mineral supplements: Rationale and safety

Accepted Manuscript Multivitamin/mineral supplements: rationale and safety Hans K. Biesalski, Jana Tinz PII:

S0899-9007(16)30092-2

DOI:

10.1016/j.nut.2016.06.003

Reference:

NUT 9788

To appear in:

Nutrition

Received Date: 17 February 2016 Revised Date:

12 May 2016

Accepted Date: 3 June 2016

Please cite this article as: Biesalski HK, Tinz J, Multivitamin/mineral supplements: rationale and safety, Nutrition (2016), doi: 10.1016/j.nut.2016.06.003. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Title: Multivitamin/mineral supplements: rationale and safety

Corresponding author: Hans K. Biesalski, Prof. Dr. University of Hohenheim Institute of Biological Chemistry and Nutrition (140)

Phone: +49 711 459-24112 Fax: +49 711 459-23822

M AN U

D-70593 Stuttgart, Germany

SC

RI PT

Running title: Multivitamin/mineral supplements: rationale and safety

TE D

List of all authors (first name, middle initial, last name) and their institutional affiliation: Hans K. Biesalski, University of Hohenheim; Jana Tinz, University of Hohenheim

EP

List of all authors' last names exactly as they should appear for PubMed

AC C

indexing: Biesalski HK; Tinz J

Word count for the entire manuscript (title through references (includes abstract, text, legends, acknowledgments, and references): CURRENT COUNT = 6,819

Keywords: multivitamins, minerals, supplements, safety, long-term use, randomized controlled studies (RCTs)

Number of tables: 1 table + 1 supplemental table

ACCEPTED MANUSCRIPT List of all sources of financial support: Editorial support was provided by Peloton Advantage and funded by Pfizer.

Conflict of interest:

AC C

EP

TE D

M AN U

SC

manuscript. Jana Tinz no conflicts of interest.

RI PT

Prof. Dr. Biesalski received an honorarium from Pfizer for the development of this

ACCEPTED MANUSCRIPT Abbreviation footnote

2

Abbreviations used: AE, adverse effect; AI, Adequate Intake; CHD, coronary heart

3

disease; CVD, cardiovascular disease; DRI, Dietary Reference Intake; EAR,

4

Estimated Average Requirement; IFA, folic acid+iron; MVM, multivitamin/mineral(s),

5

multivitamin/mineral supplement; NHANES, National Health and Nutrition

6

Examination Survey; NIH, National Institutes of Health; PHS II, Physicians’ Health

7

Study II; RCT, randomized, controlled trial; RDA, Recommended Dietary Allowance;

8

UL, tolerable upper intake level.

AC C

EP

TE D

M AN U

SC

RI PT

1

3

ACCEPTED MANUSCRIPT 9

Abstract Multivitamin/mineral supplements (MVMs) are widely used in many populations. In

11

particular, in pregnant women, MVMs together with iron and folic acid are

12

recommended to improve birth outcome and reduce low birth weight and rates of

13

miscarriage. However, MVM use is common in the general population, as well. The

14

objective of the review was to evaluate the safety of long-term use of these

15

supplements. To examine the safety of MVM use we performed a literature search

16

for randomized, controlled studies that supplemented with a combination of at least 9

17

vitamins and 3 minerals at a maximum concentration of 100% of the Recommended

18

Dietary Allowance (RDA). We found nine studies evaluating the use and efficacy of

19

MVMs in pregnant women and healthy adults and six studies in the elderly where

20

adverse effects (AE) were explicitly addressed. Only minor AEs (e.g., unspecific

21

gastrointestinal symptoms) were reported in all studies. In particular, there were no

22

significant differences between treatment and placebo groups. MVM use within the

23

range of the Dietary Reference Intake (DRI) will not result in excess intake, even

24

when including the impact of food and fortified food, and does not increase mortality.

25

Taken together, these findings indicate that MVMs can be safe for long-term use

26

(more than 10 years).

AC C

EP

TE D

M AN U

SC

RI PT

10

4

ACCEPTED MANUSCRIPT Introduction

28

There is an increasing number of individuals taking multivitamin/minerals (MVMs) to

29

maintain good health and to be protected from different diseases (e.g.,

30

cardiovascular disease [CVD], cancer, cognitive decline). The US Department of

31

Agriculture/Department of Health and Human Services 2010 Dietary Guidelines for

32

Americans acknowledge that "supplements containing combinations of certain

33

nutrients may be beneficial in reducing the risks of some chronic diseases when

34

used by special populations," yet also state that excessive use of certain

35

supplements has the potential to be harmful [1]. This raises the question of MVM

36

safety.

37

M AN U

SC

RI PT

27

Definition of an MVM supplement

39

According to a National Institutes of Health (NIH) State of the Science Panel, an

40

"MVM refers to any supplement containing 3 or more vitamins and minerals but no

41

herbs, hormones or drugs, with each component at a dose less than the tolerable

42

upper level determined by the Food and Nutrition Board—the maximum daily intake

43

likely to pose no risk for adverse health effects" [2]. In an NIH fact sheet regarding

44

MVMs [3] a more differentiated approach defines groups of MVMs as follows:

45 46 47

•

AC C

EP

TE D

38

Many MVMs are taken once daily and contain all or most recognized vitamins

and minerals at levels close to Daily Values, Recommended Dietary

48

Allowances (RDAs), or Adequate Intakes (AIs). Basic formulations are for

49

broad-spectrum use. Formulations for special populations such as children,

50

pregnant women, and seniors provide the same vitamins and minerals in

51

amounts tailored to those populations’ specific needs.

5

ACCEPTED MANUSCRIPT 52

•

Other MVMs contain vitamins and minerals at levels substantially higher than

53

the recommended values and may also include other nutritional and herbal

54

ingredients. These are sometimes packaged in multiple-pill packs to be taken

55

each day.

57

•

RI PT

56

Specialized MVMs, used, for example, to enhance performance or improve immune function, or for weight control, are often composed of vitamins and

59

minerals in combination with herbal or specialty ingredients such as

60

coenzyme Q10, probiotics, and glucosamine. These may also include

61

nutrients at levels substantially above recommended levels.

M AN U

SC

58

62

Using the NIH State of the Science Panel and the Agency for Healthcare Research

64

and Quality definition of a multivitamin as containing 3 or more vitamins and/or

65

minerals [4], the supplements listed in the second and third bullets above are indeed

66

MVMs.

