![[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}
|
|
|
|
|
||
* Department of Epidemiology, German Institute of Human Nutrition Potsdam-Rehbruecke, Germany, and
Cancer Research Center of Hawaii, Honolulu
2To whom correspondence should be addressed. E-mail: mschulz{at}mail.dife.de.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • food pattern • weight change • reduced rank regression • European Prospective Investigation into Cancer and Nutrition • EPIC-Potsdam study
Weight gain results from an imbalance between energy intake and energy expenditure. Current recommendations to avoid weight gain include limiting dietary energy from fat to 35% and meeting at least 45% of energy requirements with carbohydrates (1). However, considering the dramatic development of obesity rates in the United States and other westernized countries, despite a decline in fat intake as a percentage of total energy over the past decades (2–4), the applicability and the success of these recommendations may be debatable. The question arises as to whether the focus of body-weight control should be on energy density or on more practical characteristics of diet, such as the choice of specific foods or food groups. Although a number of reviews of the association between fat intake and body-weight development revealed a direct association (5–8), the scientific evidence has been challenged (9–11).
In recent years, the integration and the compression of full dietary information using dietary pattern approaches to assess the association between diet and body-weight development has become of increasing interest among nutritional epidemiologists (12). From cross-sectional studies, we learned that no consistent associations could be identified between BMI or obesity and dietary patterns (13). However, a prospective study design relating baseline dietary information to subsequent change in body weight may shed new light on the association between diet and body weight. To date, prospective investigations to examine this association are scarce. For subjects of the Baltimore Longitudinal Study of Aging, a healthy cluster and a reduced-fat dairy, fruits, and fiber pattern were associated with significantly lower weight gains over time (14,15). A longitudinal study of food-intake patterns and BMI change in Danish adults revealed inconsistent and only minor associations. For instance, a dietary pattern characterized by high intakes of whole-grain products, fruits, and vegetables (labeled as "green" dietary pattern) was not associated at all with prospective weight change (16). Cluster analysis and factor analysis constituted the prevailing approaches of dietary pattern analysis in previous studies (12).
The aim of the present study was to identify a food pattern based on dietary composition information to predict subsequent annual weight change over a 4-y follow-up period in a German middle-aged population of the European Prospective Investigation into Cancer and Nutrition (EPIC)3-Potsdam study (17). We applied a new dietary pattern approach, reduced rank regression (RRR), proposed for nutritional epidemiology research by Hoffmann et al. (18), which has proven useful in identifying dietary patterns in relation to chronic diseases (18,19) and which is superior to principal component analysis (PCA) if variation in the selected nutrient densities is more relevant for weight change than the unspecified variation in intake of all foods (20).
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Approval for all study procedures was given by the Ethical Committee of the State of Brandenburg, Germany, and informed consent was obtained from all study participants.
For this analysis, men aged 24–69 y and women aged 19–70 y, with complete data on body weight at baseline and who returned the first and second follow-up questionnaire, were eligible (n = 25,079). Pregnant women either at baseline or at follow-up were excluded from the analysis (n = 118), as well as 3 subjects with implausible follow-up weights, leaving 24,958 subjects for the statistical analysis. The mean follow-up time was 4.4 y, ranging between 1.6 and 7.5 y.
Assessment of body weight and diet. Baseline body weight was measured without shoes and in light clothes on a calibrated scale to the nearest 0.1 kg by trained technicians. Weight at follow-up examinations at 2 y and 4 y was self-reported by the participants in the respective follow-up questionnaires.
A self-administered, validated FFQ was applied to obtain information on the habitual diet of the past 12 mo before baseline (24). The FFQ consisted of 148 items, included questions about individual portion sizes, and additional questions about fat content of certain food items. The food items were collapsed into 49 food groups based on culinary usage or nutrient profiles (25), and absolute intakes in gram per day were computed. Dietary changes that occurred during follow-up were assessed at both follow-up examinations. Subjects were asked if they had changed their dietary habits compared with that of 2 y ago (yes; yes, partly; no). The answers "yes" and "yes, partly" were collapsed to yes for the analysis.
Information about sociodemographic characteristics, physical activity, medications, previous illnesses, and occupation were assessed at baseline with a questionnaire and in a face-to-face interview. Self-reports on incident diseases were obtained at each follow-up and subsequently were verified.
