![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
,3
* Steno Diabetes Center, DK-2820 Gentofte, Denmark;
Department of Human Nutrition, LMC Centre for Advanced Food Studies, The Royal Veterinary & Agricultural University, DK-1958 Frederiksberg C, Denmark; and
** Research Centre for Prevention and Health, DK-2600 Glostrup, Denmark
3To whom correspondence should be addressed. E-mail: krif{at}steno.dk.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • diabetes • risk • diet • macronutrient • statistical approach
During the past decades, the prevalence of type 2 diabetes has increased worldwide (1,2), and this trend is expected to continue (3). Type 2 diabetes results from a complex interaction between genetics and pre- and postnatal environmental etiological factors resulting in the presence of several defects of glucose homeostasis. Modifiable lifestyle factors such as diet and physical inactivity play a major role in the development of diabetes (4,5). Several epidemiologic prospective studies found positive associations between diabetes and intake of total and saturated fat (6–8), and inverse associations between diabetes and dietary fiber intake (9–12), moderate alcohol consumption (13–15), and coffee consumption (16–18). In addition, associations with intake of total carbohydrate, protein, and tea were examined, but these dietary components were not related to diabetes risk in most studies (9–11,16,17,19).
Most studies analyzing diet-disease relations have focused on single dietary factors (e.g., carbohydrates) in the statistical analysis rather than combining different nutrients into the same model. When single macronutrients are analyzed, the interpretation of the results becomes difficult because the effect of a specific nutrient may depend on the other nutrient(s) it replaces (in models adjusted for total energy intake). Analyzing the effect of one specific nutrient in a statistical model, which takes into consideration the other nutrient the dietary factor under study is substituted with, may help clarify this complex relation. Furthermore, such a statistical model may provide information about macronutrient substitutions, which could be used in the planning of intervention studies.
The aim of this study was to identify dietary factors associated with the probability of having diabetes identified by screening (SDM)4 in Danish men and women aged 30–60 y. A specific objective was to examine whether an alternative statistical approach could provide additional information to already existing statistical approaches used in nutritional epidemiology.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
This study uses baseline data from the Danish population-based Inter99 study, which is an intervention study on diet and lifestyle factors in relation to type 2 diabetes and cardiovascular diseases. The aim, data collection method, and nondietary baseline results of the Inter99 study were reported elsewhere (20,21).
In 1999 the study population comprised 61,301 individuals born in 1939–40, 1944–45, 1949–50, 1954–55, 1959–60, 1964–65, and 1969–70. They were all living in 11 municipalities in the southwestern part of Copenhagen County. All individuals were drawn from the Civil Registration System. An age- and sex-stratified random sample of 13,016 persons was drawn from the study population and 12,934 were eligible for further examination. All of these individuals were invited to a screening program at the Research Centre for Prevention and Health in Glostrup.
All participants gave written consent before taking part in the Inter99 study. The protocol was in accordance with the Helsinki declaration and approved by the Scientific Ethical Committee for Copenhagen County (KA 98155).
Dietary assessment
To assess the participants’ habitual diet, a self-administered FFQ was completed by the participants. The participants were asked to report their dietary intake during the past month. The FFQ consisted of 198 food items and beverages with additional questions regarding portion sizes of some selected food items. When no portion size was specified, a standard portion size for women and men, respectively, was used (22; A. Biltoft-Jensen, Danish Veterinary and Food Administration, personal communication). The quantity consumed was obtained by multiplying the portion size by the corresponding consumption frequency. A detailed description of the questionnaire is published elsewhere (23).
All food items in the FFQ were linked to food items in the Danish Food Composition Databank (24). Estimation of dietary intakes as kJ, g, and mL/d as well as energy percentages (En%) for each subject was based on calculations made using the software program FoodCalc, version 1.3 (25).
