week, c-peptide concentrations were 0.08 ng/mL and 0.29 ng/mL lower (p-trend = 0.04) in men who reported consuming alcohol on four to five days per week and six to seven days per week, respectively. Men who consumed alcohol on most days also had lower fasting insulin levels than more irregular drinkers (p-trend = 0.05).

Conclusions: Our results suggest that frequent alcohol consumption is inversely related to fasting c-peptide and insulin concentrations.

Key words: insulin, c-peptide, HbA1c, alcohol, diabetes, men


   INTRODUCTION
TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 REFERENCES
 
Results from several large prospective studies suggest a reduced risk of type 2 diabetes among moderate drinkers [1–4]. Similarly, inverse associations between moderate alcohol consumption and plasma glucose, insulin levels and insulin resistance have been reported in cross-sectional data [5–11]. However, findings from both cross-sectional and prospective cohort studies of biomarkers and insulin sensitivity have been inconsistent. Some find either no association [11–15] or a positive association [4,13,16,17] between alcohol consumption and glycemic control.

Previous epidemiologic studies may be limited because few biological parameters of glycemic status have been assessed and only average daily alcohol consumption was examined. To our knowledge, no study has yet reported on the relation of usual pattern of drinking, including the frequency of consumption and drinking with meals, on insulin sensitivity. More detailed measurements of glycemic control and alcohol consumption may contribute to a fuller understanding of the potential effects of alcohol.

In the current study, we examine the cross-sectional associations between drinking patterns and three markers of glycemic control (fasting insulin, fasting c-peptide and HbA1c levels) in 462 men from the Health Professionals’ Follow-up Study (HPFS). Average alcohol consumption was assessed with a semi-quantitative food frequency questionnaire. Unique to our study are data that further characterize typical drinking patterns with respect to drinking with meals and the regularity of consumption, which allowed us to examine the independent and joint effects of these three aspects of drinking behavior.


   MATERIALS AND METHODS
TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 REFERENCES
 
Subjects
The Health Professionals’ Follow-up Study (HPFS) is a prospective cohort study of 51,529 US male health professionals, 40 to 75 years of age at baseline. Cohort members completed a detailed baseline questionnaire in 1986, which assessed average dietary intake, other health behaviors and newly diagnosed diseases. Biennial follow-up questionnaires have been mailed to study participants with an average follow-up rate of 94% of the total possible person-years through 1994. In 1993–94, all study participants were invited to provide a blood sample, and 18,225 participants returned a sample in a mailed blood kit. Fig. 1 outlines this and subsequent steps to the selection of the study sample from the entire HPFS cohort.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 1. Selection of study population from the larger Health Professionals’ Follow-up Study cohort.

 
We initiated a study of the effects of alcohol consumption patterns on biologic markers among a sample of eligible HPFS cohort members. Biologic markers of glycemic control included fasting insulin, fasting c-peptide—an indicator of insulin secretion—and hemoglobin A1c (HbA1c)—a time integrated measure of glucose control. Other markers of glycemic control were unavailable. Eligible participants were those who provided a blood sample in 1993–1994, returned all follow-up questionnaires between 1986 and 1994, had complete data on drinking variables over the course of this follow-up and, prior to 1994, were free of self-reported illnesses that may have influenced either alcohol consumption or glycemic status, including myocardial infarction, angina pectoris, stroke, diabetes mellitus, intermittent claudication, gastric or duodenal ulcers, gall bladder removal, liver disease and all cancers except non-melanoma skin cancer. Of the 18,225 cohort members who provided blood samples in 1993–1994, 8,922 were excluded because they did not have complete questionnaire information for alcohol consumption from 1986 through 1994. Although we used data only from 1994 in the analysis, we wanted to select subjects based on their long-standing pattern of consumption. Thus, we used surveys from 1986–1994 only in the selection process to ensure that the alcohol consumption patterns that were reported in 1994 and used in the analysis truly reflected subjects’ long-term patterns. From the remaining 9,303 participants, we additionally excluded 208 men who had at least one of the self-reported medical conditions. We classified the remaining 9,095 participants according to their alcohol consumption patterns and then randomly sampled 468 men from within these categories of alcohol consumption patterns to ensure adequate numbers for statistical analysis. We sampled study members from several different drinking pattern profiles classified by their average consumption, the percentage of alcohol consumed with meals, and the average number of days per week on which alcohol was consumed. We excluded four of the 468 study participants from all statistical analysis because their responses to the alcohol consumption portion of the follow-up questionnaires were inconsistent.

