ORIGINAL ARTICLE
Inge Tetens1*, Anja Biltoft-Jensen1, Camilla Spagner1,2, Tue Christensen1, Maj-Britt Gille1, Susanne Bügel2 and Lone Banke Rasmussen1
1Division of Nutrition, National Food Institute, Technical University of Denmark, Denmark; 2Department of Human Nutrition, Faculty of Life Sciences, Centre for Advanced Food Studies, University of Copenhagen, Frederiksberg, Denmark
Abstract
Objectives: To evaluate the intake of micronutrients from the diet and from supplements in users and non-users of dietary supplements, respectively, in a representative sample of the Danish adult population. A specific objective was to identify the determinants of supplement use.
Design: A cross-sectional representative national study of the intake of vitamins and minerals from the diet and from dietary supplements.
Method: The Danish National Survey of Dietary Habits and Physical Activity, 2000–2004. Participants (n=4,479; 53% females) aged 18–75 years gave information about the use of dietary supplements in a personal interview. The quantification of the micronutrient contribution from supplements was estimated from a generic supplement constructed from data on household purchases. Nutrient intakes from the diet were obtained from a self-administered 7-day pre-coded dietary record. Median intakes of total nutrients from the diets of users and non-users of supplements were analysed using the Wilcoxon rank-sum test.
Results: Sixty percent of females and 51% of males were users of supplements. With the exception of vitamin D, the intake of micronutrients from the diet was adequate at the group level for all age and gender groups. Among females in the age group 18–49 years, the micronutrient intake from the diet was significantly higher compared with the non-users of dietary supplements. The use of dietary supplements increased with age and with ‘intention to eat healthy.’
Conclusion: Intake of micronutrients from the diet alone was considered adequate for both users and non-users of dietary supplements. Younger females who were supplement users had a more micronutrient-dense diet compared to non-users.
Keywords:
vitamins;
minerals;
food supplements;
adults
(Received: 4 April 2011; Revised: 4 July 2011; Accepted: 18 July 2011; Published: 7 September 2011)
Citation: Food & Nutrition Research 2011, 55: 7153 - DOI: 10.3402/fnr.v55i0.7153
Food & Nutrition Research 2011. © 2011 Inge Tetens et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
A varied diet that fulfils the food-based dietary guidelines will cover the needs for most vitamins and minerals in the population (1), with the exception of vitamin D, iodine, folate, and iron. Studies in several European countries have shown that the pattern of dietary supplement use differs considerably among the countries and that Finland and Denmark had the highest prevalence of supplement users (2, 3). In Denmark, the use of supplements is common practice and earlier studies have reported that between 39 and 60% of adults were users of dietary supplements (4–8).
In view of the relatively high prevalence of supplement use, dietary supplements may contribute considerably to the daily nutrient intake. It has been reported that nutrient intakes from dietary supplements vary considerably among individuals, and that variation in composition of the supplements may lead to significant differences in nutrient intakes among users of supplements (9, 10). Some studies have suggested that the use of supplements is often associated with certain lifestyle factors and especially that users of dietary supplements have healthier lifestyle than non-users. This paradox – referred to as the ‘inverse supplement hypothesis’– indicates that those taking dietary supplements have higher estimated vitamin and mineral intakes from food sources alone than non-users of supplements (11).
The methodologies for reporting the use of dietary supplements and for estimating the micronutrient contribution from these supplements vary between studies (2, 12). In some countries nutrient intakes from supplements are calculated by using tables on micronutrient contents of specific supplements or from label information obtained during a personal interview. The reference period may also differ. There is a need to develop improved methods for estimating intakes of micronutrients from food supplements (2).
Little is known about the actual contribution of dietary supplements to total micronutrient intake in the general adult Danish population. The objective of this study was therefore to evaluate the intake of micronutrients in users and non-users of dietary supplements in a representative sample of the Danish adult population; and furthermore, to identify the determinants of supplement use.
The data for the present study are derived from The Danish National Survey of Dietary Habits and Physical Activity, which is a nation-wide representative cross-sectional survey. The method is described elsewhere (13–16). The dietary data were collected between 2000 and 2004. Information on diet was obtained from a pre-coded 7-day dietary record using household measures and validated pictures to estimate portion sizes. Completed dietary records were checked and intakes of nutrients from the diet (including iodine from mandatory fortification) were calculated using the General Intake Estimation System Programme (GIES; The National Food Institute, Denmark). To assess the nutrient intake adequacy at group level, the estimated average requirement (AR) was used; and where no values were available, the recommended intake (RI) was used (1). The upper level (UL) values were used to assess the probability of adverse effects occurring at some specified level of long-term exposure (1).
In a personal interview, the subjects provided information about their social background including education and marital status, details about their dietary practices and attitudes, smoking habits, and other lifestyle variables including physical activity in leisure time and present physical activity level and intention to eat healthy. The interview included questions about the subjects’ usual intake of dietary supplements. Subjects were classified as supplement users if they reported that they habitually consumed at least one vitamin, mineral, or other supplement containing a vitamin and/or mineral, irrespective of the frequency of use. Brand names were not asked about. If the subject reported being a supplement user, he/she was asked about the type of supplement, the number, and frequency.
