Prevalence and Predictors of Selenium and Vitamin E Supplementation in a Urology Population

Abstract

Objectives:

Preclinical and epidemiological studies have suggested the use of supplements such as selenium and vitamin E for prostate cancer (PCa) prevention; however, clinical trials have not demonstrated clear benefit in patients. This study aims to investigate the current prevalence and predictors for use of these supplements in men in a urology population.

Design, subjects, and outcomes measured:

Three hundred and twelve (312) men visiting the Princess Margaret Hospital Ambulatory Urology Clinic were enrolled in this University Health Network Research Ethics Board–approved questionnaire-based study investigating supplement use, reasons for use and demographic characteristics.

Results:

It was observed that 13.5% and 20.8% of the participants are currently using selenium and vitamin E, respectively, while 10.6% and 15.7% previously used selenium and vitamin E, respectively. Both education (percentage of users comparing less than college education versus college or above education: selenium: 14% versus 28%; p=0.008, vitamin E: 26% versus 41%; p=0.013) and health literacy (mean scores±standard error of the mean of users versus nonusers: selenium question 1: 1.4507±0.09576 versus 1.6083±0.07211; p=0.023, selenium question 2: 2.8750±0.04395 versus 2.7106±0.03774; p=0.000, selenium question 3: 1.4583±0.08377 versus 1.7064±0.06278; p=0.025, vitamin E question 1: 2.8036±0.04545 versus 2.7179±0.04097; p=0.010, vitamin E question 2: 1.5446±0.06698 versus 1.7077±0.07241; p=0.006) are predictors of selenium and vitamin E use on univariable analysis. On multivariable analysis education (selenium odds ratio=2.095, 95% confidence interval=1.019–4.305, p=0.044; vitamin E odds ratio=1.855, 95% confidence interval=1.015–3.338, p=0.045) remains a significant predictor of selenium and vitamin E use. Examining the data on use by education, it was found that more men with a higher education attributed their use of selenium to urologist advice (24%), and those with a lower education attributed their use of selenium to naturopath/homeopath advice (28%).

Conclusions:

Many men who visit urology clinics use selenium and vitamin E despite the lack of clinical support for chemoprevention. Education and health literacy are important variables in determining the use of these supplements in these men. This information may aid in addressing the needs of the diverse patient population using these supplements for the prevention of PCa.

Introduction

P rostate cancer (PCa) is one of the most common malignancies facing Canadian men. Aside from family history and race, diet has long been implicated as having a role in PCa. Diets rich in beans, rice, cornbread, sweet potatoes, and okra and low in cooked processed meats have been associated with reduced risk of PCa,^1,2 while diets high in dairy protein and saturated fats are associated with increased risk of PCa.³ From this influential role of diet in PCa, an interest in vitamins and minerals such as selenium and vitamin E in the possible prevention of the disease has evolved.⁴

In 1996, the Nutritional Cancer Study investigating the effect of selenium supplementation on carcinoma of the skin discovered secondarily that selenium supplementation was associated with a lower incidence of PCa.⁵ Selenium, obtained naturally from plant and animal products, is a cofactor for antioxidant enzymes.⁶ In preclinical studies, selenium has been shown to be an activator of p53, which in turn regulates the DNA repair/apoptosis axis⁶ and when administered at levels higher than nutritional needs has been shown to inhibit growth of LNCaP xenograft tumors.⁷ In 1998, the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Trial assessing lung cancer as a primary endpoint discovered secondarily an association of vitamin E with decreased risk of PCa.⁸ Vitamin E, a fat-soluble vitamin obtained from nuts, vegetable oil, and egg yolks, is a free-radical antioxidant shown to inhibit NF-κB, androgen receptor function, cell cycle, and DNA damage in in vitro PCa cell models as well as slow growth of PCa xenografts in vivo. ^9,10 Because selenium and vitamin E are hypothesized to act through different signaling cascades, their potential additive effect may be important in the clinical prevention of PCa, and this hypothesis has been supported experimentally: combined selenium and vitamin E inhibited PCa cell growth by 78% as compared to 47% with vitamin E alone and 37% with selenium alone.¹⁰ These preclinical and epidemiological results were the basis of a large scale randomized clinical trial evaluating selenium and vitamin E chemoprevention. In this Selenium and Vitamin E Cancer Prevention Trial (SELECT), 35,533 men received placebo, selenium alone, vitamin E alone, or selenium and vitamin E in combination.¹¹ Results from the trial could not clearly demonstrate that selenium and vitamin E either alone or in combination are associated with decreased risk of PCa incidence. However, it remains uncertain whether selenium and vitamin E may have a clinical benefit in preventing PCa in a select group of patients who are at higher risk for developing the disease.¹² Regardless of the SELECT results, the use of supplements such as selenium and vitamin E have gained widespread attention over the past 30 years, increasing daily use of vitamin E in the United States by 7.3% in 1987–2000 and as high as 54% prevalence by 2005.^13,14 This study set out to assess the current prevalence, and predictors of selenium and vitamin E use in men in a urology population.

