German Food Pyramid

Learn more information about german food pyramid. In this article we'll discuss german food pyramid.

The German food pyramid was created to promote and communicate healthy food choices.

Methods:

Compliance with the food pyramid recommendations was translated into an index (German Food Pyramid Index (GFPI)) by estimating the ratio of consumed and recommended daily rations of eight food groups, with higher scores showing greater adherence. The GFPI was calculated for 23 531 subjects who participated in the European Prospective Cancer and Nutrition Survey-Potsdam and were appointed between 1994 and 1998. Associations between quintiles of GFPI results and risk of incidental cardiovascular disease (CVD), type 2 diabetes (T2D) and cancer were assessed using proportional regression models of Cox's risk. During the follow-up of 183,740 human years, 363 cases of CVD (myocardial infarction or stroke), 837 cases of T2D and 844 cases of cancer occurred.

Results:

GFPI is inversely related to the risk of CVD in men (multivariate corrected hazard ratio (HR) for the highest versus the lowest quintiles = 0.56; 95% confidence interval (CI): 0.34-0.94), but not in women (HR = 1, 39; 95% CI: 0, 76-2, 55). No association was observed between GFPI and cancer. Feedback between GFPI and T2D (men: HR = 0, 71 (0, 52–0, 97); women: HR = 0, 69 (0, 50–0, 96)) in age-adapted models was significantly enhanced after multivariate adjustments, in particular by body mass index (BMI) (men: HR = 0, 94 (0, 69–1, 30); women: HR = 1, 09 (0, 77–1, 54) ). The same is observed with the overall risk of major chronic disease (CVD, T2D and general cancer).

Conclusion:

Adherence to the German food pyramid guidelines is not associated with a reduced risk of chronic disease when BMI is considered a confederate, with the exception of CVD in men.

Introduction

Diet is considered an important risk factor for chronic diseases, including type 2 diabetes (T2D), cardiovascular disease (CVD), and cancer (WHO, 2003). However, it is still uncertain which aspects of the diet contribute significantly to the onset of disease and have the greatest potential for the prevention of chronic diseases.

Over the last decade, several national dietary recommendations have been supplemented by food-based dietary guidelines and translated into indices that have allowed the actual diet of humans to be compared with these dietary recommendations (Kennedy et al., 1995; Harnack et al., 2002 Shatenstein et al., 2005; Guenther et al., 2007; McNaughton et al., 2008; Taechangam et al., 2008; Woodruff and Hanning, 2010). However, whether compliance with dietary recommendations, as measured by these indicators, is always associated with a lower risk of major chronic diseases. Some studies show that the effects of reducing the risk of diet according to the recommendations are small. The U.S. Department of Agriculture's well-known Healthy Eating Index (HEI), for example, was associated with only a relatively small reduction in the risk of chronic disease (McCullough et al., 2000a, 2000b). In general, indices based on both food and nutrients, such as US-HEI, appear to predict the risk of chronic disease better than indices based only on food groups such as the Health Food Index (Osler et al. al., 2001, 2002), indices based solely on food groups have been less intensively studied (Waijers et al., 2007).

In 2007, the German State Consumer Information Agency (Infodienst Ernährung, Landwirtschaft und Verbraucherschutz) published food recommendations for Germany using a food pyramid (Koelsch and Brüggemann, 2007). It is not known whether adherence to this food pyramid is associated with a lower risk of chronic disease. Thus, we aimed to develop a comprehensive diet quality index to assess compliance with food pyramid recommendations and to examine the relationship of this index to the risk of chronic diseases. We looked at the incidence of CVD, T2D and cancer in a study based on the German population, the European Prospective Cancer and Nutrition Survey (EPIC) - Potsdam.

Subjects and methods

Population survey

The EPIC-Potsdam study is a prospective cohort study of 27,548 participants from the general population of Potsdam, Germany, who were mainly 35–65 years of age in recruitment (1994–98) (Boeing et al., 1999a). Baseline examinations included a self-administered food frequency questionnaire (FFQ), a computer-guided interview on lifestyle habits and medical history, and anthropometric measurements conducted by trained personnel. Every 2-3 years, participants receive a questionnaire sent for a follow-up question to assess incident disease (Bergmann et al., 1999; Boeing et al., 1999b). The response rate for each follow-up round is between 93 and 96%.