Assessment of physical activity and smoking
Lifestyle information about physical activity and smoking was obtained by a self-administered general questionnaire completed before the participants’ first visit to the research center. Based on answers about physical activity level during work and leisure time, all individuals were categorized as either being "physically active" or "physically inactive." Those in the physically inactive group had little activity at work and during leisure time, only minor physical activity at work (sitting/walking) combined with no activity during leisure time, or no activity at work combined with minor activity during leisure time (sitting/walking/cycling). Those categorized as physically active had at least moderate amounts of physical activity at work (walking stairways/heavy work) or during leisure time (sport/competitive sport) (26).
The participants were categorized into 3 groups of smoking status: "daily and occasional smokers," "ex-smokers," and "never-smokers." Missing values of physical activity (n = 58) and smoking status (n = 34) were classified in separate categories and included in the analyses.
Anthropometric measurements
Weight was measured to the nearest 0.1 kg using either an electronic or mechanical standard weight (Seca 707, Seca 710) with the participant wearing light indoor clothes without shoes. Height was measured to the nearest 0.5 cm with the participant wearing no shoes. BMI was calculated as weight in kg divided by the height in m2. Measurement of waist circumference was taken with a tape measure to the nearest 0.5 cm halfway between the lowest point of the costal margin and highest point of the iliac crest.
Oral glucose tolerance test
After an overnight fast, participants without known diabetes underwent a standard 75-g oral glucose tolerance test (OGTT; 75 g anhydrous glucose in 250 mL water). Venous plasma glucose was measured before glucose ingestion and after 120 min. The samples were placed in a tube containing sodium-fluoride, put on ice immediately, and centrifuged within 60 min in a cool-centrifuge. The glucose was analyzed using the hexokinase/G6P-DH technique (Boehringer Mannheim).
Classification of glucose tolerance status
With the use of the fasting plasma glucose (FPG) and the 2-h plasma glucose (2-h PG) values from the OGTT, the participants were classified into categories of glucose tolerance according to the 1999 WHO criteria (27). Individuals with an FPG concentration 
7.0 mmol/L or a 2-h PG concentration 11.1 mmol/L were classified as having SDM. Those who had an FPG concentration < 6.1 mmol/L and a 2-h PG concentration < 7.8 mmol/L were defined as having normal glucose tolerance (NGT). Impaired fasting glycemia (IFG) was defined as FPG between 6.1 and 6.9 mmol/L and a 2-h PG < 7.8 mmol/L, whereas impaired glucose tolerance (IGT) was defined as FPG < 7.0 mmol/L and 2-h PG between 7.8 and 11.0 mmol/L.
Statistical analysis
The dietary intakes of individuals with SDM were analyzed relative to the dietary intakes of individuals with NGT by use of multiple logistic regression models. SAS 8.2 (SAS Institute) was used for data handling, and a P-value of 0.05 was considered significant.
Baseline characteristics. Student’s t test was used to test for significant differences between individuals with NGT and SDM. Because intake of energy, fat, and alcohol were not normally distributed, they were log-transformed before analysis.
Single dietary factor models. The dietary variables included in these models were: intake of total fat (En%), saturated fat (En%), protein (En%), alcohol (En%), carbohydrates (En%), dietary fiber (g), coffee (mL), and tea (mL). All macronutrients and selected food items were analyzed as continuous variables in logistic regression models with adjustment for: 1) age, sex, smoking, physical activity, and total energy intake; and 2) age, sex, smoking, physical activity, total energy intake, BMI, and waist circumference. In these models, the coefficients for the dietary factors describe the increase in odds of SDM associated with a specific change in the dietary factor while total energy intake is kept constant. The increase in one dietary factor is, thus, at the expense of the same amount of energy from other unspecified dietary factors. In the present analysis, increases of 3 En% were chosen for fat, protein, alcohol, and carbohydrate, whereas an increase of 10 g was chosen for dietary fiber and 200 mL for coffee and tea. These increases represent changes that are reasonable according to the range of the data.
Substitution model.