67

There is indeed a strong difference regarding safety issues if MVMs with

68

concentrations at or below the RDA are compared with MVMs having doses near the

69

UL or with one or more components that exceed the RDA. The major risk, if any, of

70

MVMs at or below RDA may be an insufficient supply of one or more micronutrients

71

that might not cover the individual need in case of a higher demand or inadequate

72

supply through nutrition. MVMs with concentrations above the RDA might exert AEs,

73

particularly with long-term use. We evaluated only safety aspects in the healthy

74

population with dosages within the RDA because if an MVM shows a beneficial

75

effect, it could be the result of a compensation for an inadequate intake (below RDA)

76

of one or more micronutrients.

AC C

EP

TE D

63

6

ACCEPTED MANUSCRIPT 77 Limitations in setting dietary recommendations

79

A major problem in understanding the effect of micronutrients is that the body´s

80

physiological need is not fully understood across all organs and systems. With few

81

exceptions, micronutrients are needed in the cells and tissues and might undergo a

82

specific metabolism within the cells. Blood is the delivery route for nearly all nutrients

83

to cells and tissues. Therefore the blood may not reflect real adequacy or

84

inadequacy. Typically, a primary indicator (biomarker or blood concentration)

85

selected for setting dietary recommendations may not reflect the need to maintain

86

adequate function across all cells and tissues. Bruce Ames’ proposed “triage theory”

87

shows that, as a consequence of a higher need or low intake, micronutrients are

88

delivered from tissues to other organs depending on their need, and priority is given

89

to survival [7]. Also, the traditional targets, the blood and urine (and sometimes hair),

90

do not always give the most useful information, even though they are often the most

91

accessible biological samples. Indeed, there is a need to develop more sophisticated

92

techniques to assess nutrient status. Russell [8] summarized the main pitfalls in

93

setting recommended dietary intake levels as follows:

95

•

AC C

94

EP

TE D

M AN U

SC

RI PT

78

There is a lack of uniform criteria for selection of the endpoints used for

judging adequacy or toxicity; some of the endpoints chosen are disease

96

endpoints, some are biochemical, some are factorial.

97 98 99 100

•

The extrapolations to children are often not valid but used due to the fact that there is such a dearth of studies on children. The DRI numbers (Estimated

7

ACCEPTED MANUSCRIPT 101

Average Requirement [EAR]; RDA, and AI) and their uses are not well

102

understood and are debated by the experts even now.

103 104

•

Regarding the extrapolations to children, the same applies to the extrapolation to pregnancy. For these groups, more or less rough calculations are made

106

because scientific data to derive RDAs are often not available and/or robust.

107

Further, there may be genetic, health, and lifestyle differences that drive

108

varied requirements for individuals in addition to age and gender differences.

SC

RI PT

105

109

Dietary recommendations may not exactly represent the individual need. While the

111

present RDAs will continue to be revised as new information becomes available,

112

they represent the most current scientific data that are relevant to the population.

113

RDAs in Europe, the United States, and Canada are set using very similar

114

approaches and sets of data. It is estimated that an intake within the range of two

115

standard deviations above the EAR will not only prevent a deficiency but also cover

116

differences within the individual need. An intake below the EAR, however, is

117

inadequate and places the individual at risk to develop a real deficiency.

TE D

EP

AC C

118

M AN U

110

119

Micronutrient inadequacy - the micronutrient gap

120

A micronutrient gap may be defined as an intake below the EAR. What does this

121

mean? Those with a micronutrient intake below the EAR might develop clinical signs

122

and symptoms of deficiency depending on the duration of the gap and individual

123

need. Clinical signs of deficiency are the "endpoint" of a deficiency disease. Before

124

symptoms occur, hidden hunger exists, which means that traditional deficiency

125

symptoms are missing, but health problems may occur in either the short or long

8

ACCEPTED MANUSCRIPT term. Depending on the diet and how often dietary changes occur, the gaps can be

127

more or less severe and last for a few days to even weeks. Gaps of one or more

128

micronutrients might fluctuate over time, appearing or disappearing. Depending on

129

the half-life, this fluctuation can be harmless or critical. Nevertheless, due to a lack of

130

adequate analytical methods, it is in most cases impossible to detect and clearly

131

define the magnitude of a micronutrient gap. Consequently, a questionnaire might be

132

worthwhile to help elucidate these details.

133

A rough estimation regarding micronutrient gaps can be taken from the NHANES

134

data. A recently published analysis evaluated the contribution of MVM supplements

135

to the overall supply with micronutrients during 2007–2010 [9]. The results showed

136

that 51% of Americans consumed multivitamins containing more than nine

137

micronutrients (vitamins/minerals) in a dose equivalent to 100% of the RDA. From

138

the NHANES data, the authors calculated the percentage of individuals aged > 4

139

years with total nutrient intakes from food and MVM supplements falling below the

140

EAR as follows: vitamin D 74%, vitamin E 67%, magnesium 46%, calcium 39%,

141

vitamin A 35%, and vitamin C 31%. Smaller proportions of the population had

142

intakes below the EAR for (in decreasing prevalence) zinc, folic acid, vitamin B6,

143

iron, copper, thiamin, vitamin B12, vitamin B2, phosphorus, niacin, and selenium.

144

A recent study using data from national surveys of 13 European countries to

145

evaluate micronutrient gaps documented a high prevalence of insufficient intake of

146

selected nutrients [10]. Depending on the country, the prevalence of vitamin C intake

147

below the EAR was 8-40%, vitamin D, 47-100%, vitamin B12, 0-40%, folic acid, 10-

148

91%, calcium, 0-61%, iron (males only), 0-18%, zinc, 1-31%, and selenium, 8-47% in

149

adults aged 19-64 years. In the population over 64 years of age there was a similar

AC C

EP

TE D

M AN U

SC

RI PT

126

9

ACCEPTED MANUSCRIPT prevalence, with the exception of vitamin D (33-100%) and calcium (48-100%).

151

Similar data are available from the United States [9].

152

These data and others [11-13] show that micronutrient gaps indeed exist in the

153

general population. In specific risk groups (e.g., the elderly; low income; low

154

education level; one-sided diet, including people with dietary obesity treatment as

155

well as bariatric surgery; and self-selected diet) these gaps might be even larger.

156

A low intake of micronutrients might have long-term health consequences and may

157

contribute to the risk for chronic diseases. This aspect has been addressed recently

158

in a review paper by Angelo et al [14] showing that MVM supplements exert

159

beneficial effects in a number of diseases.

M AN U

SC

RI PT

150

160 Pregnancy

162

A major group that bears a risk of inadequacy in the short term are pregnant women.