Assignments of nutritive values for supplements, as defined in the EU (17), were based on purchased supplements reported by GfK ConsumerScan, which registers the brand names of all supplements. The GfK household panel is comprised of a randomly selected number of geographically representative households that register all purchases made by the household for a whole week.
The GfK data survey includes data from July 1999 to June 2000, covering a whole year in order to account for seasonal variation in consumption patterns. The nutrient contents of the supplements were obtained from the Danish Veterinary and Food Administration Nutrition Database. Product information not included in the database was obtained from health stores, pharmacies, supermarkets, and information search on the Internet. Further product information was obtained directly from manufacturers of dietary supplements and from their homepages.
From the supplements that are targeted for adults and contain vitamins and/or minerals, generic dietary supplement products were formulated containing vitamins and minerals weighed according to dose and purchases of the different supplements. These generic supplements were used to estimate micronutrient intakes from supplements. Several assumptions were made concerning the type of nutrients in the supplement. Vitamin A in the dietary supplements was assumed to be preformed vitamin A (i.e. retinol), where one ‘retinol activity equivalent’ (RAE) equals 1 µg of supplemental preformed vitamin A. For niacin it was assumed that the value from all sources (niacin, nicotinic acid, and nicotinamide) could be added, and folate and folic acid were used interchangeably. Each supplement reported in the personal interview was coded with a generic dietary supplement.
All data were registered and processed for each individual in the sample. As a graphic examination of the nutrient intake showed that the data distribution deviated grossly from a bell-shaped normal distribution, a Kolmogorov-Smirnov Test was carried out to compare two experimentally determined distributions comparing the experimental distribution against a hypothetical ideal. The distributions of the micronutrients were skewed and tests of non-normality data were used. The median vitamin and mineral intake with 5th and 95th percentiles, with and without supplemental intake was calculated for two different age groups (18–49 years and 50–75 years) of men and women. The median intakes, exclusively from the diets of users and non-users of supplements, were examined and compared using the Wilcoxon rank-sum test.
To test the hypothesis that supplement users would have a healthier lifestyle profile than non-users, the association between socio-demographic and lifestyle factors and the use of dietary supplements was analysed using a multiple logistic regression. Socio-demographic and lifestyle factors for inclusion in the logistic regression model were selected on the basis that the dependent variable was for the ‘use of dietary supplements’ (yes/no). The continuous variable age was grouped and included in the model. Data on education level, marital status, healthy eating, leisure time activity level, physical activity level, self-perceived physical form, smoking, and health status were derived from the background interview. The number of individuals included in the analysis was 4,479. Of these, 3,338 observations were used since 1,141 observations had to be deleted due to missing values for the dependent or explanatory variables. Stepwise backward logistic regression analyses were used and a reduced model was calculated including only the variables from the full model with P-values less than 0.05. The results in Table 1 are presented as OR and 95% CI. A P-value of less than 0.05 was considered to be significant in all statistical tests. All analysis was carried out in SAS Enterprise Guide version 3.0 or 4.0 (SAS Institute Inc., Cary, NC, USA).
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Among the 4,479 subjects with complete registered-details, 60% of the females and 51% of the males were classified as users of supplements containing vitamins and/or minerals (Table 1). The categorical variables: education levels, marital status, leisure time activity level, physical activity level, self-perceived physical form, smoking, and health status showed no significant association with the use of dietary supplements (results not shown). On the contrary, gender, age group, and intention to eat healthy all showed significant associations with use of supplements (Table 1). Subjects in the age group 50–75 years were significantly more likely to use supplements in comparison to subjects in the age group 18–49 years (results not shown). Dividing the age groups further into subgroups, the results showed that subjects in the age group 64–75 years were significantly more likely to use supplements compared to all younger age groups (Table 1). The subjects who reported that they often or most often had the intention to eat healthy were significantly more likely to be supplement users compared with subjects that sometimes or never had the intention to eat healthy.
The intake of vitamins and minerals derived exclusively from the diet for users and non-users of dietary supplements is shown for men and women separately and for the younger (18–49 years) and older age group (50–75 years; Tables 2–5). In contrast to the older women and all men, women in the age group 18–49 years who reported to be supplement users had a significantly higher intake of micronutrients from the diet compared with non-supplement users in the same age group (Table 3) except for vitamin C, niacin, potassium, and sodium. Men in both age groups, 18–49 years and 50–75 years, who reported to be supplement users had significantly higher intakes of vitamin C, folate, and calcium from diet compared with non-supplement-users (Tables 2 and 4). The older women in the age group (50–75 years) who reported to be supplement users had a significantly higher intake of calcium compared with non-supplement-users (Table 5).
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Among all men and women, the intake of vitamins and minerals from the diet was considered adequate with median intakes above the average requirement (AR; Tables 2–5). As no AR has been established for vitamin D, calcium, and potassium, the median intakes were compared with the recommended intake (RI). Intake of vitamin D from the diet was below the RI for both men and women in all age groups.