We excluded participants who reported common T2D, CVD, or cancer, withheld information about important dietary and lifestyle variables, missed follow-up time, or reported incredibly high or low energy intake (6000 kcal / day; equivalent to about 25,120 kJ respectively). After all exceptions, 23,531 participants remained (9,098 men and 14,433 women).

Dietary assessment

Dietary intake was assessed when recruiting staff with semi-quality FFQ, who was asked about the frequency of intake and the portion size of 148 foods consumed in the previous 12 months. The frequency scale varies from "never" to "five times a day", and portion sizes are estimated using photographs or standardized household measures (one slice, teaspoon or piece). Respondents had to either choose relative portion sizes by assigning the average portion size to half, the same, double, or three times the amount shown, or choose absolute portion sizes from three photos visualizing small, medium, and large portions. food or dishes, The actual portion sizes were obtained from a representative national nutrition survey (Nationale Verzehrsstudie). Additional questions on fat content in dairy products and types of fat used in cooking were included. Daily food and drink intake (g / day) is calculated by multiplying the frequency of the day and the portion size.

Details on the reproducibility and validity of FFQ have been published previously (Boeing et al., 1997; Bohlscheid-Thomas et al., 1997a, 1997b; Kroke et al., 1999). Briefly, reproducibility was assessed by repeated subgroup administration of subjects after 6 months. The correlations of the Spearman test for retesting food intake ranged from 0.49 for bread to 0.89 for alcoholic beverages (mean = 0.70). Spearman's correlations for comparing intake between FFQ and the mean intake of twelve 24-hour reminders performed in a subset of monthly intervals ranged from 0.14 for legumes to 0.90 for alcoholic beverages (mean = 0.45) ( Bohlscheid-Thomas et al., 1997b).

German Food Pyramid Index (GFPI)

GFPI is based on eight of the nine food groups for which the German food pyramid provides recommended intakes (Koelsch and Brüggemann, 2007): soft drinks, vegetables, fruits, cereals (including bread and garnishes such as pasta, rice and potatoes), dairy products (milk, yoghurt and cheese), meat / sausages / fish / eggs, added fat / oil and pastries / snacks. Alcohol is not included in the GFPI, as this nutrient is considered a life factor instead of an essential and recommended component of the human diet. Therefore, alcohol was considered a factor in multivariate analyzes.

Based on the EPIC-Soft program (Slimani et al., 2000), a computerized diet assessment tool developed for 24-hour EPIC reminders, recipes, and mixed meals (e.g., pizza or casserole) were divided into their ingredients and single foods were distributed in the respective food groups.

The actual frequency of portions consumed was calculated by dividing the estimated total food intake obtained by FFQ by the portion size proposed in the German food pyramid (Table 1). The agency reporting the pyramid issued recommendations on the number and size of servings for each of the food groups, provided that the energy intake is about 1900 kcal for women and 2400 kcal for men. This reference energy level is based on the average energy intake of the adult population with a physical activity level (PAL) of 1, 4 (DGE, 2008). In addition, we adjusted the recommended daily portions to individual energy needs by multiplying the initial recommended number of servings by the ratio of energy consumption (in kcal) to 1900 kcal (women) or 2400 kcal (men). The energy requirement of each individual is calculated as a product of the basic metabolism and PAL. Baseline metabolic rate was assessed using gender and age equations using weight and height (Schofield, 1985), and PAL was assessed by questions of activity at baseline (Haftenberger et al., 2002). PAL determines the average energy expenditure during daily activities, sports or professional activities, defined as a multiple of the basic metabolism. To calculate the mean PAL per person, the time spent on the respective activities in hours per day is multiplied by metabolic equivalent values ​​from the literature. One metabolic equivalent is equivalent to an energy expenditure of 1 kcal per kilogram of body weight per hour. using weight and height (Schofield, 1985), and PAL is assessed by questions of activity at baseline (Haftenberger et al., 2002). PAL determines the average energy expenditure during daily activities, sports or professional activities, defined as a multiple of the basic metabolism. To calculate the mean PAL per person, the time spent on the respective activities in hours per day is multiplied by metabolic equivalent values ​​from the literature. One metabolic equivalent is equivalent to an energy expenditure of 1 kcal per kilogram of body weight per hour. using weight and height (Schofield, 1985), and PAL is assessed by questions of activity at baseline (Haftenberger et al., 2002). PAL determines the average energy expenditure during daily activities, sports or professional activities, defined as a multiple of the basic metabolism. To calculate the mean PAL per person, the time spent on the respective activities in hours per day is multiplied by metabolic equivalent values ​​from the literature. One metabolic equivalent is equivalent to an energy expenditure of 1 kcal per kilogram of body weight per hour. To calculate the mean PAL per person, the time spent on the respective activities in hours per day is multiplied by metabolic equivalent values ​​from the literature. One metabolic equivalent is equivalent to an energy expenditure of 1 kcal per kilogram of body weight per hour. To calculate the mean PAL per person, the time spent on the respective activities in hours per day is multiplied by metabolic equivalent values ​​from the literature. One metabolic equivalent is equivalent to an energy expenditure of 1 kcal per kilogram of body weight per hour.