163

Pregnancy and early childhood are vulnerable periods for an inadequate supply of

164

micronutrients. There is increasing knowledge that an inadequate supply of certain

165

micronutrients may exert negative effects on birth weight and time of delivery. An

166

extensive review of micronutrient intakes in women living in resource-poor settings

167

revealed that pregnant and non-pregnant, non-lactating women were at the risk of

168

inadequate intakes for most of the micronutrients included in the review (vitamins A,

169

C, B6, and B12, thiamin, riboflavin, niacin, folic acid, iron, and zinc) [15]. Regional

170

differences in the extent of a risk for an inadequate supply of vitamins A, C, and B6

171

and riboflavin were apparent, with women in Latin America at lower risk than women

172

in Asia.

173

Data depicted in Table 1 show that iron, iodine, vitamin A, and zinc deficiencies in

174

pregnant women and children below the age of five are not only present in low-

AC C

EP

TE D

161

10

ACCEPTED MANUSCRIPT income countries, but also in Europe to an extent, which is not a peripheral matter

176

with respect to childhood development [16]. In some cases, the prevalence of a

177

deficiency is similar between low- and high-income countries. A deficiency might

178

result in different consequences of various severities; however, its cause is always

179

the same. We need to understand that poor diet quality and poor diet diversity are

180

some of the major reasons for micronutrient inadequacies. Iron deficiency or vitamin

181

A inadequacy signal a diet that is poor in food containing iron or vitamin A, or even

182

both (e.g., meat, liver, eggs). A detected micronutrient deficiency, thus, can be taken

183

as a biomarker for an inadequate diet that might result in more deficiencies than in

184

iron or vitamin A alone. Consequently, increasing diet diversity will close the known

185

and unknown micronutrient gaps.

186

As reported in a recent Cochrane review [17], supplementation with MVMs in young

187

females from low- to middle-income countries prior to or at an early time point in

188

pregnancy results in a significant decrease of newborns with low birth weight (small-

189

for-gestational-age) or preterm delivery. The review analyzed 21 trials (involving

190

75,785 women), and when compared with iron and folate supplementation only,

191

MVM supplementation resulted in a statistically significant decrease in the number of

192

low birth weight babies (RR: 0.89; 95% CI: 0.83-0.94) and small-for-gestational-age

193

babies (RR: 0.87; 95% CI: 0.81–0.95). Regarding safety, the authors claimed that

194

"There is insufficient evidence to identify adverse effects and to say that excess

195

multiple-micronutrient supplementation during pregnancy is harmful to the mother or

196

the fetus."

AC C

EP

TE D

M AN U

SC

RI PT

175

197 198

Safety

11

ACCEPTED MANUSCRIPT In high-income countries in Europe and the United States, multivitamins are the most

200

commonly used vitamin supplements. In particular, the increasing numbers of

201

healthy elderly are the major consumers of multivitamins to improve or maintain their

202

health. However, concerns are raised that multivitamins might create harmful effects,

203

and their long-term safety is in question.

204

The European Food Safety Authority [18] assesses the safety of supplement use

205

based on the risk of exceeding an established upper limit, defined as follows: •

Tolerable upper intake level (UL): the maximum level of total chronic daily

SC

206

RI PT

199

intake of a nutrient (from all sources) judged to be unlikely to pose a risk of

208

adverse health effects to humans.

M AN U

207

Nutrients possess some characteristics that distinguish them from other food

210

chemicals for the purpose of risk assessment. Nutrients are essential for human

211

well-being within a certain range of intakes, and there is a long history of safe

212

consumption of nutrients at the levels found in balanced human diets. Additionally,

213

for some nutrients there may be experience of widespread chronic consumption

214

(e.g., from dietary supplements) at levels significantly above those obtained from

215

endogenous nutrients in foods without reported adverse effects. Data on adverse

216

effects of nutrients are also often available from studies in humans, which helps to

217

reduce uncertainty factors. Furthermore, many nutrients are subject to homeostatic

218

regulation of body content through adaptation of absorptive, excretory, or metabolic

219

processes, and this provides a measure of protection against exposures above usual

220

intakes from balanced diets [18].

AC C

EP

TE D

209

221 222

Literature Search and Study Selection

12

ACCEPTED MANUSCRIPT To uncover studies reporting safety issues related to MVMs, we reviewed different

224

databases (i.e., Medline, Embase, Cochrane Database of Systematic Reviews), as

225

well as the Internet and different library catalogues and document servers of the

226

University of Hohenheim. We searched mainly for articles published since 2010,

227

because we used the new USDA dietary guidelines (2010) as the basis for

228

recommendations and RDA. MVMs commonly used were within a “physiological”

229

range, i.e., a range that can be achieved via nutrition and that does not exceed the

230

RDA.

231

In order to exclude AEs that might be related to existing diseases, we only evaluated

232

trials that used MVM treatment for primary prevention in at least one study arm. Not

233

all trials documented AEs as recommended by the Consolidated Standards of

234

Reporting Trials (CONSORT) statement for non-pharmacological treatment [19,20].

235

Some of the trials selected for review addressed the question of AEs; others did not.

236

Trials listed used MVMs for primary prevention in either pregnant women or the

237

elderly. Both groups are vulnerable and sensitive; therefore, safety aspects are of

238

great importance, and any harm will be more visible than in younger or non-pregnant

239

healthy individuals. To exclude specific formulations or doses of vitamins or minerals

240

near the tolerable upper intake level (UL), we decided to focus our review only on

241

MVMs as defined in the first bullet. This is in accordance with the definitions of

242

MVMs per the National Health and Nutrition Examination Survey (NHANES) III [5]

243

(10 vitamins+calcium and iron) and the National Health Interview Survey (10

244

vitamins+5 minerals) [6].

AC C

EP

TE D

M AN U

SC

RI PT

223

245 246

Safety and pregnancy

13

ACCEPTED MANUSCRIPT The rationale for MVM supplementation during pregnancy is the co-existence of

248

multiple micronutrient deficiencies in pregnant women, in particular in resource-poor

249

settings [15]. Resources, however, are not only a question of availability but also of

250

education and knowledge.

251

Trials of MVMs in pregnant women for primary prevention are carried out to improve

252

either maternal health or birth outcome. Most trials did not specifically address AEs.