Materials and Methods

This study was conducted using a questionnaire-based assessment of men with PCa or at risk for developing PCa. The study protocol was approved by the Human Research Ethics Board of the University Health Network.

Participants

Participants were patients visiting the Princess Margaret Hospital Ambulatory Urology Clinic in downtown Toronto for a prostate condition (i.e., abnormal prostate specific antigen test, abnormal biopsy, previously diagnosed PCa). Participants were included in the study if they were male, over the age of 18, and consented to participation.

Data collection

Participants were administered a two-part questionnaire. The first section contained a question regarding current or previous use of supplements (specifically, vitamins A, B, C, D, and E, calcium, iron, magnesium, selenium, green tea, ginseng, lycopene, omega 3, saw palmetto, and soy). The remaining questions addressed reasons for current or previous use of these supplements. Options for use included external recommendations by family member, family doctor, urologist, naturopath/homeopath, other health care professional, or self-recommendation based on information learned via newspaper, periodical, Internet, TV/radio, and other. The second section contained questions regarding birthplace, age, ethnicity, relationship status, education, income, and health literacy. To assess health literacy, three questions were used: Q1 How often do you have someone help you read hospital materials? (answer options: 1=never, 2=rarely, 3=sometimes, and 4=frequently). Q2 How confident are you filling out medical forms by yourself? (answer options: 1=not very, 2=rarely, and 3=very). Q3 How often do you have problems learning about a medical condition because of difficulty understanding written information? (answer options: 1=never, 2=rarely, 3=sometimes, and 4=frequently). These three questions were chosen based on a positive meta-analysis comparison of various surveys used to assess health literacy.¹⁵

Data and statistical analysis

Prevalence of supplement use and characteristics of the complete participant sample were calculated. For statistical testing of means and proportions, Student's t-test and Pearson's χ² analyses were conducted, respectively. Multivariable logistic regression analyses were also utilized to evaluate the relationship between supplement use and relevant covariates. Variables included birthplace, age, ethnicity, relationship status, education, income, and health literacy. Subgrouping of participants by these variables for comparison was conducted in accord with the Canadian Community Health Survey.¹⁶ For example, participants were split into less than college (lower education level) and greater than or equal to college (higher education level) groups for education assessment, while income assessment was split into groups of those who make less than $100,000 per year from those who make $100,000 or more per year. Participants who were currently married or in a marriage-like relationship were grouped together (married) and compared to those in the remaining categories (not married), and participants born outside of Canada were grouped together (other) and compared to those born in Canada (Canada). Scoring for health literacy Q2 is opposite that of Q1 and Q3. These questions were assessed on a scale from 1 to 3 or 4, where 3 indicates the highest level of health literacy in Q2 and 1 indicates the highest level of health literacy in Q1 and Q3. All statistical analysis was conducted using SPSS, (IBM Corporation, Armonk, New York); p<0.05 was considered statistically significant.