 

The GFPI score was calculated by dividing the portions consumed into the recommended portions and has eight subcomponents with a score from 0 (= non-compliance with the recommended intake) to 10 (= perfect compliance with the recommendations). Less than a perfect match leads to lower results. For the GFPI categories "beverages", "vegetables" and "fruits", the result is calculated using the following equation:

The contribution of 'cereals', 'dairy', 'meat / sausages / fish / eggs' and 'added fats and oils' to the GFPI assessment is also calculated according to equation (1). However, unlike the first three food groups, exceeding the recommendations led to a proportional deduction of points using equation (2). These four food groups have a relatively high energy content and excessive consumption can easily lead to a positive energy balance. For these three food groups we allowed up to 10 additional points according to equation (1) if the intake exceeds the recommendations for each of these three food categories, thus taking into account the potential health benefits of intake, beyond the recommendations. In doing so, it was taken into account

The results of the component (food group) were added to obtain an overall score that could range from 0 to 110 (80 points plus 30 additional points). Thus, higher scores reflect greater adherence to the food pyramid's recommendations. The "pastries and snacks" category is rated back according to equation (2), ie the higher the intake, the lower the the assigned result. A maximum score of 10 is set if the intake is below the recommendation.

Establishment of the case

Information on incidental illnesses was obtained using all available sources during follow-up, including self-reports of new medical diagnosis, medication use, or reasons for reported dietary change in subsequent questionnaires. In the case of multiple diseases, only the first clinical event is considered for the analysis.

All potential accident cases were investigated with information from the treating physician, medical records, and cancer registries (Bergmann et al., 1999). Diseases are coded based on the International Classification of Diseases (ICD-10 I21 for myocardial infarction, I60, I61, I63, I64 for stroke, E11 for diabetes and C00-C97 for cancer (except C44: non-melanoma skin cancer) .

Statistical analysis

For the current analysis, the follow-up time of each participant was extended to the first diagnosis of one of the three chronic diseases (CVD, T2D, or cancer), death, or April 2007. with proportional regression of the Cox hazard, with the age at recruitment and the age at the end of follow-up or diagnosis being time-dependent variables. The lowest quintile of the GFPI score serves as a guide. In multivariate analyzes, we corrected age (years), smoking status (non-smoker, ex-smoker, smoker), alcohol intake (g / day), physical activity in leisure time (h / week), history of high blood lipid levels, or hypertension in the beginning (yes / no), education (vocational training or lower than commercial school, technical school or university degree), vitamin supplementation and total energy intake (kJ / day). In an additional model, HRs were further adjusted for body mass index (BMI) 2  ). The coefficients are defined on a priori knowledge of the main risk factors for the studied results. In principle, the same variables were included in each model, except for a history of hypertension or high blood lipid levels, which were omitted in the multivariate cancer models.

Trend tests were performed by including the mean of each quintile as a continuous variable in the models. In addition, the influence of the components of the single index on incidental chronic diseases was investigated by introducing each component separately in the multivariate models.

Interactions of the result (in quintiles as a continuous variable, coded 1–5) with sex, smoking (never, past, current), age (<60 years versus 60 years) and BMI (35 kg / m  2  ) were assessed . using probability coefficient tests

The P values ​​presented were based on bilateral tests and we applied a significance level of 5% for all tests. All statistical analyzes were performed with SAS 9.2 software (SAS Institute Inc., Cary, NC, USA).