253

Nevertheless, differences in birth outcome can be taken as harmful or as AEs. In a

254

double-blind randomized, controlled trial (RCT) [21] the effect of MVM vs. folic acid

255

(400 µg) alone vs. folic acid+iron (IFA) (30 mg) was tested in 18,775 nulliparous

256

pregnant women. Prenatal MVM and IFA did not affect perinatal mortality (primary

257

outcome) or preterm delivery, birth weight, birth length, or gestational duration

258

(secondary outcomes). In a double-blind cluster-randomized trial, MVM (n = 15,804)

259

or IFA (30 mg/400 µg) (n = 15,486) supplementation were used to study the effect on

260

early infant mortality (primary outcome) or neonatal mortality, fetal loss, or low birth

261

weight (secondary outcomes) [22]. Compared with IFA supplementation, MVM

262

significantly reduced early infant mortality by 18%. In addition, all secondary

263

outcomes were significantly improved in the MVM group. Specific AEs were not

264

reported. In different studies from the same group [23-25], the effect of MVM on

265

pregnancy outcome was evaluated. MVM decreased the incidence of neural tube

266

defects and other cleft formations. AEs were not reported. MVM vs. IFA (60 mg/400

267

µg) during pregnancy showed a significant increase in birth weight in the MVM group

268

[26]. AEs were not reported.

269

Studies explicitly documenting AEs or addressing specific safety aspects are

270

summarized in Supplemental Table 1A. A Danish cohort study [27] documented

271

that regular use of MVM around the time of conception was associated with a

AC C

EP

TE D

M AN U

SC

RI PT

247

14

ACCEPTED MANUSCRIPT decreased risk for preterm birth and small-for-gestational-age birth. Any AEs were

273

not reported, but the authors concluded that "multivitamin use around the time of

274

conception could be a safe and simple strategy.” In a RCT with either MVM (n = 600)

275

or IFA (n = 600) from 12 weeks’ gestation until delivery [28], morbidity during

276

pregnancy was taken as an AE measure. Typical antenatal problems frequently

277

occurring during pregnancy, such as nausea, dyspepsia, and abdominal pain,

278

occurred in both groups without any significant difference. MVM supplementation

279

was associated with increased birth weight when compared with IFA

280

supplementation. In this study dealing with females from Nepal, nutrition status was

281

not assessed, but might not have been optimal. In another study from Nepal [29],

282

undernourished women were selected for MVM supplementation (n = 99) or placebo

283

(n = 101). Both groups received IFA (60 mg/500 µg). AEs of supplementation

284

(nausea, vomiting, diarrhea, abdominal pain, and anorexia) were documented in 7

285

subjects receiving MVM and in 13 in the IFA control group. In a randomized double-

286

blind RCT, the effect of MVM use (n = 55) vs. placebo (n = 59) on subjective health

287

and well-being was elucidated [30]. Those who supplemented with MVM

288

experienced an increased energy level (P = 0.022) and enhanced mood (P = 0.027),

289

both significant vs. placebo. One participant in the MVM group reported minor

290

gastrointestinal symptoms (nausea).

291

Taken together, in controlled clinical trials with MVM supplementation before and

292

during pregnancy, AEs were not reported. The safety and improvement of pregnancy

293

outcomes justify the recommendation to supplement MVM from early conception

294

until delivery. The Genetics Committee of the Society of Obstetricians and

295

Gynaecologists of Canada made clear recommendations: Women in the

296

reproductive age group should be advised of the benefits of folic acid in addition to a

AC C

EP

TE D

M AN U

SC

RI PT

272

15

ACCEPTED MANUSCRIPT multivitamin supplement during wellness visits, especially if pregnancy is

298

contemplated [31]. This recommendation is based on an extensive review of articles

299

published between 1985 and 2007 and related to MVM supplementation and its

300

impact on birth defects. Supplementation with MVMs including iron and folate

301

significantly reduced congenital anomalies (anencephaly, myelomeningocele,

302

meningocele, oral facial cleft, structural heart disease, limb defect, urinary tract

303

anomaly, and hydrocephalus).

304

SC

RI PT

297

Safety for healthy adults and elderly

306

Few clinical trials exist that explicitly address the question of AEs as a possible

307

consequence of MVM supplementation (Supplemental Table 1B). In the largest

308

randomized clinical trial with adult male physicians (Physicians’ Health Study [PHS]

309

II; mean age 64 years) [32] the authors assessed a number of potential AEs of daily

310

multivitamin use and found no significant effect on gastrointestinal tract symptoms

311

(peptic ulcer, constipation, gastritis, and nausea), fatigue, drowsiness, skin

312

discoloration, or migraine. Minor AEs such as rashes were more likely in men taking

313

MVMs (2125 men in the multivitamin group vs. 2002 men in the placebo group;

314

hazard ratio: 1.07; 95% CI: 1.01– 1.14; P = 0.03), and the MVM cohort showed an

315

increase in epistaxis (1579 men in the multivitamin group and 1451 men in the

316

placebo group; hazard ratio: 1.10; 95% CI: 1.02–1.18; P = 0.01). This study, with the

317

longest time interval for supplementation with MVMs, clearly demonstrates that any

318

serious or long-lasting side effects did not occur as a result of MVM

319

supplementation. The trial was carried out with healthy males, and the design

320

represents the highest level of evidence. PHS II addressed additional outcomes

321

(cognition, CVD, cancer), and the results of these trials have been reported

AC C

EP

TE D

M AN U

305

16

ACCEPTED MANUSCRIPT elsewhere. AEs were not further addressed because these have been described in

323

the original publication [32].

324

In a smaller RCT, MVMs were supplied for one year in a group of healthy elderly and

325

compared with placebo [33]. AEs (headache, insomnia, gout) were described in 28

326

of 456 participants in the MVM group and in 37 of 454 in the placebo group.

327

Treatment side effects were addressed in a trial combining MVM with herbal

328

supplementation [34]. One subject experienced nausea and vomiting after

329

commencing MVM supplementation. One subject from the placebo group developed

330

a mild rash. Some trials evaluated the effect of MVM supplementation in an

331

undernourished elderly population [35], while others dealt with combinations of

332

MVMs and herbals [36]. AEs were not reported. A primary prevention trial [37] of the

333

impact of MVM supplementation on infections in an elderly population reported 2

334

cases of unspecified digestive problems in the MVM group (n = 110) and 1 case of

335

unspecific digestive problems in the placebo group (n = 108).

TE D

M AN U

SC

RI PT

322

336

Historical safety concerns

338

Concerns are frequently raised that the combination of MVMs with food, including

339

fortified foods, might result in adverse effects (AEs) and could cause long-term users

340

to exceed ULs. In an analysis of the NHANES data for 16,444 participants, the use

341

of MVM supplements reduced the percentage of the population falling below the

342

EAR but did not result in excess intake [9]. In MVM supplement users ≥4 years of

343

age the prevalence of those exceeding the UL was 1.7% for retinol and 2.5% for folic

344

acid.