Statistical analysis. Weight change per year was computed by fitting a linear regression line to each person’s body weight data and determining the slope. A positive value indicates weight gain over time, whereas a negative value indicates weight loss.
To identify a food pattern that is related to annual weight change we applied the dietary pattern technique RRR. The RRR method uses both data from the study and prior knowledge to derive dietary patterns and therefore represents a so-called a posteriori method. A detailed description of the method can be found in Hoffmann et al. (18). Briefly, RRR extracts factors that maximize the proportion of explained variation in a set of response variables, which are considered to be related to the outcome of interest. In the present study, the nutrient densities of the dietary variables total fat, total carbohydrates, and fiber (g fat per 1 MJ, g carbohydrates per 1 MJ, and g fiber per 1 MJ), which are presumed to affect prospective weight change (26–32), were chosen as response variables. Three factors were extracted. The first RRR factor explained 53.4% of total variation in all 3 response variables (43.0% of variation in fiber density, 52.0% of variation in fat density, and 65.3% of variation in carbohydrate density). The subsequent 2 RRRs explained only 20.5% and 9.5%, respectively, of response variation and therefore were not further examined. We applied the RRR method to the entire study population, as well as separately, to males and females. Because the overall pattern did not markedly differ from the gender-specific patterns, indicated by the very similar factor loadings for men and women, the overall pattern was used (Table 1). To better translate the derived dietary pattern score to a food pattern and eventually to other studies, the dietary pattern score was shortened and simplified. For this purpose, the 10 food groups that contributed most to interindividual variation in the dietary pattern score (5 food groups being directly associated with the score and 5 food groups being inversely associated with the score) were identified. The proportion of variance explained by the single food groups was obtained by multiplying the standardized score parameter with the correlation coefficient between the dietary pattern score and each food group times 100 [for details see (19)]. Also, these 10 food groups were used to derive a so-called simplified food pattern to construct a less population-dependent pattern variable of similar content as the original dietary pattern score. The simplified pattern score represents the sum of the standardized food group variables, which explained most of the interindividual variation in the original dietary pattern score. The method and its applicability have been described in detail by Schulze et al. (33). The simplified pattern score defined by the food groups given in Table 1 was used to evaluate the association between the pattern and the annual weight change and, through the remainder of the text, is referred to as the food pattern.
|
2 test) was performed for categorical variables. Analysis of covariance was used separately for men and for women to compare mean outcome levels between quintiles of the food pattern, adjusted for effects of the covariates (34). We first evaluated the impact of potential effect modifiers on the association between the food pattern and the annual weight change by fitting models, including interaction terms for the simplified pattern quintile variable, with the following variables: age at baseline (median split: <50, 50+ y), BMI status (<25, 25 to 30, 30+), and change in diet (yes, no). There was evidence of effect modification by age, BMI status, and dietary change after baseline (each P-value of interaction term < 0.05 except for age) on mean weight change per year. Thus, mean outcome levels are presented in strata of age categories, BMI status, and dietary change. In addition, the impact of potential confounders was assessed on an individual basis and then in the stratified models. Relevant confounders that were retained in the final analysis included leisure time physical activity, change in smoking status after baseline, occupation at baseline, and number of incident diseases. In addition to these variables, the stratification variables were included as covariates in the full model. The women’s models additionally controlled for hormone replacement therapy at baseline. We also adjusted our models for total energy intake to separate the quantitative effect of diet and the qualitative effect expressed by the simplified pattern. However, this adjustment had no effect on the risk estimates and was left out from the final models.
All statistical tests were two-sided, and a P-value of <0.05 was considered significant. The statistical analysis was performed using SAS for Windows V8 (SAS Institute).