On average, the 18,225 men who provided blood samples were similar in age, BMI and smoking status to members of the cohort as a whole. Medians of age and BMI for the HPFS cohort and the self-selected sample of blood providers were, respectively, 62 and 61 years for age, and 25.5 kg/m2 for both groups for BMI. Of the entire cohort 6% were current smokers, compared with 5% of those who provided blood. In turn, the 468 study subjects were similar to the 18,225 cohort members who provided blood with respect to medians for age (61 years in both groups), BMI (25.0 vs. 25.5 kg/m2) and smoking status (6% vs. 5% current smoking).

Data Collection
We queried drinking patterns in questionnaires and defined them in terms of average daily alcohol consumption, the proportion of total alcohol consumed with meals and the number of drinks consumed per drinking day. Total alcohol consumption was assessed in 1994 as part of a 131-item semi-quantitative food frequency questionnaire. Average consumption of light and regular beer, white and red wine and liquor was reported as one of nine response categories ranging from less than one serving per month up to 6+ servings per day. Standard portion sizes were specified as one glass, bottle or 12-ounce can for beer (355 mL), one 4-ounce glass for wine (118 mL), and one shot of liquor (45 mL). We calculated total alcohol intake for each participant as the sum of the reported average intakes of each alcoholic beverage. In addition, we assigned gram amounts of alcohol to each beverage: 12.8 grams of alcohol for beer, 11.3 grams for light beer, 11.0 grams for red and white wine and 14.0 grams for spirits. In the 1994 follow-up questionnaire, we asked cohort members to report the proportion of total alcohol that they typically consumed with meals; categories were <25%, 25%–49%, 50%–74%, and 75% or more of total alcohol consumed with meals. In 1986 and 1988, cohort members had also reported the frequency with which they typically consumed alcohol per week, ranging from 0–7 times per week.

The validity and reproducibility of the food frequency questionnaire were assessed in a random sample of 136 participants who had returned the 1986 baseline questionnaire and a second questionnaire in 1987 [18]. Alcohol use as measured by the first and second questionnaires was compared to that measured by two one-week diet records administered six months apart between the questionnaires. The estimated average daily alcohol intakes from the 1986 (mean = 12.0 g) and 1987 (mean = 12.5 g) food frequency questionnaires were very similar to each other and to the mean daily intake from diet records (mean = 12.8 g). The Spearman correlation coefficient for alcohol consumption calculated from the 1987 questionnaire and the diet record was 0.86, and the correlation coefficient for the first and second questionnaires was 0.92. Furthermore, the correlation between HDL-C and the 1987 questionnaire assessment of average alcohol was 0.35, a value similar to what would be expected based on metabolic studies of alcohol consumption and lipids [19]. Finally, we have evidence that self-reported alcohol was inversely associated with CHD and type 2 diabetes and positively related to colon cancer and fractures, which is similar to results found in other studies where methods of reporting have varied from self-report to interview administered.

From April 1993 through August 1995, 18,225 HPFS cohort members provided blood samples for the study of biologic markers. Blood samples were collected in liquid EDTA blood tubes and returned to the laboratory, via overnight courier, on ice packs stored in Styrofoam containers. Over 95% of the samples arrived within 24 hours of being drawn. We centrifuged, aliquoted and stored the samples in liquid nitrogen (-150°C) until analysis. Based on visual inspection, less than 15% of the samples were slightly hemolyzed and very few were moderately hemolyzed (<3%), lipemic (<1%) or not cooled upon arrival (<0.5%). We measured red blood cell HbA1c by a temperature controlled HPLC method; against standards of 5.6% and 9.6%, the coefficients of variation for HbA1c were less than 2.5% (Boehringer Mannheim). Plasma insulin and c-peptide were measured with an RIA method, which has little or no proinsulin or c-peptide cross-reactivity and has a reported average CV of less than 10% (Linco Research, St. Charles, MO). Red cell levels of HbA1c were available from laboratory analysis for 466 of the 468 study subjects; plasma levels of insulin and c-peptide were available for all 468 study subjects.