Results

Overall participant characteristics

A total of 312 of 340 men with PCa or at risk for developing PCa completed the study (Table 1) (92% response rate). Most participants were white (82.1%), born in countries other than Canada (51.9%), and were presently married at the time of the study (73.4%). The average age of the participants was 66.07±0.72, and the average income grouping of the participants was $75,000–$99,000.

Table 1.

Demographic Characteristics of the Population as Displayed by Number of Users and Percentage (%) of Total Population

Variable	Number	%
Birthplace
Canada	145	46.5
Other	162	51.9
Declined to answer	5	1.6
Ethnicity
White	256	82.1
Asian	24	7.7
Black, African-American/African-Canadian	16	5.1
Interracial	5	1.6
Aboriginal	2	0.6
Other	2	0.6
Declined to answer	7	2.2
Education
Grade school	18	5.8
Some high school	19	6.1
High school graduation	19	6.1
Vocational training	7	2.2
Some college	29	9.3
College graduation	85	27.2
Some graduate school	88	28.2
Doctoral/professional degree	43	13.8
Declined to answer	4	1.3
Marital status
Never married	15	4.8
Presently married	229	73.4
Living in a marriage-like relationship	21	6.7
Divorced or separated	29	9.3
Widowed	12	3.8
Declined to answer	6	2.0
Income
Less than $10,000	4	1.3
$10,000–19,999	13	4.2
$20000–34999	20	6.4
$35,000–49,999	18	5.8
$50,000–74,999	56	17.9
$75,000–99,999	46	14.7
$100,000–149,999	48	15.4
$150,000 or more	73	23.4
Declined to answer	34	10.9

Prevalence of current and previous selenium and vitamin E use in a urology population

Overall, it was determined that 89.1% of participants were currently using any form of supplements. In terms of selenium and vitamin E, 13.5% and 20.8% of the participants were currently using selenium and vitamin E, respectively, while 10.6% and 15.7% previously used selenium and vitamin E, respectively.

Demographic characteristics of users of selenium and vitamin E

On univariable analysis, both education (percentage of users comparing less than college education versus college or above education: selenium: 14% versus 28%; p=0.008, vitamin E: 26% versus 41%; p=0.013) (Table 2) and health literacy (mean scores±standard error of the mean of users versus nonusers: selenium question 1: 1.4507±0.09576 versus 1.6083±0.07211; p=0.023, selenium question 2: 2.8750±0.04395 versus 2.7106±0.03774; p=0.000, selenium question 3: 1.4583±0.08377 versus 1.7064±0.06278; p=0.025, vitamin E question 1: 2.8036±0.04545 versus 2.7179±0.04097; p=0.010, vitamin E question 2: 1.5446±0.06698 versus 1.7077±0.07241; p=0.006) (Table 3) are predictors of selenium and vitamin E use on univariable analysis. On multivariable analysis, education (selenium odds ratio=2.095, 95% confidence interval=1.019–4.305, p=0.044; vitamin E odds ratio=1.855, 95% confidence interval=1.015–3.338, p=0.045) remains a significant predictor of selenium and vitamin E use while health literacy did not (Tables 4 and 5).

Table 2.

Demographic Characteristics of Selenium and Vitamin E Users as Displayed by Number of Participants and Percentage (%) of Total Population

Variable	Total no. in group	No. of selenium users	No. of users in group	p-Value	No. of vitamin E users	No. of users in group	p-Value
Ethnicity
White	256	60	23%		95	37%
Other	56	15	27%		19	34%
				0.595			0.654
Education
Less than college	92	13	14%		24	26%
College or above	220	62	28%		90	41%
				0.008^*			0.013^*
Marital status
Never married	62	14	23%		26	42%
Marriage-like relationship	250	61	24%		80	32%
				0.764			0.324
Income
Less than $100,000	157	33	21%		49	31%
$100,000 or more	155	42	27%		65	42%
				0.209			0.049^*

Univariable χ² statistical analyses displayed as p-value (^* p<0.05 considered statistically significant).

Table 3.

Health Literacy Characteristics of Users and Nonusers of Selenium and Vitamin E

Question no.	Selenium user	Selenium nonuser	p-Value	Vitamin E user	Vitamin E nonuser	p-Value
Q1- Mean	1.4507	1.6083		1.5596	1.5754
SEM	±0.09576	±0.07211		±0.08590	±0.07996
			^*0.023			0.356
Q2- Mean	2.8750	2.7106		2.8036	2.7179
SEM	±0.04395	±0.03774		±0.04545	±0.04097
			^*0.000			0.010^*
Q3- Mean	1.4583	1.7064		1.5446	1.7077
SEM	±0.08377	±0.06278		±0.06698	±0.07241
			^*0.025			0.006^*

Means and standard error of the means (SEM) displayed for each question and t-test statistical analyses displayed as p-value (^* p<0.05 considered statistically significant). Health literacy questions were as follows: Q1 How often do you have someone help you read hospital materials? (answer options: 1=never, 2=rarely, 3=sometimes, 4=frequently). Q2 How confident are you filling out medical forms by yourself? (answer options: 1=not very, 2=rarely, and 3=very). Q3 How often do you have problems learning about a medical condition because of difficulty understanding written information? (answer options: same as Q1).

Table 4.

Multivariable Analysis for Users Versus Nonusers of Selenium

Variable	OR	CI	p-Value
Age	0.995	0.969–1.022	0.716
Education	2.095	1.019–4.305	0.044^*
Marital status	0.884	0.439–1.781	0.730
Income	1.093	0.615–1.941	0.762
Health literacy Q1	1.023	0.728–1.437	0.896
Health literacy Q2	1.581	0.749–3.336	0.229
Health literacy Q3	0.898	0.586–1.377	0.623

Multivariable analysis displayed as odds ratio (OR), confidence interval (CI), p-value. (^* p<0.05 considered statistically significant). Variable groups were as follows: age—continuous variable; education—less than college versus greater than or equal to college; marital status—marriage-like relationship or married versus remaining; income—less than $100,000 per year versus $100,000 or more per year and health literacy: Q1 How often do you have someone help you read hospital materials? (answer options: 1=never, 2=rarely, 3=sometimes, 4=frequently). Q2 How confident are you filling out medical forms by yourself? (answer options: 1=not very, 2=rarely, and 3=very). Q3 How often do you have problems learning about a medical condition because of difficulty understanding written information? (answer options: same as Q1).

Table 5.

Multivariable Analysis for Users Versus Nonusers of Vitamin E

Variable	OR	CI	p-Value
Age	1.005	0.982–1.030	0.658
Ethnicity	1.143	0.595–2.194	0.688
Education	1.855	1.015–3.388	0.045^*
Marital status	0.556	0.301–1.028	0.061
Income	1.405	0.840–2.352	0.195
Health literacy Q1	1.153	0.861–1.545	0.339
Health literacy Q2	1.118	0.630–1.984	0.702
Health literacy Q3	0.884	0.618–1.265	0.501

Multivariable analysis displayed as odds ratio (OR), confidence interval (CI), p-value. (^* p<0.05 considered statistically significant). Variable groups were as follows: age— continuous variable; education—less than college versus greater than or equal to college; marital status—marriage-like relationship or married versus remaining; income—less than $100,000 per year versus $100,000 or more per year and health literacy: Q1=How often do you have someone help you read hospital materials? (answer options: 1=never, 2=rarely, 3=sometimes, 4=frequently). Q2 How confident are you filling out medical forms by yourself? (answer options: 1=not very, 2=rarely, and 3=very). Q3 How often do you have problems learning about a medical condition because of difficulty understanding written information? (answer options: same as Q1).

Participant reasons for selenium and vitamin E use

Overall, a larger proportion of participants attributed their use of selenium and vitamin E to urologist advice (20.8%) and family member advice (20.1%), respectively, as compared to other reasons. Examining data on use by education, it was found that more participants with a higher education attributed their use of selenium to urologist advice (24.4%) and periodical readings (18.0%), and those with a lower education attributed their use of selenium to naturopath/homeopath advice (27.8%) and periodical readings (22.2%). In terms of current vitamin E use, regardless of education level, most men attributed their current use to family member (30.8-38.9%) and family doctor advice (21.8–50.0%).

With respect to previous selenium and vitamin E use, most participants attributed stopping of use to other reasons (46.2% and 29.4%). These other reasons recorded by participants include article discovery, loss in interest, word of mouth, development of side-effects, changed diet to take naturally, felt no benefit, too many medications and laziness.

Discussion

This study explored the current prevalence and predictors of selenium and vitamin E use in 312 men in a urology population. It was confirmed for the first time that 13.5% and 20.8% of these men continue to use selenium and vitamin E, respectively, despite their lack of proven benefit.¹¹ In 1986, before the SELECT trial, the United States National Health Interview Survey reported approximately 9% and 26% of all adults using selenium and vitamin E supplements, respectively.¹⁷ In 1999, in a similar population of patients with or at risk for PCa, 35.7% and 54.8% of respondents reported taking selenium and vitamin E, respectively.¹⁸ After the SELECT trial, a decrease in selenium (13.5%) and vitamin E (20.8%) subscription was observed compared to the pre-SELECT results in this similar cohort of patients.¹⁸ However, these results demonstrate that there still remain many patients utilizing these supplements, and this has important implications. Additional to the current lack of efficacy in PCa prevention in the general population, there is no convincing evidence for the efficacy of vitamin E as a heart protector.^19,20 Furthermore, Miller et al. reported that doses of vitamin E>400 IU/d are associated with a higher all-cause mortality rate.²⁰ The risk versus benefit ratio in terms of cardiac disease, toxicity of formulation, and PCa disease makes it difficult to prove it a good supplement for men in the general population. This will have to be assessed in future trials. With respect to selenium, it has not been shown to be associated with health benefits in a meta-analysis of randomized trials regarding several diseases.²¹ In fact, in the Nutritional Prevention of Cancer Trial, selenium supplementation was associated with an increased risk of squamous cell carcinoma and nonmelenoma skin cancer,²² and in a prospective study increased dietary selenium intake has also been associated with increased risk of type 2 diabetes.²³ Furthermore, the development of supplement formulation remains unregulated despite some people developing acute toxicities from a product in 2009 in the United States.^24,25 Although there are changing trends in use of vitamin E and selenium in 2010 since 1999, a troubling high proportion of patients are still observed with or at risk for PCa utilizing these supplements despite only clear level 1 evidence for use and no proven benefit by clinical study results.

Out of the demographic predictors assessed in this study, education and health literacy were found to be significant predictors of both selenium and vitamin E use in this population. Higher health literacy was associated with increased prevalence of selenium and vitamin E on univariable analysis. Health literacy is defined by the Canadian Public Health Association as an individual's ability to access, understand, and use information related to their health and attribute skills including ability to read and comprehend documents as well as be proficient in numeracy and document literacy.²⁶ Higher education level was statistically more predictive of selenium and vitamin E use as determined on univariable and multivariable analysis, indicating that education is a central influencer of their use. Although few studies have investigated health literacy effect on supplement use, some studies have previously correlated higher education level with increased overall supplement/unconventional therapy use in the general population.^27

–31

Interestingly, data from this study suggest that the motivators for use of vitamin E and selenium are quite different among patients. As education and health literacy were identified as predictors of selenium and vitamin E use in this urology population, the authors were interested in determining whether there are any methods that could be used to increase the awareness regarding selenium and vitamin E use in men subgrouped by these demographic variables, particularly education. Through assessments of reasons for use, it was determined that men with a higher education level were shown to take selenium for very different reasons than men with a lower education level. Urologist advice and periodical readings appear to be the most influential methods for predicting current selenium use in men with a higher education level and therefore may offer useful means for targeting. Urologist education and periodical reviews regarding selenium supplementation may help men with a higher education level improve their knowledge around the issue. In terms of vitamin E, regardless of education level, reasons for use were related to family member or family physician advice; these two facets of communication may be utilized to better provide information to men regarding the benefits and harms of utilizing vitamin E supplements. One limitation to this study is that the specific methods by which urologists, family members, and family physicians influenced patients to use/stop using vitamin E and selenium were not evaluated. Future qualitative studies should determine and optimize these methods by which urologists, family members and family physicians advise patients regarding the use of these supplements. In turn, studies evaluating the effect of targeting men through these methods will determine whether these are useful approaches to improving the health of the general population.

This study demonstrates for the first time that a significant proportion of men visiting urology clinics continue to use selenium and vitamin E despite only level 1 evidence for use. Avenues to identify and target subgroups of users in the general public to improve awareness regarding these agents were explored. Nonetheless, there are limitations to this study. First, in terms of external validity, most participants were white, presently married at the time of the study, have an average annual income of $75,000–$99,999 and were visiting a urology clinic, different from the trends in the general population in the surrounding area. Further assessment of selenium and vitamin E use in a community setting may solidify results reported here. Second, doses and preparations of selenium and vitamin E were not evaluated, and this information may help determine some of the unknown toxicities expected and provide reasons for any determined lack of benefit in clinical studies.

Conclusions

Overall, this study identified education and health literacy as significant predictors of selenium and vitamin E use in men with or at risk for developing PCa. Preparing quality periodical reviews on the subject matter and addressing urologist–patient communication in men with a higher education level as well as addressing family physician–patient communication in all men may improve the gap of understanding regarding the comprehensive use of these supplements.

Footnotes

Acknowledgments

We would like to sincerely thank Drs. Antonio Finelli, Michael Jewett, John Trachtenberg, and Alexandre Zlotta for supporting participant recruitment in their urology clinics.

Disclosure Statement

The authors have nothing to disclose.

References

Rodriguez

, McCullough

, Mondul

et al. Meat consumption among black and white men and risk of prostate cancer in the Cancer Prevention Study II Nutrition Cohort. Cancer Epidemiol Biomarkers Prev, 2006; 15:211–216.

Tseng

, Breslow

, DeVellis

et al. Dietary patterns and prostate cancer risk in the National Health and Nutrition Examination Survey Epidemiological Follow-up Study cohort. Cancer Epidemiol Biomarkers Prev, 2004; 13:71–77.

Gonzalez

, Riboli

. Diet and cancer prevention: Contributions from the European Prospective Investigation into Cancer and Nutrition (EPIC) study. Eur J Cancer, 2010; 46:2555–2562.

Platz

. Is prostate cancer prevention with selenium all in the genes? Cancer Prev Res (Phila), 2010; 3:576–578.

Clark

, Combs

Jr , Turnbull

et al. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin: A randomized controlled trial. Nutritional Prevention of Cancer Study Group. JAMA, 1996; 276:1957–1963.

Sarveswaran

, Liroff

, Zhou

et al. Selenite triggers rapid transcriptional activation of p53, and p53-mediated apoptosis in prostate cancer cells: Implication for the treatment of early-stage prostate cancer. Int J Oncol, 2010; 36:1419–1428.

Lee

, Yeon

Chun J

, Nadiminty

et al.

Monomethylated selenium inhibits growth of LNCaP human prostate cancer xenograft accompanied by a decrease in the expression of androgen receptor and prostate-specific antigen (PSA)

Prostate, 2006; 66:1070–1075.

Heinonen

, Albanes

, Virtamo

et al. Prostate cancer and supplementation with alpha-tocopherol and beta-carotene: Incidence and mortality in a controlled trial. J Natl Cancer Inst, 1998; 90:440–446.

Fleshner

, Fair

, Huryk

et al. Vitamin E inhibits the high-fat diet promoted growth of established human prostate LNCaP tumors in nude mice. J Urol, 1999; 161:1651–1654.

10.

, Chen

, Zhang

et al. Vitamin E succinate inhibits human prostate cancer cell growth via modulating cell cycle regulatory machinery. Biochem Biophys Res Commun, 2003; 300:357–363.

11.

Lippman

, Klein

, Goodman

et al.

Effect of selenium and vitamin E on risk of prostate cancer and other cancers: the Selenium and Vitamin E Cancer Prevention Trial (SELECT)

JAMA, 2009; 301:39–51.

12.

Moyad

. Selenium and vitamin E supplements for prostate cancer: evidence or embellishment? Urology, 2002; 59:9–19.

13.

Bishop

, Rea

, Lewith

et al. Complementary medicine use by men with prostate cancer: A systematic review of prevalence studies. Prostate Cancer Prostatic Dis, 2011; 14:1–13.

14.

Millen

, Dodd

, Subar

. Use of vitamin, mineral, nonvitamin, and nonmineral supplements in the United States: The 1987, 1992, and 2000 National Health Interview Survey results. J Am Diet Assoc, 2004; 104:942–950.

15.

Powers

, Trinh

, Bosworth

. Can this patient read and understand written health information? JAMA, 2010; 304:76–84.

16.

Canadian Community Health Survey (CCHS), Statistics Canada. www.statcan.gc.ca/cgi-bin/imdb/p2SV.pl?Function=getSurvey&SDDS=3226&lang=en&db=imdb&adm=8&dis=2#a4. 2011 March 15.

17.

Subar

, Block

. Use of vitamin and mineral supplements: Demographics and amounts of nutrients consumed. The 1987 Health Interview Survey. Am J Epidemiol, 1990; 132:1091–1101.

18.

Nam

, Fleshner

, Rakovitch

et al. Prevalence and patterns of the use of complementary therapies among prostate cancer patients: An epidemiological analysis. J Urol, 1999; 161:1521–1524.

19.

, Song

. Antioxidant vitamins intake and the risk of coronary heart disease: Meta-analysis of cohort studies. Eur J Cardiovasc Prev Rehabil, 2008; 15:26–34.

20.

Miller

3rd , Pastor-Barriuso

, Dalal

et al. Meta-analysis: High-dosage vitamin E supplementation may increase all-cause mortality. Ann Intern Med, 2005; 142:37–46.

21.

Navas-Acien

, Bleys

, Guallar

. Selenium intake and cardiovascular risk: What is new? Curr Opin Lipidol, 2008; 19:43–49.

22.

Duffield-Lillico

, Slate

, Reid

et al. Selenium supplementation and secondary prevention of nonmelanoma skin cancer in a randomized trial. J Natl Cancer Inst, 2003; 95:1477–1481.

23.

Stranges

, Sieri

, Vinceti

et al. A prospective study of dietary selenium intake and risk of type 2 diabetes. BMC Public Health, 2010; 10:564.

24.

Brozmanova

, Manikova

, Vlckova

et al. Selenium: A double-edged sword for defense and offence in cancer. Arch Toxicol, 2010; 84:919–938.

25.

MacFarquhar

, Broussard

, Melstrom

et al. Acute selenium toxicity associated with a dietary supplement. Arch Intern Med, 2010; 170:256–261.

26.

Health Literacy Canadian Public Health Association. www.cpha.ca/en/portals/h-l.aspx. 2011 March 15.

27.

Friel

, Kelleher

, Nolan

et al. Social diversity of Irish adults nutritional intake. Eur J Clin Nutr, 2003; 57:865–875.

28.

Sebastian

, Cleveland

, Goldman

et al. Older adults who use vitamin/mineral supplements differ from nonusers in nutrient intake adequacy and dietary attitudes. J Am Diet Assoc, 2007; 107:1322–1332.

29.

van der Pal-de Bruin

, de Walle

, de Rover

et al. Influence of educational level on determinants of folic acid use. Paediatr Perinat Epidemiol, 2003; 17:256–263.

30.

Eisenberg

, Davis

, Ettner

et al. Trends in alternative medicine use in the United States, 1990–1997: Results of a follow-up national survey. JAMA, 1998; 280:1569–1575.

31.

Eisenberg

, Kessler

, Foster

et al. Unconventional medicine in the United States. Prevalence, costs, and patterns of use. NEJM, 1993; 328:246–252.