In the substitution model, the En% of all 4 macronutrients were included as variables. Because these variables add up to 100 for any person, only 3 of them can be included in the model. If all 4 macronutrients were included, it would be impossible to estimate the effect of one macronutrient, because of aliasing with the intercept. By including 3 of the macronutrients, it is possible to estimate all relevant contrasts between macronutrient effects. The interpretation of these 3 estimated parameters is the effect of increasing the intake of 1, while keeping the other 2 constant, i.e., at the expense of the nutrient not included as a variable in the model. Similarly, the pairwise contrasts between the 3 estimated parameters are the effects of increasing the intake of 1 nutrient at the expense of another. Specifically, if P, F, C, and A represent En% of protein, fat, carbohydrate, and alcohol, respectively, the following model is considered:
 |
If 1 unit of fat is substituted for 1 unit of carbohydrate the response will be:
 |
Thus, the contrasts between 2 parameters are exactly the effect of a 1-unit substitution. Based on the 4 macronutrients, there are in all 6 such contrasts (6 possible substitutions). It should be noted, however, that these are based only on the 3 estimated parameters (and their variance-covariance matrix for the computation of the standard errors). The confounders in the substitution model were age, sex, smoking, physical activity, total energy intake, BMI, and waist circumference. The reason for not presenting the results without adjustment for BMI and waist circumference in this model was that only a possible direct effect of diet on type 2 diabetes was of interest here.
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Characteristics of the study population. Individuals with SDM were characterized by higher mean age, BMI, and waist circumference and a higher proportion of men and physically inactive persons. Intake of carbohydrate was lower, whereas intake of fat tended to be higher (P = 0.09) in individuals with SDM compared with those with NGT (Table 1). The reported intakes of energy, protein, and alcohol did not differ between the groups.
|
|
No significant associations between intake of protein or alcohol and SDM were observed (Table 2). In an additional analysis (data not shown), we included alcohol as a quadratic term in the regression model with adjustment for all confounders. In this model, the association between alcohol consumption and SDM was slightly U-shaped, although nonsignificant (P = 0.085). The lowest probability of SDM was observed at an alcohol intake of 7.2 En% [
2 beverages per day at an energy intake of 2400 kcal (10.04 MJ)].
Individuals with 200 mL greater coffee consumption had a 9% reduced probability of having SDM both before and after adjustment for BMI and waist circumference. No statistically significant association was found between the intake of tea and SDM; however, a slight inverse association (P = 0.082) was observed before adjustment for BMI and waist circumference.
Effects of substituting macronutrients. When carbohydrate replaced fat, alcohol, or protein, the odds of having SDM were lowered by 7, 10, and 16%, respectively (Table 3). No other macronutrient substitutions were significantly associated with SDM.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
A protective effect of carbohydrate was observed in the single dietary factor model. This effect disappeared when dietary fiber was included in the model, suggesting a protective effect of fiber and not carbohydrate per se. Our finding confirms observations from the Nurses’ Health Study (10) and the Iowa Womens’ Health Study (11), which also used statistical models in which the substituted macronutrient(s) was not taken into consideration. However, other studies did not report any effects of total fiber intake on diabetes risk (6,9,12). On the basis of existing metabolic studies, we suggest that soluble fibers may prevent type 2 diabetes through their ability to suppress postprandial food absorption from the small intestine (34,35), and through fermentation processes in the large intestine where SCFA are produced (36). These mechanisms will diminish the postprandial glycemic and insulinemic response and improve glucose metabolism in the liver, thereby improving blood glucose regulation and insulin sensitivity (37). Another site of action of soluble fiber is through gastric inhibitory polypeptide (GIP), which stimulates insulin release (38). After consumption of a high-fiber meal, the secretion of GIP and insulin will be lower compared with consumption of a low-fiber meal (39); in the long term, this may have beneficial effects on glucose metabolism.