This study was primarily designed to examine drinking patterns in relation to lipids and clotting factors; supplemental support to additionally measure insulin, c-peptide and HbA1c was received after blood samples had been obtained. A questionnaire was included in the blood kit that queried the date and time of the blood draw, as well as the time since last meal. A substantial number of participants (N = 199) reported having consumed a meal within six hours of the blood draw and were therefore not eligible for inclusion in the insulin and c-peptide analyses. We excluded these 199 men, and 36 who did not respond to the question, from the analyses of insulin and c-peptide. In addition, we excluded subjects with insulin levels that were implausibly high for a non-diabetic population (greater than 40 µU/mL) and all subjects whose blood was partially hemolyzed, based on visual inspection, from the insulin analyses (n = 22). After exclusions, there were 462 observations available for the analysis of HbA1c, 267 for c-peptide and 245 for insulin.

We assessed data on covariates, when possible, from the 1994 follow-up questionnaire and assigned missing covariate values the most recent, available covariate value. Other covariate information assessed in 1994 included age, self-reported diagnosis of hypertension or anti-hypertensive medication use and dietary intakes of fiber, saturated fat and polyunsaturated fat.

Statistical Analysis
We examined the population distribution of alcohol consumption and created categorical variables for each of the three alcohol drinking variables—average daily consumption, frequency of consumption and the degree to which alcohol was consumed with meals—to ensure adequate numbers of subjects in each category for analysis. We used linear regression models to assess the difference in each continuous biologic marker associated with a one level increase in each categorical drinking pattern variable. We used the robust variance to allow for valid statistical inference from linear regression models despite non-normality of continuous outcome variables [20]. All regression coefficients are presented with 95% confidence intervals; reported p-values were two-tailed.

We adjusted for potential confounding variables—age, BMI, physical activity, hypertension, smoking, dietary fiber, saturated fat and polyunsaturated fat—in multivariate regression analyses. We also included average alcohol consumption and number of drinks consumed per drinking day in the analyses of alcohol consumed with meals and average alcohol consumption and drinking with meals in the analysis of number of drinks consumed per drinking day in an attempt to isolate the effects of each drinking variable. We considered possible effect modification by age and BMI. There were insufficient numbers of current smokers to study effect modification by smoking status in multivariate analysis. Finally, we examined the specific effects of wine, beer and spirits by simultaneously including three-level indicator variables (0 grams/day, 0 < grams/day < 15, and 15+ grams/day) for each beverage type.


   RESULTS
TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
REFERENCES
 
A description of the study population across categories of average daily alcohol consumption is shown in Table 1. The number of drinks consumed per drinking day increased with average daily alcohol consumption; this did not vary by beverage type. Study participants with higher levels of daily alcohol consumption were more likely to be hypertensive, smoke and report lower levels of dietary fiber intake. BMI and average physical activity did not appear strongly related to total consumption.


View this table:
[in this window]
[in a new window]
  		Table 1. Selected Baseline (Mean) Diet and Lifestyle Characteristics across Categories of Average Daily Consumption of Alcohol among 464 Study Subjects1

 
We examined the associations between HbA1c, fasting insulin and fasting c-peptide and average daily alcohol consumption (Table 2). There was a modest, not statistically significant, inverse relation between HbA1c and average consumption and a small, statistically non-significant, positive association between fasting c-peptide and average consumption. The relation between insulin and average alcohol consumption was inconsistent. Multivariate linear regression estimates—all statistically non-significant—were 0.06, 1.25, 1.02 and 0.12 µU/mL different between each category of average alcohol consumption, compared to abstainers. Point estimates from analyses that excluded abstainers differed minimally from the analyses that included abstainers. The relation between average consumption and biomarker levels did not vary according to beverage type—wine, beer or spirits.


View this table:
[in this window]
[in a new window]
  		Table 2. Absolute Differences in Biomarker Levels Associated with Grouped Differences in Average Number of Servings of Alcohol Consumed per Day, Adjusting for Potential Confounding Factors1

 
We considered the possibility that, in addition to average daily consumption, the frequency of alcohol consumption may influence glycemic control. The distribution of selected variables is shown in Table 3 across levels of frequency of alcohol consumption, in number of drinking days per week. Because we oversampled extreme drinking patterns, the average number of drinks consumed per drinking day was inversely related to frequency of consumption; average consumption was 6.5 per drinking day for men who reported drinking on one to three days per week and 2.5 drinks for those drinking six to seven days per week. Frequency of alcohol consumption did not appear related to dietary variables, physical activity, smoking or BMI. Interestingly, the prevalence of hypertension was inversely related to the number of drinking days, despite the positive relation between number of drinking days and average daily consumption.