345

In analyses of data from the Hawaii-Los Angeles Multiethnic Cohort (n = 215,823

346

adults), the addition of the intake from MVMs to that from food significantly

AC C

EP

337

17

ACCEPTED MANUSCRIPT increased the average adequacy of 17 nutrients (men: 84+/-19%; women: 83+/-20%;

348

P < 0.0001 vs. intake from food only) [38]. The addition of nutrients from MVMs

349

resulted in undesirably high intakes in >10% of the participants. Intakes from vitamin

350

A, niacin, and folate exceeded the UL; however, the UL for vitamin A refers to

351

preformed retinol and not to retinol equivalents, including provitamin A from food

352

intake. Consequently, this resulted in an overestimation of vitamin A intake, and

353

according to the authors, the excess of niacin and folate also reflects a discordance

354

in units: both of these ULs only refer to an intake from fortification and

355

supplementation.

356

A further safety concern related to vitamins and vitamin combinations comes from

357

studies estimating the impact of a long-term use of single vitamins or combinations

358

of more than two vitamins, in particular antioxidants, on mortality as a clinical

359

endpoint. Increased mortality has been described in recent meta-analyses [39], but

360

all studies were carried out with either single antioxidant vitamins or combinations in

361

higher doses, some of them exceeding the UL, and not with MVMs. Clinical trials

362

using MVMs and an additional single vitamin supplement (e.g., vitamin E) in a high

363

dose indeed have shown that it is the high-dose supplement and not the MVM that

364

exerts AEs. In one trial [40] the effect of MVM or high-dose vitamin E (200 mg) on

365

the frequency of respiratory tract infections was evaluated in individuals aged 60

366

years or older (mean age 73 years). Subjects (n = 652) were assigned to four

367

treatment groups (MVM alone, vitamin E alone, MVM+vitamin E, and placebo).

368

Vitamin E alone and MVM+vitamin E, but not MVM alone, resulted in an increased

369

incidence of infections. These results clearly show that a combination of MVM with a

370

high-dose vitamin might exert AEs, but that the MVM alone would not. This is

371

supported by the 6-year follow-up data of the Linxian (China) trial of MVM

AC C

EP

TE D

M AN U

SC

RI PT

347

18

ACCEPTED MANUSCRIPT supplementation in a group of undernourished individuals aged 40-69 years [41] that

373

revealed no increase in all-cause mortality.

374

One of the greatest studies to evaluate the effects of vitamin supplements on long-

375

term health and overall mortality was the Iowa Women's Health Study [42]. In 38,772

376

women (mean age 61.6 years) supplement use (single vitamins, minerals, and

377

MVMs) was self-reported in 1986, 1997, and 2004 without information regarding the

378

vitamin/mineral concentration. In contrast to a single vitamin use, MVM did not show

379

any increased mortality risk. The age and energy adjusted hazard rates were 0.98

380

(1986–1996), 0.91 (1997–2003) and 0.83 (2004–2008). These data are confirmed by

381

a recent analysis of the NHANES III data in which an association was found between

382

MVM intake for >3–20 years and reduced CV mortality in females (hazard ratio: 0.56

383

users vs. non-users). The effect was less pronounced in males (hazard ratio: 0.79

384

users vs. non-users) [43].

385

Meta analyses, as the gold standard of evidence, have not shown any harm, in

386

particular, for higher mortality from either cancer or CVD, in MVM users. The

387

systematic review for the US Preventive Service Task Force [4] evaluated the health

388

effects of MVM supplementation in four RCTs and one cohort study. There was no

389

significant effect on mortality, but a reduced mortality in the supplemented group

390

(without significance).

391

A recent meta-analysis [44] evaluated the effects of MVMs on all-cause mortality.

392

Across all pooled studies, no significant effect of MVM supplementation on all-cause

393

mortality was observed (RR: 0.98; 95% CI: 0.94–1.02). In 13 of 21 primary

394

prevention trials, there was a trend toward mortality risk reduction in the MVM

395

groups. No significant effect of MVMs on all-cause mortality was found in the

396

secondary prevention of CVD. This meta-analysis provides the highest level of

AC C

EP

TE D

M AN U

SC

RI PT

372

19

ACCEPTED MANUSCRIPT 397

evidence that MVMs are also safe regarding the strongest clinical endpoint,

398

mortality. Indeed, the studies enumerated in Supplemental Table 1B did not show

399

any impact on mortality in the elderly.

400 MVM supplements for a healthy population?

402

MVM supplements are safe but according to the PHS II without a real benefit. That

403

raises the question: who should supplement and why? Generally spoken

404

micronutrients can compensate inadequate supply or transient micronutrient gaps.

405

The majority of the participants of PHS II celebrated a healthy life style (only few

406

smokers, most with a normal weight, a high vegetable intake and regular exercise)

407

and a supplement seems not really necessary. However, in cases of sudden disease

408

or periods of inadequate dietary diversity a supplement reduces the risk for

409

micronutrient gaps. Whereas a diet with poor quality and an unhealthy lifestyle can

410

never be compensated by any supplement.

411

Pregnant and lactating women experience a higher demand for most micronutrients,

412

not only iron and folate as usually mentioned. Based on recent meta-analyses

413

supplementation with a MVM is superior over folate and iron alone with respect to a

414

lower incidence of malformations and small for gestational age newborns [45]. In

415

particular young females practicing a vegetarian or vegan diet should be advised to

416

take a MVM in case of a planned pregnancy.

417

Special groups with an inadequate intake are elderly with a general undernutrition (<

418

1.600 kcal/day) which occurs more frequently with increasing age (> 75).

419

Undernutrition however, is always accompanied by malnutrition. Because it is not

420

really possible to determine the degree of micronutrient inadequacy for each

421

micronutrient a MVM supplement is more meaningful than a single vitamin. Elderly

AC C

EP

TE D

M AN U

SC

RI PT

401

20

ACCEPTED MANUSCRIPT are also frequently suffering from vitamin B12 and vitamin D deficiency.

423

Consequently, these micronutrients should be added separately because the

424

concentration in usual MVM supplements is too low to fulfill the demand. A

425

recommendation to take a MVM supplement for elderly is also a preventive

426

approach. The possibility to become hospitalized increases with age. An existing

427

micronutrient gap might be a harmful basis.