Results

The GFPI score ranged from 23.5 to 92.1 in our study population. The mean GFPI score was 52.7 in women compared to 49.2 in men (P <0 0001). Subjects with higher GFPI scores tend to be slightly older and more educated, have a lower BMI, are less likely to smoke, and are more likely to take vitamin supplements than those with lower scores. (Table 2). Unexpectedly, patients with larger GFPI ulcers are more likely to have a history of hypertension or to report high blood lipid levels.

 

Of the components of the diet that contribute to GFPI, the intake of added fats and oils remains relatively stable in the quintiles of GFPI, while the intake of other food groups changes in the expected direction. The findings were quite similar between men and women, except for meat, sausages, fish and eggs, with men consuming an average of about one serving per day more than women. In addition, alcohol intake decreases through the quintiles in men, but not in women.

The intake of most nutrients increases in quintiles (Table 3). The strongest positive correlations of GFPI and nutrients were found for fiber, folic acid, vitamin C, beta-carotene, iodine and potassium, while the strongest inverse correlations were observed for total fat, saturated fat, monounsaturated fat and vitamin B12 ( data not shown). Energy intake only slightly correlated with GFPI (r = 0, 015).

 

During the total follow-up period of 183,740 human years (mean follow-up period = 7, 8 years), 363 cases of HSS incident, 837 cases of T2D and 844 cases of cancer were observed as first events. When comparing subjects in the highest and those in the lowest GFPI quintile, HR corrections for major chronic disease (CVD, T2D, and cancer in combination) were 0.77 (95% confidence interval (CI): 0.63 –0, 95, P trend = 0, 005) for men and 0, 80 (95% CI: 0, 65–0, 99, P trend = 0, 095) for women. The adjustment for additional risk factors, excluding BMI, attenuated these feedbacks (men: HR = 0, 82; 95% CI: 0, 67–1, 01; P trend = 0, 040; women: HR = 0, 82; 95 % CI: 0, 66–1, 01; P tendency = 0, 074). After further consideration of BMI associations, they disappeared (model 2) (Tables 4 and 5).

  Regarding the specific outcomes of the disease, strong feedback was observed between the assessment of GFPI and the risk of CVD in men but not in women. This significant feedback in men is maintained after controlling for other risk factors, including BMI; men with the highest quintile of GFPI have a 44% lower risk of CVD than those in the lowest quintile (model 2: HR = 0.56; 95% CI: 0.34-0.94; P = 0, 026). The insignificant association in women is independent of postmenopausal status and post-menopausal hormone use.

In men and women, GFPI scores were inversely related to T2D in models adjusted for age, education, lifestyle, and history of dyslipidemia and hypertension. These associations additionally report on BMI reporting.

No consistent trends have been identified for cancer (men: HR = 1, 16; 95% CI: 0, 83–1, 62; P = 0, 402; women: HR = 0, 79; 95% CI: 0, 58– 1.08; P = 0.144).

Finally, we examined the effect of a five-point increase in the individual components of GFPI on the risk of general underlying chronic disease (Table 6). No component was significantly associated with disease risk, with the exception of the pastry and snack component, which was positively associated with the risk for women (HR = 1, 25; 95% CI: 1, 05–1, 48). This means that women with higher than recommended intake of sweets and snacks were at lower risk of underlying chronic disease. Stratification by disease-specific outcome and BMI level revealed that the inverse relationship between pastry intake and disease risk was limited to overweight and obese women and T2D (BMI: 25, 0–29, 9 kg / m  2  : HR = 1, 68, 95% CI: 1, 04–2, 71, BMI: 30–34, 9 kg / m  2 : HR = 2.33; 95% CI: 1.33-4.07).

 

discussion

Our results show that compliance with the recommendations of the German food pyramid in its current form does not have a major impact on disease risk. In our data set, which is the largest prospective study in Germany by number with an extensive dietary assessment, the outcome values ​​are inversely related to the incidence of BPD in men but not in women. Although inverse associations of T2D risk have been observed in both men and women, they are largely explained by lower BMI in those with higher GFPI scores. The risk of cancer is not related to the results of GFPI.

One advantage of GFPI is that adherence to the German food pyramid is expressed in one figure. The one-component scores were calculated in proportion to the intake, instead of using simple breakpoints with scores of 0 and 1. An additional strength of GFPI is the integration of deduction points for accounting for energy-intensive food consumption and additional points for beneficial food groups such as drinks, fruits and vegetables. However, the assessment is based only on food consumption; nutrient intake is not used. In addition, we hypothesized that the individual evaluation components are independent of each other, although they may be related. It is further known that self-administered FFQs have their limitations, especially in terms of quantity (Kristal et al., 2005). Thus, evaluation values ​​should be interpreted as ranking variables and not as reflecting actual food intake. In general, misclassification of food intake can lead to a reduced ability to detect true associations. This may be a limitation of this study, despite the promising design, the high follow-up rate and the inclusion of confirmed cases of accidental disease.

Similar to our findings, studies in the United States show that compliance with US food guidelines, as assessed by the United States Department of Agriculture (USDA) -HEI, is relatively weakly related to the overall risk of chronic disease. Given the single diseases, the strongest feedback was observed in CVD. Similar to our findings, USDA-HEI refers primarily to a lower risk of CVD in men, while this association is less pronounced in women (McCullough et al., 2000a, 2000b). USDA-HEI failed to predict the risk of cancer (McCullough et al., 2000a, 2000b) and these findings were also similar to our study. Although the dietary index was associated with a 15% reduction in the overall risk of cancer (comparing the highest and lowest quintiles), the associations were not constant, after excluding non-dietary factors from the index (Harnack et al., 2002). Thus, it may be that variables other than dietary factors mainly affect the overall risk of cancer.

Reports between compliance with national dietary recommendations and incidental diabetes have been reported rarely. It is noticeable that anthropometric characteristics reduce the risk assessments for T2D in our study. It can be argued that anthropometric characteristics of a person are a consequence rather than the cause of specific eating habits. Analyzes of weight gain determinants, which include this data set, show that ingestion of fruits and vegetables, as well as dietary fiber, is inversely related to weight gain (Buijsse et al., 2009; Du et al., 2010). ). Thus, anthropometric variables can also be considered as intermediate variables, so models that are not adjusted for BMI may also be justified. It's obvious, that not considering BMI as a supporter would lead to reduced risk assessments for T2D and overall chronic diseases in men and women with increased adherence to the food pyramid. At this stage, we chose the more conservative model, including BMI.

When considering the individual components of GFPI, the consumption of pastries and snacks is unexpectedly linked back to the main risk of chronic disease only in obese women. Because women and people who are overweight or obese do not seem to take food into account, especially foods high in fat and sugar (Heitmann and Lissner, 1996; Voss et al., 1997; Yannakoulia et al., 2007). , we cannot exclude that for this association the selective non-reporting is responsible. If obese women at higher risk for chronic diseases are selectively accounted for by eating sweets, false reverse risk relationships can occur. Furthermore, we cannot completely rule out the effect of residual confusion as an additional possible explanation for this observation. However, our findings also do not give grounds to conclude that the consumption of these foods carries a risk of chronic diseases, in particular T2D. More research is needed on this particular issue.

The results of this study cannot be used to suggest that diet has little effect on disease risk. Recently, we could show in another analysis, using this set of data, that a simple nutritional index consisting of fruits, vegetables, whole grains and less meat intake contributes significantly to a healthy lifestyle and is associated with a reduced risk of chronic illnesses. This was particularly the case for T2D, even after BMI control. Estimates of stroke and cancer were also reduced, but not significantly (Ford et al., 2009). The nutritional potential for risk reduction appears to be limited to certain food groups and does not cover all dietary aspects discussed by nutritionists as components of a healthy diet.

 

In summary, this study illustrates the difficulty of finding an appropriate disease prevention strategy through dietary recommendations. In the current food pyramid, food groups with opposite health effects are combined because they are similar in terms of the content of certain nutrients. Further separation of these food groups such as fish and meat, whole grains and refined cereals and oils and other fats may be beneficial. In addition, other food-based dietary guidelines developed for Germany, such as the food circle of the German Nutrition Society, should be studied in a similar way to the German food pyramid.