View this table:
[in this window]
[in a new window]
  		Table 3. Selected Baseline (Mean) Diet and Lifestyle Characteristics across Categories of Frequency of Alcohol Consumption among 464 Study Subjects1

 
Age- and BMI-adjusted, and multivariate-adjusted linear regression comparisons of biomarker concentrations by categories of frequency of alcohol consumption are presented in Table 4. The frequency of alcohol consumption was unrelated to HbA1c levels. Multivariate differences in c-peptide concentrations were -0.08 ng/mL (95% CI: -0.41, 0.25) and -0.29 ng/mL (95% CI: -0.56, -0.02) (p-trend = 0.04) lower in men who consumed alcohol on four to five days per week and six to seven days per week, respectively, compared to men who consumed alcohol on one to three days per week. For insulin, without adjusting for average alcohol consumption, there was a -1.55 µU/mL (95% CI: -3.09, -0.01) difference between men who consumed alcohol six to seven days per week to those who consumed alcohol one to three days per week. Adjusting for average daily consumption strengthened the relation between fasting insulin and frequency of consumption (Fig. 2). Mean insulin concentrations were 13.0, 12.1 and 11.3 µU/mL greater (p for trend = 0.03) for men consuming alcohol one to three days per week, four to five days per week and six to seven days per week, respectively, compared to abstainers.


View this table:
[in this window]
[in a new window]
  		Table 4. Absolute Differences in Biomarker Levels Associated with Grouped Differences in the Frequency of Alcohol Consumption, Adjusting for Potential Confounding Factors1

 


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 2. Mean values of insulin and c-peptide across categories of frequency of consumption, adjusted for average daily consumption and other confounding factors listed in Table 4.

 
The reported degree to which alcohol was typically consumed with meals was not strongly associated with levels of HbA1c, c-peptide or insulin, but the data do suggest a more beneficial glycemic profile among men who consumed most of their alcohol with meals (Table 5). Men who reported consuming <25% of their total alcohol consumption with meals had a statistically non-significant 0.12% higher HbA1c concentration than those who consumed 75% or more of their alcohol with meals. Absolute insulin concentrations were 1.82 and 1.50 µU/mL higher among men who reported consuming 25% to 75% and <25% of total alcohol with meals, compared to men that reported consuming >75% of total alcohol with meals. Adjusting for the other drinking pattern variables did not substantively alter the findings for the relations between biomarker levels and frequency of consumption or percent of alcohol consumed with meals.


View this table:
[in this window]
[in a new window]
  		Table 5. Absolute Differences in Biomarker Levels Associated with Grouped Differences in the Percent of Alcohol Consumed with Meals, Adjusting for Potential Confounding Factors1

 
We further examined the relation between alcohol consumption patterns and biomarkers of glycemic control through simultaneous adjustment of drinking pattern variables and by exploring possible effect modification. Simultaneous adjustment for frequency of consumption, drinking with meals and average daily consumption changed regression estimates for frequency of consumption and drinking with meals only modestly. We considered the possibility that the beneficial effects of alcohol consumption may have been restricted to men who reported regular, rather than episodic, alcohol consumption. However, there was not evidence for an interaction between frequency of consumption and average daily consumption on the biologic measures.


   DISCUSSION
 
Although average consumption of alcohol was not associated with lower concentrations of HbA1c, fasting insulin or fasting c-peptide, men with frequent alcohol consumption (6–7 days per week) had significantly lower fasting c-peptide and insulin levels than less frequent drinkers (1–3 days per week). Frequent consumers drank on average 2.5 drinks per drinking day, while infrequent consumers reported an average of 6.5 drinks per drinking day. Men who consumed alcohol with meals also tended to have a better glycemic profile. These findings did not vary according to the type of alcohol consumed.

Our finding that average alcohol consumption was not inversely related to insulin runs counter to several cross-sectional studies [5–11] and prospective cohort studies which support an inverse relation between average alcohol consumption and incident type 2 diabetes drinkers [1–4,21,22]. In 1995, we reported that HPFS cohort members who consumed 30.0–49.9 g of alcohol daily had a relative risk of diabetes of 0.61 (95% confidence interval 0.44 to 0.91) compared to abstainers over six years of follow-up [1].