428

There are a couple of disease-related micronutrient deficiencies which are not

429

discussed within this paper. There is only one group which has been treated to

430

become healthy but might develop different micronutrient deficiencies which at least

431

are unhealthy thus questioning the initial approach: obese individuals undergoing

432

bariatric surgery. Depending on the time after surgery and the method used, different

433

deficiencies have been described [46, 47]. Deficiencies which might have existed

434

before the surgical intervention and will exacerbate [48,49,50]. Whether or not

435

morbid obese people should be recommended to take a MVM prior surgery is a

436

matter of debate. However following bariatric surgery a MVM supplement should be

437

strongly recommended.

EP

438

TE D

M AN U

SC

RI PT

422

Conclusions

440

Despite a balanced and overall healthy diet, micronutrient gaps may occur from time

441

to time. Any impact on health depends on the gap’s duration and severity. Individual

442

needs differ based on health, lifestyle, genetics, and other factors. The precise

443

individual need is difficult to measure and requires further investigation to improve

444

methods of assessment. Nevertheless, MVM supplement consumption has been

445

shown to reduce dietary intake gaps and to improve measures of nutritional status

AC C

439

21

ACCEPTED MANUSCRIPT without exceeding the DRI or UL. Overcoming these nutritional gaps can help to

447

counteract potential health issues caused by inadequacy.

448

A small number of MVM intervention studies have been published that haven’t

449

reported safety issues and AEs. However, in studies reporting AEs, none have

450

revealed any serious effects of MVM supplementation. MVMs are safe at

451

physiological doses (100% DRI) in the short and the long term, whereas AEs may

452

occur if single vitamins at high doses are consumed.

453

Because the individual need may be greater in cases of different chronic or acute

454

diseases, more or less severe micronutrient gaps may occur from time to time. An

455

MVM can help to improve the nutrient supply and overcome problems of inadequacy

456

without concern for its long-term safety.

AC C

EP

TE D

M AN U

SC

RI PT

446

22

ACCEPTED MANUSCRIPT Acknowledgments

458

Editorial support was provided by Peloton Advantage and funded by Pfizer. All

459

authors have read and approved the final manuscript.

AC C

EP

TE D

M AN U

SC

RI PT

457

23

ACCEPTED MANUSCRIPT 460

Literature Cited [1] Dietary Guidelines for Americans. 2010, [www.dietaryguidelines.gov].

462

[2] National Institutes of Health. National Institutes of Health State-of-the-Science

463

Conference Statement: multivitamin/mineral supplements and chronic disease

464

prevention. Am J Clin Nutr 2007;85:257S-264S.

467

[https://ods.od.nih.gov/factsheets/MVMS-HealthProfessional/].

SC

466

[3] Multivitamin/Mineral Supplements: Fact Sheet for Health Professionals,

[4] Fortmann SP, Burda BU, Senger CA, Lin JS, Beil TL, O'Connor E et al.

M AN U

465

RI PT

461

Vitamin, mineral, and multivitamin supplements for the primary prevention of

469

cardiovascular disease and cancer: a systematic evidence review for the U.S.

470

Preventive Services Task Force. Evidence Report No. 108. AHRQ Publication

471

No. 14-05199-EF-1. Rockville, MD: Agency for Healthcare Research and

472

Quality; 2013.

473

TE D

468

[5] Radimer K, Bindewald B, Hughes J, Ervin B, Swanson C, Picciano MF. Dietary supplement use by US adults: data from the National Health and

475

Nutrition Examination Survey, 1999-2000. Am J Epidemiol 2004;160:339-349.

476

[6] Moss AJ, Levy AS, Kim I, Park YK. Use of vitamin and mineral supplements in

477

the United States: current users, types of products, and nutrients. Advance

479

480

AC C

478

EP

474

data from vital and health statistics; no 174. Hyattsville, MD: National Center for Health Statistics; 1989.

[7| Ames BN. Low micronutrient intake may accelerate the degenerative

481

diseases of aging through allocation of scarce micronutrients by triage. Proc

482

Natl Acad Sci U S A 2006;103:17589-17594.

24

ACCEPTED MANUSCRIPT 483 484

[8] Russell RM. Setting dietary intake levels: problems and pitfalls. Novartis Found Symp 2007;282:29-36. [9] Wallace TC, McBurney M, Fulgoni VL, III. Multivitamin/mineral supplement

486

contribution to micronutrient intakes in the United States, 2007-2010. J Am

487

Coll Nutr 2014;33:94-102.

488

RI PT

485

[10] Vinas BR, Barba LR, Ngo J, Gurinovic M, Novakovic R, Cavelaars A, de Groot LC, van't Veer P, Matthys C, Majem LS. Projected prevalence of

490

inadequate nutrient intakes in Europe. Ann Nutr Metab 2011;59:84-95.

491

[11] National Nutrition Survey II, [http://www.was-esse-ich.de/index.php?id=44].

492

[12] Bailey RL, Fulgoni VL, III, Keast DR, Lentino CV, Dwyer JT. Do dietary

M AN U

SC

489

supplements improve micronutrient sufficiency in children and adolescents? J

494

Pediatr 2012;161:837-842.

TE D

493

[13] Gallagher CM, Black LJ, Oddy WH. Micronutrient intakes from food and

496

supplements in Australian adolescents. Nutrients 2014;6:342-354.

498 499 500

501

[14] Angelo G, Drake VJ, Frei B. Efficacy of multivitamin/mineral supplementation to reduce chronic disease risk: a critical review of the evidence from

AC C

497

EP

495

observational studies and randomized controlled trials. Crit Rev Food Sci Nutr 2015;55:1968-1991.

[15] Torheim LE, Ferguson EL, Penrose K, Arimond M. Women in resource-poor

502

settings are at risk of inadequate intakes of multiple micronutrients. J Nutr

503

2010;140:2051S-2058S.

25

ACCEPTED MANUSCRIPT [16] Black RE, Victora CG, Walker SP, Bhutta ZA, Christian P, de OM, Ezzati M,

505

Grantham-McGregor S, Katz J, Martorell R, Uauy R. Maternal and child

506

undernutrition and overweight in low-income and middle-income countries.

507

Lancet 2013;382:427-451.

510 511

512

during pregnancy. Cochrane Database Syst Rev 2012;11:CD004905. [18] Tolerable upper intake levels for vitamins and minerals. Parma, Italy: European Food Safety Authority; 2006.