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
To evaluate the association between the identified food pattern and the annual weight change, we used a linear regression model approach and estimated mean weight change per year as a function of the food-pattern score controlled for other covariates (Table 3). Mean annual weight change across strata of male and female study subjects ranged from –0.36 kg/y for obese women to 0.33 kg/y for normal-weight men. With all subjects, after adjustment for covariates, we saw a weight gain of 0.12 kg/y among subjects in the lowest quintile of the food-pattern score and a nearly stable weight among subjects of the highest quintile. The annual weight changes in the 4th and 5th quintile of the food-pattern score were significantly different from the annual weight change in the first quintile. This means that subjects with a diet characterized by high intakes of whole-grain bread, fresh fruit, fruit juices, grains and cereals, and raw vegetables, and by low intakes of processed meat, butter, high-fat cheese, margarine, and meat generally maintained their body weight, whereas subjects without these dietary habits were more likely to gain weight. This association was found for all subjects combined, as well as for males and females separately. However, the association of the food pattern and the annual weight change differed according to strata of the effect modifying variables.
|
In females, the most pronounced difference in predictability of the food pattern was found in BMI status. The food pattern predicted annual weight change for the normal-weight range (P < 0.0001) better than for the group of obese women (P = 0.1841). Worth mentioning is the direction of the weight change across BMI classes; whereas normal-weight and overweight women on average gained weight over time, the obese women reported a weight loss. The weight gained among normal-weight women with a high food-pattern score was significantly lower than the weight gain of women of the lowest quintile of the food-pattern score. A change in diet did not affect the association of the food pattern with annual weight change. However, those who changed their diet, on average, maintained their body weight and even lost some weight with an increasing food-pattern score, whereas women with no change in diet on average gained weight with significantly lower weight gains in the upper quintiles than in the bottom quintile of the food-pattern score.
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
We compared our findings with findings from previous studies of dietary patterns and prospective weight change. There are at least 3 studies (14–16) to date that examined empirically derived dietary patterns in relation to prospective weight change. The analyses by Newby et al. [cluster analysis (14), factor analysis (15)] of the Baltimore study to derive dietary patterns each identified a healthy dietary pattern, characterized by high intake of high-fiber grains and cereals, reduced fat dairy, fruits, and vegetables, and by low intake of refined grains, meats, soda, and high-fat dairy, which was associated with substantially less weight gain over time. Our findings are in agreement with these studies, assuming that dietary practices of subjects in the highest quintile of the food-pattern score of the present study resemble those in the healthy dietary pattern in the Baltimore study. A study (16) from Denmark found no association of a "green" dietary pattern, characterized by a high consumption of whole grains, fruits, and vegetables, and thus resembles a high score of the food pattern identified in the present study, with prospective weight change. This study suffered from a high drop-out rate at follow-up examination, which may have diluted the association under investigation. Other points of reference are studies that identified food groups predictive of prospective weight gain. In a previous study (35), we showed that, among women, high-fat foods such as fats, sauces, and meat predicted subsequent weight gain, whereas the consumption of cereals was predictive of weight loss. Although in this study we did not use the full dietary information in a pattern approach but used single food-group intakes, it strengthens the importance of the food pattern identified in the present study with regard to body-weight development. Further studies integrating food-intake information were equivocal: some studies found evidence that consumption of meat or sweets was related to increases in body weight (36,37), whereas other investigators reported opposite or null associations between examined food groups and weight change (38,39).
An important aspect of the present study is that the association of the food pattern we identified with annual weight change of the subsequent 4-y period was confined to nonobese subjects. This merits special attention, because the impact of a diet characterized by high-fiber low-fat food choices on weight change appears to be prevention of weight gain instead of predicting weight loss. The obese individuals of the present study reported follow-up weights that were lower than their measured baseline weights, indicating a weight loss. This might be the reason why we did not find an association among the obese subjects. Another possible explanation can be misclassification bias of either the exposure (diet) or the outcome (annual weight change). It is known that diet is reported biased, especially for foods considered salubrious (fruits and vegetables) or foods considered unhealthy (fats and meat). This phenomenon is even more pronounced among overweight subjects (40,41). Correspondingly, overweight subjects tend to underreport their body weight (42). Because we had to rely on self-reported weights at follow-up, underreporting may have introduced misclassification of the outcome, which, as a result, diluted any potential association between diet and weight change detected. Furthermore, there is always the possibility of residual confounding, and we can never rule out that unmeasured or imprecisely measured factors have confounded the association.
By choosing energy density variables as the base for the identification of the dietary pattern in the present study, we assumed that the composition of the diet may affect body-weight development. This assumption affected the selection of the food groups representing the food pattern and is subject to discussion. There are at least 7 prospective investigations examining the association between dietary composition and subsequent weight change. Attention focused on the effect of dietary fat as percent energy in relation to weight change. However, results are inconclusive: whereas some studies detected a positive association (27,29,32,43,44), others reported no association (45,46). At least one study reported an inverse association between percent energy from carbohydrates or fiber density and prospective weight change (29). Although it is a highly debated topic, whether total energy intake or the relative contribution of macronutrients to energy is to be recommended to avoid weight gain, our results suggest that dietary behavior, which reflects the current recommendation to limit fat intake to 35% of energy and to increase fiber intake (1), can avoid or at least minimize weight gain in an adult population.
We applied a new dietary-pattern approach to examine the association between dietary habits and prospective weight change by focusing on dietary composition. The benefit of using RRR compared with directly choosing energy densities as predictors of prospective weight change is that the effect of increasing or decreasing the usual intake of a specific food group on body weight can be estimated. This advantage follows from the fact that for any change of habitual dietary intake, the corresponding change of pattern score can be calculated. The RRR method is a powerful technique, because it combines both a priori and a posteriori information of the exposure-outcome association by defining a set of variables assumed to be linked to the outcome and explaining variation in those variables with the dietary information at hand. This is novel in nutritional epidemiology and has proven informative in examining associations between dietary patterns and some outcome (18,19). Previous pattern approaches, with the exception of confirmatory factor analysis, used only either a priori information or the data at hand.
A question often raised, given the more data-driven nature of the RRR method, as well as of other pattern methods, is to what extent is the identified dietary pattern entirely unique to the underlying study population, and thus of limited value for inferences. Because of this and the ease of understanding and of disseminating the dietary pattern into public health practices, we chose the simplified food-pattern approach. Using this approach, the major weakness of the pattern technique was limited by concisely naming the contributing food groups, along with their intakes and direction of association with the pattern score. Thus, the food pattern identified in the present study is easily reproducible in other study populations with similar quality of dietary variables.
Nevertheless, our application of the RRR method has several limitations. First, although we tried to define homogeneous food groups, the number and definitions of the food groups are somewhat arbitrary and may affect the derived patterns. Grouping may arbitrarily limit the variation, and groups composed of several foods are considered to be similar to groups consisting of only one food. Moreover, RRR and PCA work with the default of centered and scaled food groups, which implicitly gives equal weight to all food groups, regardless whether the foods are consumed frequently in high amounts or infrequently in small amounts. However, in contrast to PCA patterns, the RRR patterns are not affected by standardization of food groups, because they focus on the variation in the selected nutrient densities. Second, there is no well-accepted statistical criterion to specify the number of derived patterns. In the present study, we reported and analyzed only the first pattern. This seems to be unusual, because in previous PCA applications, at least 2 patterns were reported. However, the RRR method is a dimension-reduction procedure that starts from the dimension of responses, which is only 3 in our study and therefore is much lower than the number of food groups, which is the starting point of PCA. Therefore, a reduction from dimension 3 to 1 seems reasonable. Moreover, although the response variations explained by the second and third RRR patterns were 20.5% and 9.5%, respectively, both patterns explained together much less variation in responses than the first RRR pattern alone. Third, the choice of the 3 nutrient densities as responses in the RRR analysis is neither objective nor unique. Choosing different dietary components as responses will clearly result in distinct patterns that must not be related to subsequent weight gain or loss. However, minor changes of the set of response variables, such as adding a fourth nutrient density, will not have a dramatic effect on the results, by our calculations (not shown), and a comprehensive sensitivity analysis focused on this issue (19).
Given the relatively small mean weight change in our study, which appears to contrast the general raise of obesity rates (47), the question is whether the association we found between the food pattern and the annual weight change is relevant to other populations; to address this concern, consider the source of the follow-up information on weight presented in this study. These were self-reported and were compared with measured body weights at baseline. This is a clear limitation of the study. However, we still detected a significant association between the baseline diet and subsequent weight change, which is considered a rather conservative estimate of the true association, given the measurement error in the weight-change variable. Furthermore, our study population is a middle-aged population, which is not characterized by massive adult weight gain. In Germany, adult weight gain is seen among young men between the ages 30 y and 40 y (47). Because only a small percentage of our study population is in this age range, we did not expect to see large mean weight gains. We also compared our annual weight changes with weight changes measured in other studies. Togo et al. (16) reported a mean 5-y change in BMI of 0.2 kg/m2 among 60-y-old subjects. This would be 0.12 kg/y, assuming a mean height of 1.70 m and resembles our observed annual weight gain. The annual BMI change in the Baltimore study (14) was reported as 0.11 kg/m2 per year, which converts to about 0.32 kg/y for a person of average height. Unlike our study, in these studies, both baseline and follow-up weights were measured. Taking these weights into account, we do not find that the associations detected in the present study are of limited relevance. However, the absolute value of mean annual weight change of the present study needs to be interpreted with caution, because only baseline weights were measured.
In summary, from our data, we conclude that a food pattern characterized by low-fat, high-fiber, high-carbohydrate food choices, such as whole-grain bread, fruits, grains, and cereals, may help to maintain body weight or to prevent excess weight gain over time. Although the predictive value of this dietary pattern was confined to nonobese subjects, this study lends support to the current recommendations regarding macronutrient composition of the diet. Following the food pattern identified in the present study will help meet the recommendations for dietary fat, carbohydrate, and fiber intake and, consequently, may not result in excess energy intake and thus in weight gain. The association between dietary practices and weight change among obese individuals warrants further investigations.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
3 Abbreviations used: EPIC, European Prospective Investigation into Cancer and Nutrition; PCA, principal component analysis; RRR, reduced rank regression.
Manuscript received 21 October 2004. Initial review completed 16 November 2004. Revision accepted 23 January 2005.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Panel on Dietary Reference Intakes for Macronutrients (2002) Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids (Macronutrients) 2002 Institute of Medicine of the National Academies. National Academy Press (http://www.nap.edu).
2. Daily dietary fat and total food-energy intakes—Third National Health and Nutrition Examination Survey, Phase 1, 1988–91 (1994) Morb. Mortal Wkly. Rep. 43:116-117, 123–125.[Medline]
3. Flegal, K. M., Carroll, M. D., Kuczmarski, R. J. & Johnson, C. L. (1998) Overweight and obesity in the United States: prevalence and trends, 1960–1994. Int. J. Obes. Relat. Metab. Disord. 22:39-47.[Medline]
4. Flegal, K. M., Carroll, M. D., Ogden, C. L. & Johnson, C. L. (2002) Prevalence and trends in obesity among US adults, 1999–2000. J. Am. Med. Assoc. 288:1723-1727.
5. Peters, J. C. (2003) Dietary fat and body weight control. Lipids 38:123-127.[Medline]
6. Astrup, A., Grunwald, G. K., Melanson, E. L., Saris, W. H. & Hill, J. O. (2000) The role of low-fat diets in body weight control: a meta-analysis of ad libitum dietary intervention studies. Int. J. Obes. Relat. Metab. Disord. 24:1545-1552.[Medline]
7. Lissner, L. & Heitmann, B. L. (1995) The dietary fat: carbohydrate ratio in relation to body weight. Curr. Opin. Lipidol. 6:8-13.[Medline]
8. Rolls, B. J. & Shide, D. J. (1992) The influence of dietary fat on food intake and body weight. Nutr. Rev. 50:283-290.[Medline]
9. Willett, W. C. (1998) Is dietary fat a major determinant of body fat?. Am. J. Clin. Nutr. 67:556S-562S.[Abstract]
10. Stern, L., Iqbal, N., Seshadri, P., Chicano, K. L., Daily, D. A., McGrory, J., Williams, M., Gracely, E. J. & Samaha, F. F. (2004) The effects of low-carbohydrate versus conventional weight loss diets in severely obese adults: one-year follow-up of a randomized trial. Ann. Intern. Med. 140:778-785.
11. Yancy, W. S., Jr, Olsen, M. K., Guyton, J. R., Bakst, R. P. & Westman, E. C. (2004) A low-carbohydrate, ketogenic diet versus a low-fat diet to treat obesity and hyperlipidemia: a randomized, controlled trial. Ann. Intern. Med. 140:769-777.
12. Newby, P. K. & Tucker, K. L. (2004) Empirically derived eating patterns using factor or cluster analysis: a review. Nutr. Rev. 62:177-203.[Medline]
13. Togo, P., Osler, M., Sorensen, T. I. & Heitmann, B. L. (2001) Food intake patterns and body mass index in observational studies. Int. J. Obes. Relat. Metab. Disord. 25:1741-1751.[Medline]
14. Newby, P. K., Muller, D., Hallfrisch, J., Qiao, N., Andres, R. & Tucker, K. L. (2003) Dietary patterns and changes in body mass index and waist circumference in adults. Am. J. Clin. Nutr. 77:1417-1425.
15. Newby, P., Muller, D., Hallfrisch, J., Andres, R. & Tucker, K. L. (2004) Food patterns measured by factor analysis and anthropometric changes in adults. Am. J. Clin. Nutr. 80:504-513.
16. Togo, P., Osler, M., Sorensen, T. I. & Heitmann, B. L. (2004) A longitudinal study of food intake patterns and obesity in adult Danish men and women. Int. J. Obes. Relat. Metab. Disord. 28:583-593.[Medline]
17. Boeing, H., Wahrendorf, J. & Becker, N. (1999) EPIC-Germany–a source for studies into diet and risk of chronic diseases. European Investigation into Cancer and Nutrition. Ann. Nutr. Metab. 43:195-204.[Medline]
18. Hoffmann, K., Schulze, M. B., Schienkiewitz, A., Nothlings, U. & Boeing, H. (2004) Application of a new statistical method to derive dietary patterns in nutritional epidemiology. Am. J. Epidemiol. 159:935-944.
19. Hoffmann, K., Zyriax, B. C., Boeing, H. & Windler, E. (2004) A dietary pattern derived to explain biomarker variation is strongly associated with the risk of coronary artery disease. Am. J. Clin. Nutr. 80:633-640.
20. Kroke, A. (2004) Re: "Application of a new statistical method to derive dietary patterns in nutritional epidemiology.". Am. J. Epidemiol. 160:1132 author reply 1132–1133.
21. Boeing, H., Korfmann, A. & Bergmann, M. M. (1999) Recruitment procedures of EPIC-Germany. European Investigation into Cancer and Nutrition. Ann. Nutr. Metab. 43:205-215.[Medline]
22. Riboli, E., Hunt, K. J., Slimani, N., Ferrari, P., Norat, T., Fahey, M., Charrondiere, U. R., Hemon, B. & Casagrande, C., et al (2002) European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection. Public Health Nutr. 5:1113-1124.[Medline]
23. Bergmann, M. M., Bussas, U. & Boeing, H. (1999) Follow-up procedures in EPIC–Germany–data quality aspects. Ann. Nutr. Metab. 43:225-234.[Medline]
24. Kroke, A., Klipstein-Grobusch, K., Voss, S., Moseneder, J., Thielecke, F., Noack, R. & Boeing, H. (1999) Validation of a self-administered food-frequency questionnaire administered in the European Prospective Investigation into Cancer and Nutrition (EPIC) Study: comparison of energy, protein, and macronutrient intakes estimated with the doubly labeled water, urinary nitrogen, and repeated 24-h dietary recall methods. Am. J. Clin. Nutr. 70:439-447.
25. Schulze, M. B., Hoffmann, K., Kroke, A. & Boeing, H. (2001) Dietary patterns and their association with food and nutrient intake in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam Study. Br. J. Nutr. 85:363-373.[Medline]
26. Birketvedt, G. S., Aaseth, J., Florholmen, J. R. & Ryttig, K. (2000) Long-term effect of fibre supplement and reduced energy intake on body weight and blood lipids in overweight subjects. Acta Medica (Hradec Kralove) 43:129-132.
27. Lissner, L., Heitmann, B. L. & Bengtsson, C. (1997) Low-fat diets may prevent weight gain in sedentary women: prospective observations from the population study of women in Gothenburg, Sweden. Obes. Res. 5:43-48.[Medline]
28. Liu, S., Willett, W. C., Manson, J. E., Hu, F. B., Rosner, B. & Colditz, G. (2003) Relation between changes in intakes of dietary fiber and grain products and changes in weight and development of obesity among middle-aged women. Am. J. Clin. Nutr. 78:920-927.
29. Ludwig, D. S., Pereira, M. A., Kroenke, C. H., Hilner, J. E., Van Horn, L., Slattery, M. L. & Jacobs, D. R., Jr (1999) Dietary fiber, weight gain, and cardiovascular disease risk factors in young adults. J. Am. Med. Assoc. 282:1539-1546.
30. Mueller-Cunningham, W. M., Quintana, R. & Kasim-Karakas, S. E. (2003) An ad libitum, very low-fat diet results in weight loss and changes in nutrient intakes in postmenopausal women. J. Am. Diet Assoc. 103:1600-1606.[Medline]
31. Paeratakul, S., Popkin, B. M., Keyou, G., Adair, L. S. & Stevens, J. (1998) Changes in diet and physical activity affect the body mass index of Chinese adults. Int. J. Obes. Relat. Metab. Disord. 22:424-431.[Medline]
32. Sherwood, N. E., Jeffery, R. W., French, S. A., Hannan, P. J. & Murray, D. M. (2000) Predictors of weight gain in the Pound of Prevention Study. Int. J. Obes. Relat. Metab. Disord. 24:395-403.[Medline]
33. Schulze, M. B., Hoffmann, K., Kroke, A. & Boeing, H. (2003) An approach to construct simplified measures of dietary patterns from exploratory factor analysis. Br. J. Nutr. 89:409-419.[Medline]
34. Kleinbaum, D., Kupper, L. L. & Muller, K. E. (1988) Applied regression analysis and other multivariable methods 1988 PWS-KENT Publishing Boston, MA.
35. Schulz, M., Kroke, A., Liese, A. D., Hoffmann, K., Bergmann, M. M. & Boeing, H. (2002) Food groups as predictors for short-term weight changes in men and women of the EPIC-Potsdam cohort. J. Nutr. 132:1335-1340.
36. Kahn, H. S., Tatham, L. M., Rodriguez, C., Calle, E. E., Thun, M. J. & Heath, C. W., Jr (1997) Stable behaviors associated with adults’ 10-year change in body mass index and likelihood of gain at the waist. Am. J. Public Health 87:747-754.
37. French, S. A., Jeffery, R. W., Forster, J. L., McGovern, P. G., Kelder, S. H. & Baxter, J. E. (1994) Predictors of weight change over two years among a population of working adults: the Healthy Worker Project. Int. J. Obes. Relat. Metab. Disord. 18:145-154.[Medline]
38. Parker, D. R., Gonzalez, S., Derby, C. A., Gans, K. M., Lasater, T. M. & Carleton, R. A. (1997) Dietary factors in relation to weight change among men and women from two southeastern New England communities. Int. J. Obes. Relat. Metab. Disord. 21:103-109.[Medline]
39. Fogelholm, M., Kujala, U., Kaprio, J. & Sarna, S. (2000) Predictors of weight change in middle-aged and old men. Obes. Res. 8:367-373.[Medline]
40. Voss, S., Kroke, A., Klipstein-Grobusch, K. & Boeing, H. (1998) Is macronutrient composition of dietary intake data affected by underreporting? Results from the EPIC-Potsdam Study. European Prospective Investigation into Cancer and Nutrition. Eur. J. Clin. Nutr. 52:119-126.[Medline]
41. Pomerleau, J., Ostbye, T. & Bright-See, E. (1999) Potential underreporting of energy intake in the Ontario Health Survey and its relationship with nutrient and food intakes. Eur. J. Epidemiol. 15:553-557.[Medline]
42. Niedhammer, I., Bugel, I., Bonenfant, S., Goldberg, M. & Leclerc, A. (2000) Validity of self-reported weight and height in the French GAZEL cohort. Int. J. Obes. Relat. Metab. Disord. 24:1111-1118.[Medline]
43. Kant, A. K., Graubard, B. I., Schatzkin, A. & Ballard-Barbash, R. (1995) Proportion of energy intake from fat and subsequent weight change in the NHANES I Epidemiologic Follow-up Study. Am. J. Clin. Nutr. 61:11-17.
44. Klesges, R. C., Klesges, L. M., Haddock, C. K. & Eck, L. H. (1992) A longitudinal analysis of the impact of dietary intake and physical activity on weight change in adults. Am. J. Clin. Nutr. 55:818-822.
45. Field, A. E., Wing, R. R., Manson, J. E., Spiegelman, D. L. & Willett, W. C. (2001) Relationship of a large weight loss to long-term weight change among young and middle-aged US women. Int. J. Obes. Relat. Metab. Disord. 25:1113-1121.[Medline]
46. Jorgensen, L. M., Sorensen, T. I., Schroll, M. & Larsen, S. (1995) Influence of dietary factors on weight change assessed by multivariate graphical models. Int. J. Obes. Relat. Metab. Disord. 19:909-915.[Medline]
47. Bergmann, K. E. & Mensink, G. B. (1999) [Anthropometric data and obesity]. Gesundheitswesen :S115-S120 61 Spec No:.
This article has been cited by other articles:
|
|
 |
|
  E. Kesse-Guyot, K. Castetbon, C. Estaquio, S. Czernichow, P. Galan, and S. Hercberg Association Between the French Nutritional Guideline-based Score and 6-Year Anthropometric Changes in a French Middle-aged Adult Cohort Am. J. Epidemiol., September 15, 2009; 170(6): 757 - 765. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Buijsse, E. J. Feskens, M. B Schulze, N. G Forouhi, N. J Wareham, S. Sharp, D. Palli, G. Tognon, J. Halkjaer, A. Tjonneland, et al. Fruit and vegetable intakes and subsequent changes in body weight in European populations: results from the project on Diet, Obesity, and Genes (DiOGenes) Am. J. Clinical Nutrition, July 1, 2009; 90(1): 202 - 209. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Tucker and K. S. Thomas Increasing Total Fiber Intake Reduces Risk of Weight and Fat Gains in Women J. Nutr., March 1, 2009; 139(3): 576 - 581. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Nothlings, S. P Murphy, L. R Wilkens, H. Boeing, M. B Schulze, H B. Bueno-de-Mesquita, D. S Michaud, A. Roddam, S. Rohrmann, A. Tjonneland, et al. A food pattern that is predictive of flavonol intake and risk of pancreatic cancer Am. J. Clinical Nutrition, December 1, 2008; 88(6): 1653 - 1662. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. McNaughton, G. D. Mishra, and E. J. Brunner Dietary Patterns, Insulin Resistance, and Incidence of Type 2 Diabetes in the Whitehall II Study Diabetes Care, July 1, 2008; 31(7): 1343 - 1348. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Johnson, A. P Mander, L. R Jones, P. M Emmett, and S. A Jebb Energy-dense, low-fiber, high-fat dietary pattern is associated with increased fatness in childhood Am. J. Clinical Nutrition, April 1, 2008; 87(4): 846 - 854. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Esmaillzadeh and L. Azadbakht Major Dietary Patterns in Relation to General Obesity and Central Adiposity among Iranian Women J. Nutr., February 1, 2008; 138(2): 358 - 363. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E McCann, W. E McCann, C.-C. Hong, J. R Marshall, S. B Edge, M. Trevisan, P. Muti, and J. L Freudenheim Dietary patterns related to glycemic index and load and risk of premenopausal and postmenopausal breast cancer in the Western New York Exposure and Breast Cancer Study Am. J. Clinical Nutrition, August 1, 2007; 86(2): 465 - 471. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Drogan, K. Hoffmann, M. Schulz, M. M. Bergmann, H. Boeing, and C. Weikert A Food Pattern Predicting Prospective Weight Change Is Associated with Risk of Fatal but Not with Nonfatal Cardiovascular Disease J. Nutr., August 1, 2007; 137(8): 1961 - 1967. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. McNaughton, G. D. Mishra, A. M. Stephen, and M. E. J. Wadsworth Dietary Patterns Throughout Adult Life Are Associated with Body Mass Index, Waist Circumference, Blood Pressure, and Red Cell Folate J. Nutr., January 1, 2007; 137(1): 99 - 105. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Mendez, B. M. Popkin, P. Jakszyn, A. Berenguer, M. J. Tormo, M. J. Sanchez, J. R. Quiros, G. Pera, C. Navarro, C. Martinez, et al. Adherence to a Mediterranean Diet Is Associated with Reduced 3-Year Incidence of Obesity J. Nutr., November 1, 2006; 136(11): 2934 - 2938. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Weismayer, J. G. Anderson, and A. Wolk Changes in the Stability of Dietary Patterns in a Study of Middle-Aged Swedish Women J. Nutr., June 1, 2006; 136(6): 1582 - 1587. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