In the present study, the lack of an inverse relation between alcohol intake and insulin concentrations is unlikely to be attributable to either confounding bias or measurement error. We controlled for variables believed to affect insulin levels, including BMI, physical activity and smoking. A validation study of the alcohol questions on the FFQ demonstrated similar estimated mean intakes from the FFQ and two one-week diet records [18]. Although we did not conduct a formal validation study of the insulin measure, the direction and magnitude of the associations between insulin and other covariates—including physical activity [23] and obesity [24]—were consistent with expectation in our study population.

The oversampling of episodic and heavy drinkers from the HPFS cohort also would not explain the differences between our findings and those of other research groups. Our analyses suggested that the study sample was comparable to the cohort population with respect to several correlates of insulin sensitivity and alcohol consumption. Further, we adjusted for potential confounding variables, including characteristics of alcohol consumption, in multivariate regression models. Rather than impair the validity of this study, our sampling scheme would optimize the statistical power for examining a range of drinking patterns in this population and to control for confounding of average consumption with frequency.

Although we expected an inverse relation between insulin and average alcohol consumption, the literature has not consistently supported an inverse association. Several previous reports suggested a U-shaped association between alcohol and insulin resistance or diabetes [25,26], and still other studies have shown a positive association between alcohol consumption and insulin or glucose concentrations [13,17]. The importance of the pattern of drinking, rarely reported, may explain some of the apparent conflict in past literature on the effect of alcohol on glycemic control and on subsequent risk of diabetes. For example, the relation between average consumption and glycemic control may be confounded by other characteristics of alcohol consumption, such as frequency of consumption.

We found an inverse association between frequency of alcohol consumption and levels of c-peptide and insulin. Compared to abstainers, the mean insulin concentration was 1.62 µU/mL lower in those who consumed alcohol five to seven days a week, which would be clinically significant on a population level [27]. These results are consistent with findings from a parallel analysis in the HPFS cohort. After 12 years of follow-up, we documented 1,571 cases of type 2 diabetes mellitus among 46,892 men free of chronic disease at baseline. Men who reported alcohol consumption on 5+ days per week had a significantly lower incidence of diabetes (RR = 0.60; 95% CI: 0.43–0.84) compared to men who drank one to two days per week, controlling for the number of drinks consumed per day. In this analysis, average daily alcohol consumption was not strongly related to diabetes incidence once it was adjusted for frequency of alcohol consumption. That is, within strata of frequency of alcohol consumption, the risk of diabetes was relatively constant over categories of drinks consumed per drinking day [28].

There are several potential mechanisms through which alcohol may affect the insulin carbohydrate system. Alcohol acts as a preferred fuel, being metabolized in preference to other calorie sources in the body, thus reducing glucose disposal [29,30]. However, alcohol also inhibits gluconeogenesis [31–33] and thus when consumed in moderation may result in better maintenance of normoglycemia. In laboratory studies, the effect of alcohol on the insulin carbohydrate system may vary both with the amount of alcohol consumed and also from the fasting to the fed state. In fasting subjects, high doses of alcohol can lead to hypoglycemia, possibly due to reduced gluconeogenesis [34,35]. But in the non-fasting state, high doses of alcohol may reduce glucose disposal causing resistance to the effects of insulin [29,36]. Higher levels of alcohol may reduce insulin binding and inhibit intracellular signaling related to insulin [37,38], but moderate levels appear to be related to enhanced sensitivity [6,7]. Our findings that frequent consumption of alcohol may be more beneficial than more infrequent consumption has not been explored in metabolic studies. Jequier hypothesized that alcohol may induce increases in energy expenditure [39,40], which, if spread over most days, could be more influential on long term glucose utilization and insulin sensitivity than infrequent consumption.

Like many other studies of this association [5–11], our study was cross-sectional, and determinations of causality may not be warranted. However, findings from several prospective cohort studies have supported an inverse relation between alcohol and measures of glycemic control [1–4,21,22]. Further, recent findings from a randomized metabolic trial showed that daily moderate alcohol consumption significantly reduced fasting insulin and increased insulin sensitivity [41]. Together, these results are strongly suggestive of cause and effect. Our study sample was small, and one possibility is that we simply lacked sufficient power to detect meaningful associations with average alcohol consumption.

