Coffee Consumption and Risk for Type 2 Diabetes Mellitus

By E. Salazar-Martínez, W. Willett, A. Ascherio, JoAnn E. Manson, M. Leitzmann, MR Stampfer, F. B. Hu. Annals of Internal Medicine, 2004

657 citations recorded by the source corpus.

Context A recently published study in a Dutch population identified an association between higher coffee consumption and lower risk for type 2 diabetes mellitus but did not distinguish between intake of regular and decaffeinated coffee. Contribution This large epidemiologic study in a U.S. sample confirmed the Dutch study findings. It also found a statistically significant protective association between total caffeine intake and type 2 diabetes mellitus and a modest inverse association with decaffeinated coffee consumption. Implications Consumption of caffeinated coffee is associated with a reduced risk for type 2 diabetes mellitus. Further studies should explore the long-term effects of caffeine on glucose metabolism. The Editors The prevalence of type 2 diabetes mellitus has increased dramatically in the past decades, and the disease now affects approximately 8% of U.S. adults (1). Diet and lifestyle factors are primary determinants of risk for type 2 diabetes (2). In addition to overweight and obesity, other independent behavioral and lifestyle factors include physical inactivity, smoking, and a diet higher in glycemic load and trans fatty acids and lower in fiber and polyunsaturated fatty acids. Coffee consumption may be related to diabetes because short-term metabolic studies have suggested that caffeine adversely affects insulin sensitivity and glucose metabolism (3). Coffee consumption is widespread; more than 50% of Americans drink coffee, and average per capita intake is about 2 cups per day (4). Coffee is the primary source of caffeine intake, providing approximately 210 mg/d per person in the United States. In humans, acute administration of caffeine decreases insulin sensitivity and impairs glucose tolerance (3, 5-8). On the other hand, caffeine stimulates thermogenesis and increases energy expenditure (9-11), which may facilitate weight reduction and maintenance. Because of these complex physiologic effects of caffeine and because tolerance to the humoral and hemodynamic effects of caffeine typically develops with long-term use (12), it is difficult to extrapolate findings from short-term metabolic studies to long-term use of coffee and other caffeinated beverages. A recent epidemiologic study has found a statistically significant inverse association between coffee consumption and risk for type 2 diabetes in a sample of Dutch participants (13). After adjustment for potential confounders, the relative risk for type 2 diabetes among participants consuming at least 7 cups of coffee per day as compared with those consuming 2 cups or less per day was 0.50 (95% CI, 0.35 to 0.72; P < 0.001 for trend). The study, however, could not distinguish regular coffee from decaffeinated coffee and did not evaluate the association with total caffeine intake. In our study, we examined long-term intake of coffee and other caffeinated beverages and decaffeinated coffee in relation to incidence of type 2 diabetes in 2 large prospective cohorts of men and women. In particular, we examined whether the associations were modified by smoking and body mass index. Methods Study Sample The Health Professionals Follow-up Study The Health Professionals Follow-up Study (HPFS) was established in 1986 when 51 529 male health professionals, including dentists, optometrists, veterinarians, osteopathic physicians, podiatrists, and pharmacists, who were 40 to 75 years of age returned a mailed questionnaire providing detailed information about their medical history, lifestyle, and other risk factors (14). Information on the cohort is updated every 2 years to identify newly diagnosed diseases. For our analysis, we excluded men with a previous diagnosis of type 2 diabetes, coronary heart disease, or cancer at baseline. Participants who did not complete more than 70 of 131 food items in the 1986 dietary questionnaire or had extreme scores for total daily intake of energy (<800 kcal or >4200 kcal) were excluded. After exclusions, the analysis included 41 934 eligible men who were followed from 1986 to 1998. The Nurses' Health Study The Nurses' Health Study (NHS) was established in 1976 when 121 700 female nurses 30 to 55 years of age living in 11 large states completed a mailed questionnaire providing detailed information about their medical history, lifestyles, and other risk factors (15). The information is updated every 2 years to identify newly diagnosed diseases. In 1980, a semiquantitative food frequency questionnaire was added. For our analysis, we excluded women with a previous diagnosis of type 2 diabetes, coronary heart disease, or cancer at baseline. Participants who did not complete more than 10 items on the 1980 dietary questionnaire or had extreme scores for total daily intake of energy (<500 kcal or >3500 kcal) were also excluded, leaving 84 276 eligible women who were followed from 1980 to 1998. Assessment of Coffee and Caffeine Intake Validated dietary questionnaires were sent to the HPFS participants in 1986, 1990, and 1994 and to the NHS participants in 1980, 1984, 1986, 1990, and 1994. In all the questionnaires, participants were asked how often on average during the previous year they had consumed coffee and tea. Decaffeinated coffee and different types of caffeinated soft drinks were first assessed in 1986 in the HPFS and in 1984 in the NHS. We assessed the total intake of caffeine by summing the caffeine content for a specific amount of each food during the previous year (1 cup for coffee or tea, one 12-ounce bottle or can for carbonated beverages, and 1 ounce for chocolate) multiplied by a weight proportional to the frequency of its use. The participants could choose from 9 frequency responses (never, 1 to 3 per month, 1 per week, 2 to 4 per week, 5 to 6 per week, 1 per day, 2 to 3 per day, 4 to 5 per day, and 6 or more per day). Using the U.S. Department of Agriculture food composition sources, we estimated that the caffeine content was 137 mg per cup of coffee, 47 mg per cup of tea, 46 mg per bottle or can of cola beverage, and 7 mg per serving of chocolate candy. In our validation studies, high correlations were recorded for coffee and other caffeinated beverage intake assessed by the food frequency questionnaire and several 1-week diet records (coffee, r = 0.78; tea, r = 0.93; and caffeinated sodas, r = 0.85) (16). For these analyses, coffee consumption was categorized into 5 groups: never, less than 1 cup per day, 1 to 3 cups per day, 4 to 5 cups per day, and 6 cups or more per day. Caffeine intake was categorized into quintiles. Assessment of Diabetes Cases Participants who reported having diabetes were sent a supplementary mailed questionnaire that asked about symptoms, diagnostic tests, and treatment. We considered any 1 of the following criteria to represent a new case of diabetes: classic symptoms and elevated glucose levels (fasting plasma glucose concentration 7.8 mmol/L [ 140 mg/dL] or randomly measured concentration 11.1 mmol/L [ 200 mg/dL]); at least 2 elevated plasma glucose concentrations on different occasions in the absence of symptoms (plasma glucose level 11.1 mmol/L [ 200 mg/dL] 2 or more hours after oral glucose tolerance testing); or treatment with oral hypoglycemic agents or insulin. Our criteria for diabetes classification are consistent with those proposed by the National Diabetes Data Group (17) because most cases were diagnosed before 1997. The validity of this diagnostic procedure has been verified in a subsample of this study sample (18). The diagnosis of type 2 diabetes was confirmed by medical records in 98% of the participants. In addition, another substudy assessing the prevalence of undiagnosed diabetes suggested a very low rate of false-negative results (19). Statistical Analysis We calculated the follow-up period from the return of the baseline questionnaires (in 1986 for men and in 1980 for women) to the diagnosis of type 2 diabetes, death, or the end of follow-up (January 1998 for men and June 1998 for women), whichever occurred first. Incidence rates for diabetes were calculated by dividing incident cases by person-years of follow-up in each category of coffee intake. Relative risks were calculated as the rate of occurrence of type 2 diabetes in each quintile of caffeine intake divided by the corresponding rate in the lowest quintile, and 95% CIs were calculated. To inspect potential confounding variables for the analyses, we examined baseline characteristics (directly standardized to the age distribution of each cohort) according to coffee consumption levels. We used proportional hazards models (20) to adjust for several risk factors, including age in 5 categories (<50, 50 to 54, 55 to 59, 60 to 64, or 65 years of age); body mass index (weight divided by height [kg/m2]) in 8 categories (<23, 23 to 24, 25 to 28, 29 to 30, 31 to 32, 33 to 34, 35 to 39, or 40 kg/m2); physical activity (hours of moderate or vigorous exercise in NHS or quintiles of metabolic equivalents in HPFS); family history of diabetes (yes or no); hormone use (never, current, or past [in NHS only]); smoking status (never, past, or current smokers [1 to 14, 15 to 24, or 25 cigarettes per day]); alcohol consumption (0, 0.1 to 4.9, 5 to 9.9, 10 to 14.9, or 15 g/d); total calorie intake (continuous); and quintiles of trans fat, glycemic load, and cereal fiber intake. We also adjusted for magnesium intake because our previous analyses have shown an inverse association between dietary magnesium and risk for diabetes (21). Information on covariates was updated periodically during follow-up. We performed a test for trend by treating median values of each category of caffeine and coffee intake as continuous variables. To represent long-term intake of coffee and caffeine and to reduce measurement error, we conducted analyses using cumulative updated caffeine and coffee intake from all questionnaires (22). For example, in the NHS, diabetes incidence between 1980 and 1984 was related to the coffee intake from the 1980 qu

Subfields: Pathology and Forensic Medicine, Pharmacology

View Paper: https://www.semanticscholar.org/paper/53eed9cf3c2f7e1ddaa500ee22603f047575e22f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Pathology and Forensic Medicine

  1. Augustin Scalbert
  2. Veronica Graham
  3. Claudine Manach
  4. F. B. Hu
  5. JoAnn E. Manson
  6. B. Frei
  7. W. Willett
  8. J. Higdon

In Pharmacology

  1. Augustin Scalbert
  2. Claudine Manach
  3. Veronica Graham
  4. David S. Wishart

Citations in the corpus