For all adults, supplement use resulted in considerably higher intake of all micronutrients compared with intake of non-supplement users. For vitamin D, supplement users achieved a median intake slightly above the RI of 7.5 µg/d for 18–60 years old (Tables 2 and 3) but below the RI of 10 µg/d for 61 and above (Tables 4 and 5). Users of supplements among both sexes had an intake of vitamin A and zinc with P95 values close to or slightly above the UL values of 3,000 RE/d and 25 mg/d, respectively (Tables 2–5).
Our results show that in the general adult Danish population more than half were categorised as supplement users. Users and non-users of dietary supplements had similar intakes of micronutrients from the diet with the exception of the women in the reproductive age between 18–49 years of age. In this group the users of supplements had a significantly higher intake of most but not all micronutrients and, thus, a more micronutrient-dense diet compared to non-users. For this age group, our results are consistent with those from previous studies showing that supplement users consume more nutrient dense diets than non-users (18–21). In our study, both users and non-users of dietary supplements appear to obtain adequate amounts of vitamins and minerals from the diet when compared with the estimated average requirements. However, both users and non-users of supplements had shortfalls in vitamin D intake from diet alone and supplementation was thus most beneficial in improving adequacy of this particular vitamin.
In the present study we estimated the contribution to micronutrient intake from dietary supplements by using a novel approach where we generated generic supplement products based on purchase data adjusted to the relevant time period. One of the advantages of this method is that the subjects do not have to recall the name(s) of the supplement brand that they take; it can therefore be applied to other dietary assessment methods (i.e. recalls, food frequency, etc.). We acknowledge that a coupling of an exact supplement product would have been optimal. An American study (9) reported that nutrient intakes from multivitamins with high variability among individuals and variation in composition can lead to considerable differences among users in terms of nutrient intakes from supplements. A disadvantage of using the present approach might be that subjects could be misclassified as having a low or moderate intake, although they actually have higher intakes. Also, the non-vitamin/mineral supplements excluded from the present study may contain small doses of vitamins or minerals, thus perhaps underestimating the micronutrient intake.
In the present study, total intake of selected micronutrients among the users of dietary supplements of vitamins E and D was increased fourfold and, for iron, selenium, and zinc two- to threefold compared with non-users. The 5th percentiles among the users illustrate that several users, also in the lower range, markedly elevated the total intakes.
A common problem in the assessment of diet in surveys is the under-reporting of total food intake indicated by reports of low energy intake. Part of the present study population was included in the Danish nationwide dietary survey 2000/2001, where 24% of adults (18 to 75 years) were classified as under-reporters (13). The effects of under-reporting on the estimated intake of micronutrients may depend on the food items under-reported. In a recent study carried out in a similar study population of 138 healthy volunteers between the ages 20 and 59 years, we found that the under-reporting was biased with a significant under-reporting of energy from added sugar and over-reporting of energy from protein among the under-reporters (16). It is noteworthy that the dietary survey method and food database used in the above studies were the same, eliminating possible biases from these sources. A German national survey conducted from 1997 to 1999 examined the impact of under-reporting on the results. When the under-reporters were excluded from the analyses, a 13% higher vitamin intake was found among the population of both users and non-users of supplements (19). Results of a more recent study from the United States suggest that energy consumption was under-reported by both men and women, but that protein consumption was not under-reported by women and only slightly by men (21). It is not clear, however, how under-reporting of consumption of selected foods can affect the estimated vitamin and mineral intake. Under-reporting of food energy does not necessarily lead to under-reporting of specific micronutrients if the food omitted from the response is not rich in the specific micronutrient.
The present study identified the females in the age group 50–75 years with intention to eat healthy as the most frequent supplement user in the Danish population. Previous studies from Denmark (4, 7), Sweden (22, 23), and Germany (20) reported similar results. Likewise, studies from other European countries report that supplement use increased with age 25.
To conclude, the present study confirms that use of dietary supplements is common behaviour in the general Danish adult population, in spite of the fact that the intake of micronutrients for both users and non-users of supplements is considered adequate from the diet alone. Use of supplements is higher among women than among men and increases with age and the intention to eat healthy. Female users of dietary supplements between the ages 18 and 49 years have a significantly higher intake of micronutrients from their diet than non-users, whereas no differences were seen in men or in elderly women.
The authors would like to acknowledge senior researcher Margit Groth, senior advisor Sisse Fagt, and scientist Jeppe Matthiessen who, together with the author ABJ, were responsible for the collection of data. CS and ABJ were responsible for analyses of data. IT and CS wrote the first draft and all authors participated in the discussion of the results and revision of the manuscript.
The authors have not received any funding or benefits from industry or elsewhere to conduct this study
*Inge Tetens
Head of Division of Nutrition
National Food Institute
Technical University of Denmark
Mørkhøj Bygade 19
DK 286 Søborg, Denmark
Tel: +45 35887400
Fax: +45 35887401
Email: intet@food.dtu.dk