Our findings demonstrate the potential importance of frequency of alcohol consumption on measures of glycemic control, specifically c-peptide and insulin, and warrant assessing not only average daily consumption, but also drinking patterns in examining the effects of alcohol on biological markers and disease outcomes.


   FOOTNOTES
TOP
 FOOTNOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 REFERENCES
 
Supported by research grants HL35464, CA55075, and AA11181 from the National Institutes of Health.

Received March 20, 2002. Accepted October 7, 2002.


   REFERENCES
TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 REFERENCES
 

  1. Rimm EB, Chan J, Stampfer MJ, Colditz GA, Willett WC: Prospective study of cigarette smoking, alcohol use, and the risk of diabetes in men. BMJ310 :555 –559,1995 .[Abstract/Free Full Text]
  2. Stampfer MJ, Colditz GA, Willett WC, Manson JE, Arky RA, Hennekens CH, Speizer FE: A prospective study of moderate alcohol drinking and risk of diabetes in women. Am J Epidemiol128 :549 –558,1988 .[Abstract/Free Full Text]
  3. Perry E, Wannamethee SG, Walker MK, Thomson AG, Whincup PH, Shaper AG: Prospective study of risk factors for development of non-insulin dependent diabetes in middle-aged British men. BMJ310 :560 –564,1995 .[Abstract/Free Full Text]
  4. Tsumura K, Kayashi T, Suematsu C, Endo G, Fujii S, Okada K: Daily alcohol consumption and the risk of type 2 diabetes in Japanese men: the Osaka Health Survey. Diabetes Care22 :1432 –1437,1999 .[Abstract/Free Full Text]
  5. Mayer EJ, Newman B, Quesenberry Jr CP, Friedman GD, Selby JV: Alcohol consumption and insulin concentrations. Role of insulin in associations of alcohol intake with high-density lipoprotein cholesterol and triglycerides. Circulation88 :2190 –2197,1993 .[Abstract/Free Full Text]
  6. Facchini F, Chen Y-D, Reaven GM: Light-to-moderate alcohol intake is associated with enhanced insulin sensitivity. Diabetes Care17 :115 –119,1994 .[Abstract]
  7. Kiechl S, Willeit J, Poewe W, Egger G, Oberhollenzer F, Muggeo M, Bonora E: Insulin sensitivity and regular alcohol consumption: large, prospective, cross sectional population study (Bruneck study). BMJ313 :1040 –1044,1996 .[Abstract/Free Full Text]
  8. Mooy JM, Grootenhuis PA, de Vries H, Valdenburg HA, Bouter LM, Kostense PJ, Heine RJ: Prevalence and determinants of glucose intolerance in a Dutch caucasian population. Diabetes Care18 :1270 –1273,1995 .[Abstract]
  9. Razay G, Heaton KW, Bolton CH, Hughes AO: Alcohol consumption and its relation to cardiovascular risk factors in British women. BMJ304 :80 –83,1992 .
  10. Vitelli LL, Folsom AR, Shahar E, Winkhart SP, Shimakawa T, Stevens J, Duncan BB, Chambless LE, for the Atherosclerosis Risk in Communities (ARIC) Study Investigators: Association of dietary composition with fasting serum insulin level: the ARIC study. Nutr Metab Cardiovasc Dis6 :194 –202,1996 .
  11. Feskens EJ, Loeber JG, Kromhout D: Diet and physical activity as determinants of hyperinsulinemia: the Zutphen Elderly Study. Am J Epidemiol140 :350 –360,1994 .[Abstract/Free Full Text]
  12. Feskens EJ, Kromhout D: Cardiovascular risk factors and the 25-year incidence of diabetes mellitus in middle-aged men. The Zutphen Study. Am J Epidemiol130 :1101 –1108,1989 .[Abstract/Free Full Text]
  13. Hodge AM, Dowse GK, Collins VR, Zimmet PZ: Abnormal glucose tolerance and alcohol consumption in three populations at high risk of non-insulin-dependent diabetes mellitus. Am J Epidemiol137 :178 –189,1993 .[Abstract/Free Full Text]