In the substitution model that included all macronutrients, the odds of having diabetes decreased when carbohydrates replaced fat. This supports the adverse effect of fat and the protective effect of carbohydrates observed in the single dietary factor models. Because most population-based studies failed to show any relation between total carbohydrate intake and the development of diabetes (9–11,40), we suggest that the protective effect of carbohydrate observed in this study was caused either by dietary fiber intake and/or by a concomitant decrease in fat intake.
The slightly U-shaped association between alcohol consumption and SDM suggests that alcohol intake > 7 En% may be detrimental for the development of type 2 diabetes. Due to methodological differences, comparison with previous studies is difficult, but the beneficial effect of low-to-moderate alcohol consumption is supported by other studies (13–15). In the substitution model, replacement of alcohol with carbohydrate decreased the odds of having SDM. However, the substitution model was not perfectly suited to include alcohol because the relation between alcohol consumption and diabetes tended to be U-shaped and not linear. The substitution model, though, provides evidence that substitution of alcohol with carbohydrate (i.e., dietary fiber) is beneficial for the prevention of diabetes whether alcohol consumption is reduced from 10 to 7 En%, for example, or from 3 to 0 En%.
On the basis of findings from the model with single dietary components, it is not likely that protein influences the probability of having SDM. This confirms findings from the Nurses’ Health Study (19) and the Finnish and Dutch cohorts of the Seven Country Study (6). However, the substitution model indicates that substitution of protein with carbohydrate is more beneficial than substitution of fat or alcohol with carbohydrate, suggesting that protein could have detrimental effects on glucose metabolism. It is important, however, to reproduce the findings from the substitution model in other studies before any firm conclusions regarding the health aspects of protein intake can be drawn.
In addition to the macronutrients and dietary fiber, we also examined the effect of coffee and tea on the probability of having SDM. We found that higher intake of coffee was associated with decreased odds of SDM. This finding supports prospective studies from Finland (18), the Netherlands (17) and the United States (16) showing that consumption of coffee decreased the risk of developing type 2 diabetes. Coffee contains active components including caffeine, chlorogenic acid, and other micronutrients that influence blood glucose regulation (41). Given the lack of intervention studies, the limited understanding of pathogenic mechanisms, and the limited number of studies, we still find it too premature to give recommendations with respect to coffee consumption. No effect of tea consumption was found in the present study. However, because tea may play a role in blood glucose regulation as demonstrated in a short-term intervention study (42), we recommend further analyses in this area.
The Inter99 study proposes interventions based on lifestyle modification with diet, physical activity, and smoking cessation to individuals at high risk of developing cardiovascular disease or diabetes. Hence, it is possible that those who were obese and overweight or had an unhealthy lifestyle were more likely to participate in the intervention program for lifestyle modification than those who considered themselves to have a healthy lifestyle. The representativeness of the Inter99 population was examined by comparing morbidity between the participants and nonparticipants. This analysis revealed that participants in the Inter99 study had lower morbidity based on hospitalizations than the whole population invited (21). This suggests that the Inter99 population is not more ill than the entire Danish population; therefore, the associations observed in this study may be applied to the entire Danish population.
As in every other observational study, this study has certain limitations. The degree of physical activity was estimated from answers about physical activity level at work and during leisure time. This approach has not been validated, but we found that the combined physical activity variable was very strongly correlated with both insulin resistance [estimated by the homeostasis model assessment (43)] and type 2 diabetes in univariate analyses (data not published), suggesting that this variable can be used as an estimate of physical activity in observational studies.
Underreporting is also a major issue to consider in observational studies using self-reported dietary intake. The energy intake in individuals with SDM was similar to that of individuals with NGT, even though those with SDM had a higher BMI (Table 1). This observation supports the observation that obese individuals underreport their energy intake to a greater degree than normal weight individuals (44). However, based on calculations of Goldberg’s cutoff (45), which is reported elsewhere (23), we did not exclude any individuals, and this may have overestimated some of the observed effects (44).
Many studies examining the relation between diet and diabetes have categorized the dietary variable of interest into quartiles or quintiles. The advantages of this are that it allows calculation of relative risks for each category of dietary intake, and it allows for nonlinear diet-disease relations. The disadvantages are that the interpretation is not straightforward and it affects the choice of statistical approach (46). Furthermore, the use of categorized variables may result in the loss of information; the findings become difficult to generalize because other populations are likely to have different dietary intakes. Thus, continuous-variable models are preferable in studying diet-disease relations when these models fit data and correspond to likely biological explanations of the diet-disease relation under study (46).
Within nutritional epidemiology, there has been a recent shift away from individual nutrients and toward food-based approaches with a focus on dietary patterns, food groups, and combinations of dietary factors. We examined 2 different statistical models, a model with single dietary factors and a model taking more than one nutrient into consideration. Our findings from the single dietary factor model supported findings from most other studies examining the effect of single dietary factors on diabetes. The second statistical approach, the substitution model, was designed to examine the effect of substituting nutrients for each other, thereby adding information to already existing knowledge about the role of diet in type 2 diabetes. This model gave considerable additional information to the models with single dietary factors. For protein, in particular, the substitution model provided new information. The results from such a statistical model may be used in the planning of intervention programs and recommendations aimed at reducing the development of type 2 diabetes. Based on the results from this study it could be expected, for example, that a population-based strategy aimed at increasing the intake of carbohydrate by 3En% at the expense of dietary fat, would decrease the odds of diabetes by 7%. However, the substitution model should be analyzed in other studies to verify that the approach is useful in other populations and settings. We suggest that studies analyzing relations between diet and other chronic diseases also analyze their data in statistical models taking more than one nutrient into consideration because this approach may be useful in all kinds of diet-disease relations.
In conclusion, intakes of total fat and saturated fat were positively associated with SDM. Intake of dietary fiber was inversely associated with SDM; to a large extent, the intake of dietary fiber also explained the inverse association between carbohydrate intake and SDM. Furthermore, we found that coffee consumption was inversely associated with SDM. A statistical approach, in which macronutrients were substituted for each other, provided additional information to a model analyzing single dietary factors. The complex role of diet and other lifestyle factors in the development of type 2 diabetes warrants future studies that use complex statistical approaches, such as factor or cluster analysis, because these kinds of analyses offer flexible models and make characterization of different food and lifestyle patterns possible. Due to the cross-sectional nature of the present study, the associations found should be confirmed in prospective studies and ideally through randomized, long-term intervention studies.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Supported by the Danish Medical Research Council, the Danish Centre for Evaluation and Health Technology Assessment, Novo Nordisk A/S, Copenhagen County, the Danish Heart Foundation, the Danish Diabetes Association, the Danish Pharmaceutical Association, the Augustinus Foundation, the Ib Henriksen Foundation, and the Becket Foundation.
4 Abbreviations used: En%, energy percentage; FPG, fasting plasma glucose; GIP, gastric inhibitory polypeptide; IFG, impaired fasting glycemia; IGT, impaired glucose tolerance; NGT, normal glucose tolerance; OGTT, oral glucose tolerance test; SDM, diabetes identified by screening; 2-h PG, 2-h plasma glucose.
Manuscript received 9 December 2004. Initial review completed 12 January 2005. Revision accepted 18 January 2005.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Drivsholm, T., Ibsen, H., Schroll, M., Davidsen, M. & Borch-Johnsen, K. (2001) Increasing prevalence of diabetes mellitus and impaired glucose tolerance among 60-year-old Danes. Diabet. Med. 18:126-132.[Medline]
2. Dunstan, D. W., Zimmet, P. Z., Welborn, T. A., De-Court Cameron, A. J., Sicree, R. A., Dwyer, T., Colagiuri, S. & Jolley, D., et al (2002) The rising prevalence of diabetes and impaired glucose tolerance—the Australian diabetes, obesity and lifestyle study. Diabetes Care 25:829-834.
3. Wild, S., Roglic, G., Green, A., Sicree, R. & King, H. (2004) Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 27:1047-1053.
4. Astrup, A. (2001) Healthy lifestyles in Europe: prevention of obesity and type II diabetes by diet and physical activity. Public Health Nutr. 4:499-515.[Medline]
5. Steyn, N. P., Mann, J., Bennett, P. H., Temple, N., Zimmet, P., Tuomilehto, J., Lindstrom, J. & Louheranta, A. (2004) Diet, nutrition and the prevention of type 2 diabetes. Public Health Nutr. 7:147-165.[Medline]
6. Feskens, E. J., Virtanen, S. M., Räsänen, L., Tuomilehto, J., Stengård, J., Pekkanen, J., Nissinen, A. & Kromhout, D. (1995) Dietary factors determining diabetes and impaired glucose tolerance. A 20-year follow-up of the Finnish and Dutch cohorts of the Seven Countries Study. Diabetes Care 18:1104-1112.[Abstract]
7. Van Dam, R. M., Stampfer, M., Willett, W. C., Hu, F. B. & Rimm, E. B. (2002) Dietary fat and meat intake in relation to risk of type 2 diabetes in men. Diabetes Care 25:417-424.
8. Marshall, J. A., Hoag, S., Shetterly, S. & Hamman, R. F. (1994) Dietary fat predicts conversion from impaired glucose tolerance to NIDDM. The San Luis Valley Diabetes Study. Diabetes Care 17:50-56.[Abstract]
9. Salmeron, J., Ascherio, A., Rimm, E. B., Colditz, G. A., Spiegelman, D., Jenkins, D. J., Stampfer, M. J., Wing, A. L. & Willett, W. C. (1997) Dietary fiber, glycemic load, and risk of NIDDM in men. Diabetes Care 20:545-550.[Abstract]
10. Salmeron, J., Manson, J. E., Stampfer, M. J., Colditz, G. A., Wing, A. L. & Willett, W. C. (1997) Dietary fiber, glycemic load, and risk of non-insulin-dependent diabetes mellitus in women. J. Am. Med. Assoc. 277:472-477.
11. Meyer, K. A., Kushi, L. H., Jacobs, D. R., Slavin, J., Sellers, T. A. & Folsom, A. R. (2000) Carbohydrates, dietary fiber, and incident type 2 diabetes in older women. Am. J. Clin. Nutr. 71:921-930.
12. Stevens, J., Ahn, K., Juhaeri, , Houston, D., Steffan, L. & Couper, D. (2002) Dietary fiber intake and glycemic index and incidence of diabetes in African-American and white adults. Diabetes Care 25:1715-1721.
13. Wannamethee, S. G., Camargo, C. A., Manson, J. E., Willett, W. C. & Rimm, E. B. (2003) Alcohol drinking patterns and risk of type 2 diabetes mellitus among younger women. Arch. Intern. Med. 163:1329-1336.
14. Nakanishi, N., Suzuki, K. & Tatara, K. (2003) Alcohol consumption and risk for development of impaired fasting glucose or type 2 diabetes in middle-aged Japanese men. Diabetes Care 26:48-54.
15. Conigrave, K. M., Hu, B. F., Camargo, C. A., Stampfer, M. J., Willett, W. C. & Rimm, E. B. (2001) A prospective study of drinking patterns in relation to risk of type 2 diabetes among men. Diabetes 50:2390-2395.
16. Salazar-Martinez, E., Willett, W. C., Ascherio, A., Manson, J. E., Leitzmann, M. F., Stampfer, M. J. & Hu, F. B. (2004) Coffee consumption and risk for type 2 diabetes mellitus. Ann. Intern. Med. 140:1-8.
17. Van Dam, R. M. & Feskens, E.J.M. (2002) Coffee consumption and risk of type 2 diabetes mellitus. Lancet 360:1477-1478.[Medline]
18. Tuomilehto, J., Hu, G., Bidel, S., Lindstrom, J. & Jousilahti, P. (2004) Coffee consumption and risk of type 2 diabetes mellitus among middle-aged Finnish men and women. J. Am. Med. Assoc. 291:1213-1219.
19. Colditz, G. A., Manson, J. E., Stampfer, M. J., Rosner, B., Willett, W. C. & Speizer, F. E. (1992) Diet and risk of clinical diabetes in women. Am. J. Clin. Nutr. 55:1018-1023.
20. Glümer, C., Jørgensen, T. & Borch-Johnsen, K. (2003) Prevalences of diabetes and impaired glucose regulation in a Danish population: the Inter99 study. Diabetes Care 26:2335-2340.
21. Jørgensen, T., Borch-Johnsen, K., Thomsen, T. F., Ibsen, H., Glümer, C. & Charlotta, P. (2003) A randomized non-pharmacological intervention study for prevention of ischaemic heart disease: baseline results Inter99 (1). Eur. J. Cardiovasc. Prev. Rehab. 10:377-386.
22. Haraldsdóttir, J., Seppänen, R., Steinrímsdóttir, L., Trygg, K. & Hagman, U. (1998) Portion sizes—Nordic standard portions of food and foodstuffs (Portionsstorleker - Nordiska standardportioner av mat och livsmedel) 1998:1-41 Nordic Council & Council of Ministers Copenhagen, Denmark .
23. Lau, C., Færch, K., Glümer, C., Toft, U., Tetens, I., Borch-Johnsen, K. & Jørgensen, T. (2004) Evaluation of dietary intake in a Danish population: the Inter99 study. Scand. J. Nutr. 48:136-143.
24. Møller, A., Saxholt, E., Christensen, T. & Hartkopp, H. () Danish Food Composition Databank Revision 5.0 [Online] Danish Institute for Food and Veterinary Research, Ministry of Family and Consumer Affairs http://www.foodcomp.dk/fvdb_default.htm [accessed Nov 20, 2003].
25. Lauritsen, J. () FoodCalc v1.3 [Online] http://www.ibt.ku.dk/jesper/foodcalc/default.htm [accessed Feb 12, 2004].
26. Saltin, B. & Grimby, G. (1968) Physiological analysis of middle-aged and old former athletes. Circulation 38:1104-1115.
27. World Health Organization (1999) Definition, Diagnosis and Classification of Diabetes Mellitus and Its Complications. Report of a WHO Consultation. Part 1, Diagnosis and Classification of Diabetes Mellitus 1999:1-59 WHO Geneva Switzerland .
28. Randle, P. J., Garland, P. B., Hales, C. N. & Newsholme, E. A. (1963) The glucose fatty-acid cycle. Its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet 1:785-789.[Medline]
29. Storlien, L. H., Kriketos, A. D., Jenkins, A. B., Baur, L. A., Pan, D. A., Tapsell, L. C. & Calvert, G. D. (1997) Does dietary fat influence insulin action?. Ann. N.Y. Acad. Sci. 827:287-301.[Medline]
30. Pan, D. A., Lillioja, S., Kriketos, A. D., Milner, M. R., Baur, L. A., Bogardus, C., Jenkins, A. B. & Storlien, L. H. (1997) Skeletal muscle triglyceride levels are inversely related to insulin action. Diabetes 46:983-988.[Abstract]
31. Ezaki, O. (1997) Regulatory elements in the insulin-responsive glucose transporter (GLUT4) gene. Biochem. Biophys. Res. Commun. 241:1-6.[Medline]
32. Borkman, M., Storlien, L. H., Pan, D. A., Jenkins, A. B., Chisholm, D. J. & Campbell, L. V. (1993) The relation between insulin sensitivity and the fatty-acid composition of skeletal-muscle phospholipids. N. Engl. J. Med. 328:238-244.