SC

509

[17] Haider BA, Bhutta ZA. Multiple-micronutrient supplementation for women

M AN U

508

RI PT

504

[19] Boutron I, Moher D, Altman DG, Schulz KF, Ravaud P. Extending the

513

CONSORT statement to randomized trials of nonpharmacologic treatment:

514

explanation and elaboration. Ann Intern Med 2008;148:295-309. [20] Ioannidis JP, Evans SJ, Gotzsche PC, O'Neill RT, Altman DG, Schulz K,

TE D

515 516

Moher D. Better reporting of harms in randomized trials: an extension of the

517

CONSORT statement. Ann Intern Med 2004;141:781-788. [21] Liu JM, Mei Z, Ye R, Serdula MK, Ren A, Cogswell ME. Micronutrient

519

supplementation and pregnancy outcomes: double-blind randomized

AC C

520

EP

518

controlled trial in China. JAMA Intern Med 2013;173:276-282.

521

[22] Shankar AH, Jahari AB, Sebayang SK, Aditiawarman, Apriatni M, Harefa B,

522

Muadz H, Soesbandoro SD, Tjiong R, Fachry A, Shankar AV, Atmarita,

523

Prihatini S, Sofia G. Effect of maternal multiple micronutrient supplementation

524

on fetal loss and infant death in Indonesia: a double-blind cluster-randomised

525

trial. Lancet 2008;371:215-227.

26

ACCEPTED MANUSCRIPT 526

[23] Czeizel AE, Dudas I: Prevention of the first occurrence of neural-tube defects

527

by periconceptional vitamin supplementation. N Engl J Med 1992;327:1832-

528

1835.

530

531

[24] Czeizel AE. Prevention of congenital abnormalities by periconceptional

RI PT

529

multivitamin supplementation. BMJ 1993;306:1645-1648.

[25] Czeizel AE. Reduction of urinary tract and cardiovascular defects by periconceptional multivitamin supplementation. Am J Med Genet

533

1996;62:179-183.

M AN U

534

SC

532

[26] Catov JM, Bodnar LM, Olsen J, Olsen S, Nohr EA. Periconceptional

535

multivitamin use and risk of preterm or small-for-gestational-age births in the

536

Danish National Birth Cohort. Am J Clin Nutr 2011;94:906-912. [27] Osrin D, Vaidya A, Shrestha Y, Baniya RB, Manandhar DS, Adhikari RK,

TE D

537

Filteau S, Tomkins A, Costello AM. Effects of antenatal multiple micronutrient

539

supplementation on birthweight and gestational duration in Nepal: double-

540

blind, randomised controlled trial. Lancet 2005;365:955-962.

541

[28] Gupta P, Ray M, Dua T, Radhakrishnan G, Kumar R, Sachdev HP.

543 544

545

AC C

542

EP

538

Multimicronutrient supplementation for undernourished pregnant women and the birth size of their offspring: a double-blind, randomized, placebo-controlled trial. Arch Pediatr Adolesc Med 2007;161:58-64.

[29] Zeng L, Dibley MJ, Cheng Y, Dang S, Chang S, Kong L, Yan H. Impact of

546

micronutrient supplementation during pregnancy on birth weight, duration of

547

gestation, and perinatal mortality in rural western China: double blind cluster

548

randomised controlled trial. BMJ 2008;337:a2001.

27

ACCEPTED MANUSCRIPT [30] Sarris J, Cox KH, Camfield DA, Scholey A, Stough C, Fogg E, Kras M, White

550

DJ, Sali A, Pipingas A. Participant experiences from chronic administration of

551

a multivitamin versus placebo on subjective health and wellbeing: a double-

552

blind qualitative analysis of a randomised controlled trial. Nutr J 2012;11:110.

553

RI PT

549

[31] Wilson RD, Johnson JA, Wyatt P, Allen V, Gagnon A, Langlois S, Blight C, Audibert F, Desilets V, Brock JA, Koren G, Goh YI, Nguyen P, Kapur B. Pre-

555

conceptional vitamin/folic acid supplementation 2007: the use of folic acid in

556

combination with a multivitamin supplement for the prevention of neural tube

557

defects and other congenital anomalies. J Obstet Gynaecol Can

558

2007;29:1003-1026.

M AN U

559

SC

554

[32] Gaziano JM, Sesso HD, Christen WG, Bubes V, Smith JP, MacFadyen J, Schvartz M, Manson JE, Glynn RJ, Buring JE. Multivitamins in the prevention

561

of cancer in men. The Physicians' Health Study II Randomized Controlled

562

Trial. JAMA 2012;308:1871-1880.

TE D

560

[33] Avenell A, Campbell MK, Cook JA, Hannaford PC, Kilonzo MM, McNeill G,

564

Milne AC, Ramsay CR, Seymour DG, Stephen AI, Vale LD. Effect of

565

multivitamin and multimineral supplements on morbidity from infections in

567

568

AC C

566

EP

563

older people (MAVIS trial): pragmatic, randomised, double blind, placebo controlled trial. BMJ 2005;331:324-329.

[34] Macpherson H, Ellis KA, Sali A, Pipingas A. Memory improvements in elderly

569

women following 16 weeks treatment with a combined multivitamin, mineral

570

and herbal supplement: A randomized controlled trial. Psychopharmacology

571

(Berl) 2012;220:351-365.

28

ACCEPTED MANUSCRIPT 572

[35] Volkert D, Hubsch S, Oster P, Schlierf G. Nutritional support and functional

573

status in undernourished geriatric patients during hospitalization and 6-month

574

follow-up. Aging (Milano ) 1996;8:386-395. [36] Macpherson H, Silberstein R, Pipingas A. Neurocognitive effects of

RI PT

575

multivitamin supplementation on the steady state visually evoked potential

577

(SSVEP) measure of brain activity in elderly women. Physiol Behav

578

2012;107:346-354.

579

SC

576

[37] Chavance M, Herbeth B, Lemoine A, Zhu BP. Does multivitamin supplementation prevent infections in healthy elderly subjects? A controlled

581

trial. Int J Vitam Nutr Res 1993;63:11-16.

582

M AN U

580

[38] Murphy SP, White KK, Park SY, Sharma S. Multivitamin-multimineral supplements' effect on total nutrient intake. Am J Clin Nutr 2007;85:280S-

584

284S.

TE D

583

[39] Bjelakovic G, Nikolova D, Gluud LL, Simonetti RG, Gluud C. Mortality in

586

randomized trials of antioxidant supplements for primary and secondary

587

prevention: systematic review and meta-analysis. JAMA 2007;297:842-857.

AC C

EP

585

588

[40] Graat JM, Schouten EG, Kok FJ. Effect of daily vitamin E and multivitamin-

589

mineral supplementation on acute respiratory tract infections in elderly

590

591

persons: a randomized controlled trial. JAMA 2002;288:715-721.

[41] Li JY, Taylor PR, Li B, Dawsey S, Wang GQ, Ershow AG, Guo W, Liu SF,

592

Yang CS, Shen Q. Nutrition intervention trials in Linxian, China: multiple

593

vitamin/mineral supplementation, cancer incidence, and disease-specific

29

ACCEPTED MANUSCRIPT 594

mortality among adults with esophageal dysplasia. J Natl Cancer Inst

595

1993;85:1492-1498.

596

[42] Mursu J, Robien K, Harnack LJ, Park K, Jacobs DR, Jr. Dietary supplements and mortality rate in older women: the Iowa Women's Health Study. Arch

598

Intern Med 2011;171:1625-1633.

RI PT

597

[43] Bailey RL, Fakhouri TH, Park Y, Dwyer JT, Thomas PR, Gahche JJ, Miller

600

PE, Dodd KW, Sempos CT, Murray DM. Multivitamin-mineral use is

601

associated with reduced risk of cardiovascular disease mortality among

602

women in the United States. J Nutr 2015;145:572-578.

M AN U

603

SC

599

[44] Macpherson H, Pipingas A, Pase MP. Multivitamin-multimineral supplementation and mortality: a meta-analysis of randomized controlled

605

trials. Am J Clin Nutr 2013;97:437-444.

606

[45]

TE D

604

Goh YI, Bollano E, Einarson TR, Koren G. Prenatal multivitamin supplementation and risk of congenital malformation. J Obstet Gynecol Can

608

2006;28:680-689. [46]

Screening and diagnosis of micronutrient deficiencies before and after

610

bariatric surgery. Obes Surg 2013;23:1581-1589.

611

612

Gudzune KA, Huizinga MM, Chang HY, Asamoah V, Gadgil M, Clark JM.

AC C

609

EP

607

[47]

Homan J, Betzel B, Aarts EO, Dogan K, van Laarhoven KJ, Janssen IM,

613

Berends FJ. Vitamin and Mineral deficiencies after biliopancreatic diversion

614

and biliopancreatic diversion with duodenal switch-the ruler rather than the

615

exception. Obes Surg 2015;25:1626-1632.

30

ACCEPTED MANUSCRIPT 616

[48]

after weight reduction. Ann Nutr Disord & Ther 2015;2:1024-1029.

617

618

Bhatti O, Klaus Bielefeldt K, Nusrat S. Nutritional deficiencies in obesity and

[49]

Correia Horvath JD, Dias de Castro ML, Kops N, Kruger Malinoski N, Friedman R. Obesity coexists with malnutrition? Adequacy of food

620

consumption by severely obese patients to dietary reference

621

recommendations. Nutr Hosp 2014;29:292-299.

SC

Schweiger C, Weiss R, Berry E, Keidar A. Nutritional deficiencies in bariatric

TE D

M AN U

surgery candidates. Obes Surg 2009;19:83-88.

EP

623

[50]

AC C

622

RI PT

619

31

ACCEPTED MANUSCRIPT Table 1

625

Hidden hunger: (A) Vitamin A deficiency in children (< 5 years of age) and pregnant

626

women in developed countries; (B): Iodine, zinc, and iron-deficiency anemia in

627

children (< 5 years of age) and pregnant women in developed countries1

628 629

A

RI PT

624

Vitamin A deficiency

Americas and the Caribbean Asia Europe Oceania B

33.5 (30.7–36.3) 14.9 (0.1–29.7) 12.6 (6.0–19.2)

Iodine deficiency (UIC < 100 µg/L)

28.5 (28.2–28.9) 40.0 (39.4–40.6) 13.7 (12.5–14.8)

Zinc deficiency (weighted average of country means) 17.3 (15.9–18.8) 23.9 (21.1–26.8) 9.6 (6.8–12.4)

18.1 (15.6–20.8) 20.2 (18.6–21.7) 12.7 (9.8–16.0)

19.2 (17.1–21.5) 20.3 (18.3–22.4) 15.2 (11.7–18.6)

31.6 (30.7–32.5) 44.2 (43.5–45.0) 17.3 (16.6–18.1)

19.4 (16.9–22.0) 7.6 (6.2–9.1) 5.7 (1.0–10.3)

19.0 (14.5–23.4) 12.1 (7.8–16.2) 15.4 (7.0–25.2)

19.8 (15.8–23.5) 16.2 (12.6–19.7) 17.2 (9.7–25.6)

EP

Region

0.5 (0.0–1.3) 0.7 (0.0–1.5) 0.5 (0.1–1.0)

AC C

630 631

Global Africa

Americas and the Caribbean Asia Europe Oceania

SC

Africa

Pregnant women Night Serum retinol blindness < 0.70 µmol/L 7.8 15.3 (6.5–9.1) (6.0–24.6) 9.4 14.3 (8.1–10.7) (9.7–19.0) 4.4 2.0 (2.7–6.2) (0.4–3.6)

M AN U

Global

Children < 5 years of age Night Serum retinol blindness < 0.70 µmol/L 0.9 33.3 (0.1–1.8) (29.4–37.1) 2.1 41.6 (1.0–3.1) (34.4–44.9) 0.6 15.6 (0.0–1.3) (6.6–24.5)

TE D

Region

7.8 (6.6–9.0) 2.9 (1.1–4.6) 9.2 (0.3–18.2)

18.4 (5.4–31.4) 2.2 (0.0–4.3) 1.4 (0.0–4.0)

Iron-deficiency anemia (hemoglobin < 110 g/L) Children Pregnant < 5 years women of age

32

ACCEPTED MANUSCRIPT 632 633

1

634

inadequate zinc intake (2005), and iron deficiency anemia (2011). Data in % (95 %

635

CI) [16]. UIC, urine iodine concentration.

AC C

EP

TE D

M AN U

SC

RI PT

Prevalence of vitamin A deficiency (1995–2005), iodine deficiency (2013),

33

ACCEPTED MANUSCRIPT Highlights MVM use is common in the general population which raises questions regarding the safety of long-term use of these supplements

•

There is an increasing number of individuals taking multivitamin/minerals (MVM) to maintain good health and to be protected from different diseases

•

MVM use within the range of the Dietary Reference Intake (DRI) will not result in excess intake, even when including the impact of food and fortified food, and does not increase mortality

•

Despite a balanced and overall healthy diet, micronutrient gaps may occur from time to time

•

An MVM can help to improve the nutrient supply and overcome problems of inadequacy without concern for its long-term safety

•

MVM are safe for long-term use (more than 10 years) as documented in a recent clinical trial.

AC C

EP

TE D

M AN U

SC

RI PT

•