  14. Ohlson L-O, Larsson B, Bjorntorp P, Eriksson H, Svardsudd K, Welin L, Tibblin G, Wilhelmsen L: Risk factors for Type 2 (non-insulin-dependent) diabetes mellitus. Thirteen and one-half years of follow-up of the participants in a study of Swedish men born in 1913. Diabetologia31 :798 –805,1988 .[Medline]
  15. Moller LF, Jespersen J: Elevated insulin levels in men: an 11-year follow-up study. J Cardiovasc Risk4 :339 –343,1995 .
  16. Holbrook TL, Barrett-Connor E, Wingard DL: A prospective population-based study of alcohol use and non-insulin-dependent diabetes mellitus. Am J Epidemiol132 :902 –909,1990 .[Abstract/Free Full Text]
  17. Godsland IF, Leyva F, Walton C, Worthington M, Stevenson JC: Associations of smoking, alcohol and physical activity with risk factors for coronary heart disease and diabetes in the first follow-up cohort of the Heart Disease and Diabetes Risk Indicators in a Screened Cohort study (HDDRISC-1). J Intern Med244 :33 –41,1998 .[Medline]
  18. Giovannucci E, Colditz G, Stampfer MJ, Rimm EB, Litin L, Sampson L, Willett WC: The assessment of alcohol consumption by a simple self-administered questionnaire. Am J Epidemiol133 :810 –817,1991 .[Abstract/Free Full Text]
  19. Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ: A biologic basis for moderate alcohol consumption and lower coronary heart disease risk: a meta-analysis of effects on lipids and hemostatic factors. British Medical Journal1999 .
  20. White H. Maximum likelihood estimation of misspecified models. Econometrica50 :1 –26,1982 .
  21. Ajani U, Hennekens C, Spelsberg A, Manson J: Alcohol consumption and risk of type 2 diabetes mellitus among US male physicians. Arch Intern Med160 :1025 –1030,2000 .[Abstract/Free Full Text]
  22. Wei M, Gibbons L, Mitchell T, Kampert J, Blair S: Alcohol intake and incidence of type 2 diabetes in men. Diabetes Care23 :18 –22,2000 .[Abstract]
  23. Fung TT, Hu FB, Yu J, Chu NF, Spiegelman D, Tofler GH, Willett WC, Rimm EB: Leisure-time physical activity, television watching, and plasma biomarkers of obesity and cardiovascular disease risk. Am J Epidemiol152 :1171 –1178,2000 .[Abstract/Free Full Text]
  24. Chu NF, Spiegelman D, Rifai N, Hotamisligil GS, Rimm EB: Glycemic status and soluble tumor necrosis factor receptor levels in relation to plasma leptin concentrations among normal weight and overweight US men. Int J Obes24 :1085 –1092,2000 .
  25. Lazarus S, Sparrow D, Weiss S: Alcohol intake and insulin levels. The Normative Aging Study. Am J Epidemiol145 :909 –916,1997 .[Abstract/Free Full Text]
  26. Razay G, Heaton KW: Alcohol consumption and cardiovascular risk factors in middle aged men. Cardiovasc Risk Factors5 :200 –205,1995 .
  27. Despres JP, Lamarche B, Mauriege P, Cantin B, Dagenais GR, Moorjani S, Lupien PJ: Hyperinsulinemia as an independent risk factor for ischemia heart disease. N Engl J Med334 :952 –957,1996 .[Abstract/Free Full Text]
  28. Conigrave KM, Hu FB, Carmargo C, Stampfer MJ, Willett WC, Rimm EB: A prospective study of drinking patterns in relation to risk of type 2 diabetes among men. Diabetes50 :2390 –2395,2001 .[Abstract/Free Full Text]
  29. Shah JH: Alcohol decreases insulin sensitivity in healthy subjects. Alcohol Alcohol23 :103 –109,1988 .[Abstract/Free Full Text]
  30. Shelmet JJ, Reichard GA, Skutches CL, Hoeldtke RD, Owen OE, Boden G: Ethanol causes acute inhibition of carbohydrate, fat, and protein oxidation and insulin resistance. J Clin Invest81 :1137 –1145,1988 .
  31. Avogaro A, Tiengo A: Alcohol, glucose metabolism and diabetes. Diabetes Metab Rev9 :129 –146,1993 .[Medline]
  32. Siler SQ, Neese RA, Christiansen MP, Hellerstein MK: The inhibition of gluconeogenesis following alcohol in humans. Am J Physiol275(5 Pt 1) :E897 –E907,1998 .