33. Unger, R. H. (1995) Lipotoxicity in the pathogenesis of obesity-dependent NIDDM. Genetic and clinical implications. Diabetes 44:863-870.[Abstract]
34. Anderson, J. W., Gustafson, N. J., Bryant, C. A. & Tietyen-Clark, J. (1987) Dietary fiber and diabetes: a comprehensive review and practical application. J. Am. Diet. Assoc. 87:1189-1197.[Medline]
35. Hallfrisch, J. & Behall, K. M. (2000) Mechanisms of the effects of grains on insulin and glucose responses. J. Am. Coll. Nutr. 19:320S-325S.
36. Englyst, H. N. & Hudson, G. J. (2000) Carbohydrates. Garrow, J. S. James, W. P. Ralph, A. eds. Human Nutrition and Dietetics 2000:61-76 Churchill Livingstone London, UK .
37. Venter, C. S., Vorster, H. H. & Cummings, J. H. (1990) Effects of dietary propionate on carbohydrate and lipid metabolism in healthy volunteers. Am. J. Gastroenterol. 85:549-553.[Medline]
38. Brown, J. C., Dryburgh, J. R., Ross, S. A. & Dupre, J. (1975) Identification and actions of gastric inhibitory polypeptide. Recent Prog. Horm. Res. 31:487-532.
39. Jenkins, D. J., Wolever, T. M., Collier, G. R., Ocana, A., Rao, A. V., Buckley, G., Lam, Y., Mayer, A. & Thompsom, L. U. (1987) Metabolic effects of a low-glycemic-index diet. Am. J. Clin. Nutr. 46:968-975.
40. Ekblond, A., Mellemkjaer, L., Tjønneland, A., Suntum, M., Stripp, C., Overvad, K., Johansen, C. & Olsen, J. H. (2000) A cross-sectional study of dietary habits and urinary glucose excretion—a predictor of non-insulin-dependent diabetes mellitus. Eur. J. Clin. Nutr. 54:434-439.[Medline]
41. Johnston, K. L., Clifford, M. N. & Morgan, L. M. (2003) Coffee acutely modifies gastrointestinal hormone secretion and glucose tolerance in humans: glycemic effects of chlorogenic acid and caffeine. Am. J. Clin. Nutr. 78:728-733.
42. Hosoda, K., Wang, M.-F., Liao, M.-L., Chuang, C.-K., Iha, M., Clevidence, B. & Yamamoto, S. (2003) Antihyperglycemic effect of oolong tea in type 2 diabetes. Diabetes Care 26:1714-1718.
43. Matthews, D. R., Hosker, J. P., Rudenski, A. S., Naylor, B. A., Treacher, D. F. & Turner, R. C. (1985) Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28:412-419.[Medline]
44. Heitmann, B. L. (1999) Assessment of dietary intake. Guy-Grand, B. Aihaud, G. eds. Progress in Obesity Research 1999:701-706 John Libbey & Company Toronto, Canada .
45. Goldberg, G. R., Black, A. E., Jebb, S. A., Cole, T. J., Murgatroyd, P. R., Coward, W. A. & Prentice, A. M. (1991) Critical evaluation of energy intake data using fundamental principles of energy physiology: 1. Derivation of cut-off limits to identify under-recording. Eur. J. Clin. Nutr. 45:569-581.[Medline]
46. Willett, W. C., Howe, G. R. & Kushi, L. H. (1997) Adjustment for total energy intake in epidemiologic studies. Am. J. Clin. Nutr. 65(suppl. 4):1220S-1228S.
This article has been cited by other articles:
|
|
 |
|
  J. A Greenberg, C. N Boozer, and A. Geliebter Coffee, diabetes, and weight control. Am. J. Clinical Nutrition, October 1, 2006; 84(4): 682 - 693. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. van Dam, W. C. Willett, J. E. Manson, and F. B. Hu Coffee, Caffeine, and Risk of Type 2 Diabetes: A prospective cohort study in younger and middle-aged U.S. women Diabetes Care, February 1, 2006; 29(2): 398 - 403. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
