Joel E. Kimmons, PhD; Heidi Michels Blanck, MS, PhD; Beth Carlton Tohill, MPH, PhD; Jian Zhang, DrPH; Laura Kettel Khan, PhD 

Medscape General Medicine.  2006;8(3) ?2006 Medscape

Abstract

Context: There are scant data on patterns of multivitamin use among US adults in terms of body mass index (BMI) or whether one is trying to lose weight.
Objective: To examine multivitamin use and beliefs about multivitamin use among adults according to BMI and to determine whether use by body weight differs if one is trying to lose weight.
Design: Cross-sectional multivariate analysis of the HealthStyles consumer survey. The final analytic sample consisted of 2239 women and 1532 men.
Main Outcome Measures: Prevalence and odds of multivitamin use by demographic and behavioral characteristics including BMI, use by weight loss intent, and among users, reasons for use.
Results: 63.7% of women and 52.9% of men reported multivitamin use (taking 1 or more multivitamin per week). Obese women were less likely than normal-weight women to use multivitamins; no differences according to BMI category were detected for men. Among women who were not trying to lose weight, obese women were less likely than normal-weight women to use multivitamins (odds ratio = 0.63, CI 0.41-0.98). Assessment of reasons for use found that compared among women not trying to lose weight, those trying to lose weight were more likely to report multivitamin use because "It is important for my health."
Conclusions: This descriptive analysis adds to the limited literature on multivitamin use according to both body weight and attempting to lose weight. Multivitamin use was common and decreased with increasing BMI. This may be because fewer obese people consider vitamins "important for their health."

---

Readers are encouraged to respond to George Lundberg, MD, Editor of MedGenMed, for the editor's eyes only or for possible publication via email: glundberg@medscape.net

Introduction

Approximately half of US women and one third of men report they are currently trying to lose weight. Among these individuals, 50% report they are eating fewer calories as a weight loss strategy.^[1] The nutritional quality of a reduced-calorie diet may be adequate if nutrient-rich low-energy-dense fruits, vegetables, and grain products replace high-energy-dense foods.^[2] However, people who limit their caloric intake and do not eat a variety of nutrient rich, low-energy dense foods may have inadequate micronutrient intakes. Dietary assessment of US adults who intentionally consumed energy-restricted diets were found be lower in vitamin E, calcium, iron, zinc, and selenium.^[3-5] In contrast, Kant^[6] demonstrated that there was little evidence of increased risk of low nutritional biomarkers in US adults who reported trying to lose weight.

Despite a lack of consistent evidence showing low nutritional status or intake, many popular diet books and Web sites suggest that dieters should take a vitamin/mineral supplement to fill in where his/her diet might be lacking.^[7-9] A popular consumer health site states that, "Many people don't receive all of the nutrients they need from their diet because they either can't or don't eat enough, or they can't or don't eat a variety of healthy foods. For some people, including those on restrictive diets, multivitamin-mineral supplements can provide vitamins and minerals that their diets often don't."^[10]

There are scant data on patterns of multivitamin use among adults in the United States who are trying to lose weight. A Gallup survey conducted in 2004 for the March of Dimes (n = 2012) found that 44% of women aged 18-45 years who had dieted in the past 6 months consumed a vitamin containing folic acid, primarily in the form of a multivitamin.^[11] Although dieters are often advised to take multivitamins, their motivations to do so have not been formally assessed. To describe multivitamin use among US adults, particularly among those trying to lose weight, we used the 2003 HealthStyles survey to examine multivitamin use by demographic and behavioral characteristics and among people currently trying to lose weight, according to BMI category. We further considered reasons for use among people currently trying to lose weight, according to BMI category.

Methods

This study used data from the 2003 Styles cross-sectional adult marketing surveys (ConsumerStyles, HealthStyles). The Synovate, Inc. consumer mail panel consists of approximately 600,000 people who are recruited through a 4-page survey. Respondents are given small gifts (eg, a 20-minute calling card, entry in a sweepstake) in return for their participation. Stratified random sampling of this pool was used to generate a list of 10,000 potential ConsumerStyles adult respondents (≥ 18 years). This sample was then stratified (or balanced) on region, household income, population density, age, and household size to create a nationally representative sample. Adequate representation of low income/minority groups was ensured by oversampling. The response rate for ConsumerStyles was 59%, yielding a sample size of 5873. Conducted since 1995, HealthStyles is a ConsumerStyles subsurvey of health attitudes, behaviors, conditions, and knowledge. ConsumerStyles respondents were sent the HealthStyles survey, with a 69% (4035 participants) response rate. Respondents may refuse to answer questions at any time and no personal identifiers are included in the data file.

Variable Definitions

"Multivitamin user" was defined as those who reported taking a multivitamin 1 or more times per week in response to the question, "How many times a week do you take a multivitamin?" Respondents were asked to check all applicable responses to the statement, "I take multivitamins or try to take multivitamins because..." Possible responses were "Taking vitamins is important for my health," "I've made taking a vitamin part of my daily routine," "I don't always eat a balanced diet," "I feel vitamins give me more energy," and "My healthcare provider recommends vitamins.'

Intention to lose weight was determined by the question, "Are you currently trying to lose weight?" (Yes, No). BMI (weight in kilograms divided by height in meters squared) was based on self-reported height and weight. BMI category was defined according to National Heart, Lung, and Blood Institute guidelines: normal 18.5-24.9, overweight 25.0-29.9, and obese ≥ 30.0.^[12]

Ethnic groups were self-defined as non-Hispanic black, non-Hispanic white, Hispanic, or other. Age groups were 18-34, 35-54, and 55+ years. Education was grouped into high school or less, some college, and graduated college.

Physical activity was categorized as meets physical activity recommendations (30 minutes ≥ 5 times per week), irregular activity (30 minutes < 5 times per week), and inactive, in accordance with Centers for Disease Control and Prevention guidelines.^[13] The questions used to determine this included (1) during a usual week in the past month, how many days did you do moderate or vigorous physical activities such as walking, biking, running, aerobics, yard work, or anything else that causes increases in breathing or heart rate for at least 10 minutes? and (2) on the days when you exercised or did physical activities for at least 10 minutes, about how many total minutes each day did you typically spend doing these activities?

Participants were excluded if they were pregnant (n = 32), had a missing BMI (n = 167), or had a BMI less than 18.5 (n = 44). The final analytic sample consisted of 3771 subjects (2239 women and 1532 men).

Statistical Methods

Unadjusted prevalence estimates and multivariate adjusted odds ratios of multivitamin use are presented. Multivitamin use was examined in relation to trying to lose weight and BMI category in multivariate models that included age, race/ethnicity, education level, and physical activity. Estimates and models were stratified by sex, to permit sex-specific interpretations, because previous research indicates that men and women use different weight-control practices.^[1] The HealthStyles survey is weighted based on the US census figures for age, sex, ethnicity/race, income, and household size. SUDAAN software was used to account for weighting (SAS-callable version, 9.0, Research Triangle Park, NC). Statistical tests were determined to be statistically significant at P = .05. The data were weighted to adjust the sample distribution of gender, age, income, and race matched to that of the general population according to the US census. Weighting adjusts for over- or under-representation of categories within these demographic variables.

Results

More than half of women (63.7%) and men (52.9%) reported taking multivitamins 1 or more times per week. Women had a higher odds of use than men (odds ratio [OR] = 1.6, confidence interval [CI] 1.4-1.9). Less than half of all respondents, 47.4% of women and 40.1% of men, reported taking multivitamins 7 or more times per week. Among multivitamin users, the overall mean use was 6.3 times per week (women 6.3/wk, men 6.2/wk). The percentage of respondents who were currently trying to lose weight was 58.0% (women 65.7%, men 50.1%). Among those trying to lose weight, 21.2% were normal-weight, 40.7% were overweight, and 38.1% were obese. Among those not trying to lose weight, 51.9% were normal-weight, 29.5% were overweight, and 18.6% were obese.

In women, the odds of multivitamin use were higher among those ≥ 35 years than among 18- to 34-year-old women and higher among high school graduates than less educated women. Odds of use were higher among non-Hispanic white women than non-Hispanic black women and higher in women who reported at least some physical activity than among inactive women. In men, the odds of multivitamin use were higher among those ≥ 35 years than 18- to 34-year-old men and among non-Hispanic white men than non-Hispanic black men, but did not differ by education or physical activity ( Table 1 ). Normal-weight women had higher odds of use than obese women, but no differences by BMI category were detected for men ( Table 1 ).

There was not an association between BMI and vitamin use among women or men trying to lose weight. However, among women who were not trying to lose weight, obese women had a lower odds of multivitamin use than women of normal BMI (OR = 0.63, CI 0.41-0.98) ( Table 2 ). Among men, BMI category was not associated with multivitamin use by those trying or not trying to lose weight ( Table 2 ).

Table 3 and Table 4 describe the reasons for multivitamin use examined in relation to BMI category and weight loss intention, respectively. With increasing BMI category among women, the prevalence and odds decreased for reported multivitamin use because it is "important for their health," "part of their daily routine," or "gives them more energy" ( Table 3 ). Among men, reasons did not differ by BMI category. Among women trying to lose weight, there was a higher odds of responding that they took a multivitamin because it was "important for their health" than for those not trying to lose weight (women OR 1.74) ( Table 4 ). Other reasons for multivitamin use did not differ by intent to lose weight.

Discussion

In this analysis of a 2003 survey of American adults, almost two thirds of women and about half of men reported taking at least 1 multivitamin weekly. No difference in multivitamin use was found between men or women trying to lose weight and those not trying to lose weight. Furthermore, regardless of weight loss intent, the prevalence of multivitamin use decreased with increasing BMI category, with the exception of overweight men who reported currently trying to lose weight. This latter group had the highest prevalence of multivitamin use among men.

The proportion of this sample who reported taking a multivitamin is similar to most reports in the literature from various surveys. The 2001 Behavioral Risk Factor Surveillance System (BRFSS), a telephone survey of adults ≥ 18 years of age, found 56.5% reported daily use of a vitamin or other supplement, with significantly more women using supplements (62%).^[14] Additionally, those who regularly take supplements use them frequently; 83% of supplement users reported in the BRFSS taking 1-2 vitamin or supplement pills per day.^[14] The INTERMAP study, conducted between 1997 and 1999, which examined Americans aged 40 to 59 years via in-person interview, found 56% of women and 46% of men used a daily vitamin or other supplement.^[15] The National Health and Nutrition Examination Survey (1999-2000) assessed multivitamin use during the in-person interview and found 57% of women and 47% of men reported taking any supplement and 38% of women and 32% of men reported taking a multivitamin in the previous month.^[16]

The characteristics of multivitamin users reported here are similar to those described in other reports.^[15-18] Higher multivitamin use was found among older women with non-Hispanic white ethnicity, lower BMI, higher education, and greater physical activity. Among men, we only detected lower odds of multivitamin use among the younger and black non-Hispanic groups; however, the direction of the changing prevalence was similar to women for education and physical activity. Foote and colleagues^[18] also stratified these characteristics of multivitamin use by sex and reported results for both men and women similar to those of the women in our study.^[18] However, they used a much larger sample size (n = 100,196). We speculate that although multivitamin use was greater in women than men, the general patterns of use may be similar for both sexes. However, because of this study's smaller sample size, differences in multivitamin use among men by some demographic characteristics may not have been detected.

No difference in reported multivitamin use was detected between adults trying to lose weight and those not trying to lose weight. This is not consistent with a March of Dimes study that reported female dieters to be nearly 30% more likely to be taking a vitamin with folic acid than non-dieters.^[11] Also, Archer and colleagues^[15] found that people reporting consumption of special diets were more likely to use dietary supplements (vitamins, minerals, supplements, tonics). However, the March of Dimes study only examined women aged 18-45 years, and the questions used to determine weight loss behavior asked specifically about dieting. The Archer study assessed people aged 40-59 years, asked about general dietary supplements (vitamins, minerals, supplements, and tonics) and consumption of special diets, and did not distinguish if they were for weight loss. In contrast, this study examined a wider age range of men and women, asked the broader question, "Are you trying to lose weight?," specifically asked about multivitamin use, and adjusted for a number of behavioral and demographic characteristics that the March of Dimes analysis did not.

Limitations of this study include the cross-sectional nature of the HealthStyles survey, which prevents inferences on causal relationships. Also, HealthStyles uses volunteers from a panel survey. Research conducted to validate findings between studies based on paneling techniques and traditional health-research sampling have found similar prevalences for common health behaviors.^[19] Although selection of participants from survey panels, as in the Styles survey, is common in market research, it is less common in health research. Therefore, further comparison with other population-based surveys is needed. Other limitations include the low response rate among panel participants (59% for the initial study, 69% of which responded to HealthStyles survey) and small numbers of minority participants.

Limited research has examined the reasons leading to and maintaining multivitamin use, which include a variety of social, psychological, knowledge, and economic factors. Conner and colleagues^[20] reported that a major predictor of the decision to use supplements and, in turn, the motives for multivitamin use, may be formed from the behavioral beliefs underlying attitudes. One such behavioral belief is that despite the lack of clear scientific evidence supporting the benefits of multivitamins, users believe they are important for their health.^[20] In this study, this explanation of multivitamin use was a more likely response for women who were trying to lose weight than for those who were not. This was the only reason for multivitamin use that differed significantly by intent to lose weight, suggesting that for those attempting weight loss, one rationale for multivitamin use is importance for health.

In this study, multivitamin use was common and decreased with increasing body weight. What is unknown is why heavier people are less likely to use multivitamins. Our study points to the possibility that fewer obese people consider multivitamins "important for their health." In contrast, for those who are trying to lose weight, or in other words, potentially interested in their weight or health, multivitamin use is not more common. Considering the high prevalence of overweight in the US population and the number of people attempting to lose weight, more research is needed to understand whether BMI affects nutrient requirements and whether certain dieters are at increased risk of low nutritional status and thus could benefit from multivitamin use.