  33. Avogaro A, Valerio A, Miola M, Crepaldi C, Pavan P, Tiengo A, del Prato S: Ethanol impairs insulin-mediated glucose uptake by an indirect mechanism. J Clin Endocrinol Metab81 :2285 –2290,1996 .[Abstract]
  34. Arky RA: Hypoglycemia associated with liver disease and ethanol. Endocrinol Metab Clin North Am18 :75 –90,1989 .[Medline]
  35. Fauci AS, Braunwald E, Isselbacher KJ, Petersdorf RG, Wilson JD, Martin JB, Kasper DL, Hauser SL, Longo DL: "Harrison’s Principles of Internal Medicine," 14 ed. New York: McGraw Hill Inc, Health Professions Division,1998 .
  36. Yki-Jarvinen H, Nikkila EA: Ethanol decreases glucose utilization in healthy men. J Clin Endocrinol Metab61 :941 –945,1985 .[Abstract]
  37. Singh SP, Kumar Y, Snyder AK, Ellyin FE, Gilden JL: Effect of alcohol on glucose tolerance in normal and noninsulin-dependent diabetic subjects. Alcohol Clin Exp Res12 :727 –730,1988 .[Medline]
  38. de la Monte SM, Ganju N, Tanaka S, Banerjee K, Karl PJ, Brown NV: Differential effects of ethanol on insulin-signaling through the insulin receptor substrate-1. Alcohol Clin Exp Res23 :770 –777,1999 .[Medline]
  39. Jequier E: Alcohol intake and body weight: a paradox. Am J Clin Nutr69 :173 –174,1999 .[Free Full Text]
  40. Jequier E: Reply to MF McCarty. Am J Clin Nutr70 :941 –942,1999 .[Free Full Text]
  41. Davies MJ, Baer DJ, Jedd JT, Brown ED, Campbell WS, Taylor PR: Effects of moderate alcohol intake on fasting insulin and glucose concentrations and insulin sensitivity in postmenopausal women: a randomized controlled trial. JAMA287 :2559 –2562,2002 .[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Clin. Nutr.Home page
I. Baik and C. Shin
Prospective study of alcohol consumption and metabolic syndrome
Am. J. Clinical Nutrition, May 1, 2008; 87(5): 1455 - 1463.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
G. Tuncman, E. Erbay, X. Hom, I. De Vivo, H. Campos, E. B. Rimm, and G. S. Hotamisligil
A genetic variant at the fatty acid-binding protein aP2 locus reduces the risk for hypertriglyceridemia, type 2 diabetes, and cardiovascular disease
PNAS, May 2, 2006; 103(18): 6970 - 6975.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
E. K. Wei, J. Ma, M. N. Pollak, N. Rifai, C. S. Fuchs, S. E. Hankinson, and E. Giovannucci
C-Peptide, Insulin-like Growth Factor Binding Protein-1, Glycosylated Hemoglobin, and the Risk of Distal Colorectal Adenoma in Women.
Cancer Epidemiol. Biomarkers Prev., April 1, 2006; 15(4): 750 - 755.
[Abstract] [Full Text] [PDF]

Home page
 Diabetes CareHome page
T. Wu, W. C. Willett, S. E. Hankinson, and E. Giovannucci
Caffeinated Coffee, Decaffeinated Coffee, and Caffeine in Relation to Plasma C-Peptide Levels, a Marker of Insulin Secretion, in U.S. Women
Diabetes Care, June 1, 2005; 28(6): 1390 - 1396.
[Abstract] [Full Text] [PDF]

Home page
 JNCI J Natl Cancer InstHome page
E. K. Wei, E. Giovannucci, J. Selhub, C. S. Fuchs, S. E. Hankinson, and J. Ma
Plasma Vitamin B6 and the Risk of Colorectal Cancer and Adenoma in Women
J Natl Cancer Inst, May 4, 2005; 97(9): 684 - 692.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
T. Wu, E. Giovannucci, T. Pischon, S. E Hankinson, J. Ma, N. Rifai, and E. B Rimm
Fructose, glycemic load, and quantity and quality of carbohydrate in relation to plasma C-peptide concentrations in US women
Am. J. Clinical Nutrition, October 1, 2004; 80(4): 1043 - 1049.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Meyer, K. A.
Right arrow Articles by Rimm, E. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Meyer, K. A.
Right arrow Articles by Rimm, E. B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS