Although the mortality rate of patients with CHD has decreased by 52 % in men and 49 % in women from 1980 to 2002(Reference Ford, Ajani and Croft1) and the mortality rate of stroke decreased by 24·2 % from 1994 to 2004(Reference Rosamond, Flegal and Furie2), the burden of these diseases remains high. Among US adults, the prevalence of CHD was 7·3 % and the prevalence of stroke was 2·6 % in 2005.
Hyperhomocysteinaemia is associated with an increased risk for CVD(Reference Clarke, Collins and Lewington3–Reference Wald, Watt and Law5). Results from meta-analyses have shown that for a 5 μmol/l increase in serum homocysteine levels, the risk of myocardial infarction increased by 84 %(Reference Wald, Watt and Law5), the risk of CHD increased by 23–70 %(Reference Ford, Smith and Stroup4), and the risk of cerebrovascular disease increased by 58–116 %(Reference Ford, Smith and Stroup4), although prospective studies offer weaker support for this association than case–control studies(Reference Ford, Smith and Stroup4). A higher intake of folic acid or multivitamins reduces serum homocysteine levels(Reference Clarke, Frost and Sherliker6, Reference Neal, MacMahon and Ohkubo7) and is associated with reduced risk of acute coronary events and myocardial infarction(Reference Holmquist, Larsson and Wolk8–Reference Tavani, Pelucchi and Parpinel11) or reduced rates of coronary restenosis and revascularisation in patients with coronary angioplasty(Reference Schnyder, Roffi and Pin12, Reference Schnyder, Roffi and Flammer13). However, other studies including recent clinical trials failed to show any beneficial effects on the risk of recurrent CVD or on total CVD or CHD mortality when using folic acid and B vitamins to lower homocyteine levels among patients with pre-existing CVD or renal disease(Reference Bazzano, Reynolds and Holder14–Reference Muntwyler, Hennekens and Manson17) or among people at high risk for CVD (having a history of vascular disease, diabetes or other coronary risk factors)(Reference Albert, Cook and Gaziano18, Reference Lonn, Yusuf and Arnold19).
From 1990–1997 to 1998–2002, stroke-related mortality has decreased dramatically in the United States and Canada(Reference Yang, Botto and Erickson20). Folic acid fortification of flour and cereal grain initiated in 1998 may have contributed to this decline(21). In addition, folic acid supplementation could effectively reduce the risk of stroke in primary prevention(Reference Bazzano, He and Ogden22–Reference Wang, Qin and Demirtas25); however, its role in the secondary prevention of stroke remains to be elucidated.
The National Health and Nutrition Examination Survey (NHANES) data have shown that dietary supplement use increased from 23 % in NHANES I (1971–1975) to 40 % in NHANES III (1988–1994)(Reference Radimer, Bindewald and Hughes26). The 1999–2000 NHANES data also showed that 52 % of US adults reported taking dietary supplements in the past month, and 35 % reported regular use of multivitamin–multimineral supplements(Reference Radimer, Bindewald and Hughes26). However, the prevalence and correlates of taking folic acid and vitamin supplements among people with CVD are not well understood. Using a relatively large survey sample, we examined the percentage and likelihood of adults with CHD or stroke who reported taking folic acid or vitamin supplements in comparison with those observed in adults without these conditions.
Methods
Data for our analyses came from the 2006 Behavioural Risk Factor Surveillance System (BRFSS), a population-based telephone survey of health-related behaviours regarding the leading causes of death among non-institutionalised US adults aged ≥18 years. The survey was reviewed by the Human Research Protection Office at the Centers for Disease Control and Prevention and was determined to be exempt from human subject guidelines. The BRFSS survey design, sampling methods and weights have been described elsewhere(Reference Mokdad, Stroup and Giles27), and BRFSS data have consistently been found to provide valid and reliable estimates when compared with other national household surveys in the United States(Reference Mokdad, Stroup and Giles27–Reference Nelson, Powell-Griner and Town29). Further information on BRFSS is available at http://www.cdc.gov/brfss/.
We analysed the data collected from survey participants aged ≥45 years in twelve states (Delaware, Florida, Georgia, Kentucky, Minnesota, Missouri, Montana, New York, North Carolina, Virginia, Wisconsin and Wyoming) and Puerto Rico that had an optional module on folic acid and vitamin supplementation in the 2006 BRFSS. The median cooperation rate (the percentage of eligible persons contacted who completed the interview) for the 2006 BRFSS was 74·5 %.
Vitamin supplementation was assessed by asking participants whether they were currently taking any vitamin pills or supplements. Folic acid supplementation was assessed by asking participants whether and how often they were taking multivitamins or any vitamin pills/supplements that contained folic acid. Their responses were then categorised as (i) no folic acid supplements, (ii) taking folic acid supplements less than one time per d and (iii) taking folic acid supplements one to four times per d. Those who answered that they were taking folic acid supplements more than four times per d were excluded from the analyses because there were few participants in this category (<0·5 %). Self-reported supplement use of vitamins and minerals has been highly reliable and valid(Reference Satia-Abouta, Patterson and King30).
Participants’ CVD status was assessed by asking them whether they had ever been told by a doctor or other health professional that they had CHD (including heart attack or myocardial infarction, or angina pectoris) or stroke. Self-reports of physician-diagnosed CHD and stroke have an adequate validity and reliability compared with those obtained through medical records(Reference Bush, Miller and Golden31–Reference Okura, Urban and Mahoney32).
The demographic variables in our analyses included age, sex, BMI (kg/m2; self-reported), race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic and others), education (below high school diploma, high school graduate, some college or technical school and above college graduate) and household income (<$25 000, $25 000 to <$50 000 and ≥$50 000). In addition, participants’ status concerning smoking and alcohol consumption was included as a covariate in our analyses. For smoking status, participants were categorised as (i) current smokers (participants who had smoked 100 or more cigarettes during their lifetime and were still smoking), (ii) former smokers (participants who had smoked 100 or more cigarettes in their entire life but stopped) and (iii) never smoking (participants who had smoked less than 100 cigarettes during their lifetime). Heavy drinkers were defined as women who had one or more drinks per d or men who had two or more drinks per d.
Statistical analyses
After excluding from analyses the participants who responded ‘don’t know/not sure’, refused to answer or had missing responses for any of the study variables, a total of 41 792 participants remained in our analyses. The prevalence estimates for taking folic acid or vitamin supplements by CHD or stroke were age-adjusted to the standard 2000 US population. Logistic regression analyses were conducted to estimate the adjusted prevalence (AP; predicted marginal probability) of taking folic acid or vitamin supplements and to estimate the adjusted OR (AOR) with 95 % CI to assess associations of taking folic acid or vitamin supplements with CVD after adjustment for covariates. We used the Survey Data Analysis (SUDAAN) statistical software package release 9·0 (Research Triangle Institute, Research Triangle Park, NC, USA) to account for the multistage, disproportionate stratified sampling design.
Results
Of all participants, a total of 5445 (12·3 % (weighted), 2756 men (15·2 %) and 2689 women (9·6 %)) had CHD and 2076 (4·4 % (weighted), 867 men (4·2 %) and 1209 women (4·5 %)) had stroke. The characteristics of participants with or without these conditions varied significantly by sociodemographics (Table 1). Overall, those who reported having CHD or stroke were more likely to be older and to have a lower household income. Significantly more men than women reported having CHD (P < 0·001), but not stroke (P = 0·238). In addition, the percentages of adults with CHD or stroke differed significantly by participants’ educational and BMI levels. A significant racial/ethnic difference existed among participants who reported having stroke (Table 1, P < 0·05).
Table 1 Sociodemographic characteristics of study population by CHD and stroke, BRFSS, 2006
BRFSS, Behavioral Risk Factor Surveillance System.
*χ 2 tests.
Overall, 3·9 % of men and 5·0 % of women reported taking folic acid supplements less than one time per d, 49·2 % of men and 58·4 % of women reported taking folic acid supplements one to four times per d and 58·8 % of men and 70·2 % of women reported taking vitamin supplements (Table 2). By states (and US territories), the prevalence of taking folic acid supplements ranged from 1·5 % in Kentucky to 6·9 % in Puerto Rico for les than one time per d and from 38·2 % in Puerto Rico to 60·3 % in Montana for one to four times per d; the prevalence of taking vitamin supplements ranged from 52·7 % in Puerto Rico to 73·0 % in Montana.
Table 2 Age-adjusted prevalence of taking any vitamins or folic acid-containing vitamins among US men and women aged ≥45 years, by selected characteristics, BRFSS, 2006
BRFSS, Behavioral Risk Factor Surveillance System.
*,†P < 0·1, ‡P < 0·01, §P < 0·001, ∥P < 0·05 and for comparisons between adults with CVD and those without.
**,‡P < 0·01, ∥P < 0·05 and for comparisons between non-Hispanic whites and other racial/ethnic groups.
***,§P < 0·001 and ∥P < 0·05 for comparisons between current smokers and former smokers or those who have never smoked.
****,∥P < 0·05 for comparisons between adults who drink heavily and those who do not.
Women with CHD had a significantly lower prevalence (age-adjusted) of taking folic acid supplements or taking vitamin supplements than women without CHD (P < 0·01 for all comparisons). Among both men and women, the prevalence of taking folic acid supplements one to four times per d or taking vitamin supplements was significantly lower in those with stroke than in those without (P < 0·05 for both). In addition, in both men and women, regardless of their disease status, the percentages of adults who were taking folic acid supplements one to four times per d or taking vitamin supplements increased significantly with increases in age, educational levels and household income, but decreased when BMI levels increased. Non-Hispanic whites had the highest and current smokers had the lowest prevalence of taking folic acid supplements one to four times per d or taking vitamin supplements among the respective characteristics. Women who were heavy drinkers had a significantly higher prevalence of taking folic acid (one to four times per d) or taking vitamin supplements than women who were not heavy drinkers (Table 2).
Multivariate analyses showed a significant gender interaction on the association of taking folic acid or vitamin supplements with CHD (P < 0·01), but not with stroke. After adjustment for age, race/ethnicity, education, BMI, household income, smoking and heavy alcohol consumption, the AP and the AOR for taking folic acid supplements one to four times per d among men with CHD were significantly higher than among those without (P < 0·05 for both, Table 3). In contrast, the AP and the AOR for taking folic acid supplements among women with CHD were significantly lower than among those without (P < 0·001 for both, Table 3). In addition, women with CHD were significantly less likely to take vitamin supplements than those without CHD (P < 0·001, Table 4). Adults with stroke were as likely to take folic acid or vitamin supplements as those without stroke after multivariate adjustment (Tables 3 and 4).
Table 3 Adjusted prevalence and AOR for taking folic acid-containing vitamins among US adults aged ≥45 years, by gender and by CHD or stroke, BRFSS, 2006
BRFSS, Behavioral Risk Factor Surveillance System; AOR, adjusted OR.
*Adjusted for age, race/ethnicity, education, BMI, income, smoking and heavy drinking.
†P < 0·1.
‡P < 0·01.
§P < 0·05.
∥P < 0·001 for comparisons between adults with CVD and those without CVD.
Table 4 Adjusted prevalence and AOR for taking any vitamins among US adults aged ≥45 years, by gender and by coronary heart disease or stroke, BRFSS, 2006
BRFSS, Behavioral Risk Factor Surveillance System; AOR, adjusted OR.
*Adjusted for age, race/ethnicity, education, BMI, income, smoking and heavy drinking.
†P < 0·001 for comparisons between adults with CVD and those without CVD.
Discussion
Our results from a large sample showed that 58·3 % of adults aged ≥45 years were taking folic acid supplements (4·4 % for taking less than one time per d and 53·9 % for taking one to four times per d) and 64·6 % were taking vitamin supplements regardless of their CVD status. When stratified by disease status, men with CHD were 17 % more likely than those without to take folic acid supplements one to four times per d after multivariate adjustments. However, women with CHD were 44 % and 32 % less likely to take folic acid supplements less than one time per d or one to four times per d, respectively, and 34 % less likely to take vitamin supplements than those without CHD. No significant differences were observed in taking folic acid or vitamin supplements between adults with stroke and those without stroke.
The early Vitamins and Lifestyle (VITAL) study reported that adults with coronary artery disease (heart attack, bypass surgery, angioplasty or angina) were 1·4 times as likely to use folic acid supplements as those without coronary artery disease(Reference Satia-Abouta, Kristal and Patterson33). The data from the 1999–2002 NHANES showed that adults with either coronary artery disease or stroke were 3·6 times as likely to use folic acid supplements as those without any of the following conditions: coronary artery disease, stroke, diabetes, hypertension or hypercholesterolaemia(Reference Buettner, Phillips and Davis34). However, in these studies, data analyses were not stratified by sex(Reference Satia-Abouta, Kristal and Patterson33, Reference Buettner, Phillips and Davis34) or by CVD status(Reference Buettner, Phillips and Davis34). Our study demonstrates a significant gender interaction when assessing the association of taking folic acid or vitamin supplements with CHD. We found that men with CHD were significantly more likely than those without CHD to take folic acid supplements one to four times per d after adjustment for potential confounding variables. However, women with CHD were significantly less likely than those without CHD to take folic acid supplements or vitamin supplements. Although the role of folic acid and multivitamin interventions in primary and secondary prevention of CHD remains to be elucidated, the potential impact of not taking folic acid and vitamin supplements on the health of women with CHD needs to be studied further. Moreover, several studies have reported that folic acid supplementation significantly reduces the risk of stroke in patients with CHD(Reference Bønaa, Njølstad and Ueland15, Reference Lonn, Yusuf and Arnold19, Reference Sepe, Adamo and Giuliano23, Reference Liem, Reynierse-Buitenwerf and Zwinderman35). Thus, our finding that men with CHD were significantly more likely to take folic acid supplements one to four times per d is of interest because this may help reduce risk of stroke in the future.
To date, studies have shown that a higher intake of folic acid from diet or supplements may help reduce risk of stroke and stroke-related mortality in primary prevention(Reference Bazzano, He and Ogden22, Reference Virtanen, Voutilainen and Happonen24, Reference Wang, Qin and Demirtas25). However, its role in preventing recurrent stroke and CHD and death in patients with a history of stroke remains unclear(Reference Lonn, Yusuf and Arnold19, Reference Toole, Malinow and Chambless36) and requires further investigation. Our study did not show any significant differences in the likelihood of taking folic acid or vitamin supplements between adults with stroke and those without.
In the United States, although most food is now fortified with folic acid or vitamins, dietary supplement use remains common(Reference Radimer, Bindewald and Hughes26). Our results also demonstrated a higher prevalence of taking vitamin supplements and a much higher prevalence of taking folic acid supplements among participants with CHD or stroke than reported previously. In the VITAL study, 9·2 % of adults aged 50–75 years with coronary artery disease and 41·9 % of adults with any CVD risk factors were taking dietary supplements; no data were reported for taking folic acid supplements among these patients(Reference Satia-Abouta, Kristal and Patterson33). In the NHANES 1999–2002, approximately 36 % of adults aged ≥40 years with either coronary artery disease or stroke were taking multivitamins, 8 % were taking B vitamins, and 5 % were taking folic acid as single supplements(Reference Buettner, Phillips and Davis34). The inconsistent results of these studies may be largely due to the varying measures of folic acid or vitamin supplementation used. Nonetheless, our results are consistent with earlier studies in that the prevalence of taking folic acid or vitamin supplements was higher in women v. men, in former smokers v. current smokers, in non-Hispanic whites v. other racial/ethnic groups, and in those with a higher educational level or a lower BMI(Reference Radimer, Bindewald and Hughes26, Reference Satia-Abouta, Kristal and Patterson33, Reference Buettner, Phillips and Davis34). The question as to whether physicians or other health-care professionals prescribed folic acid and vitamin supplements to their patients with CVD warrants further investigation.
At present, several mechanisms may help explain the possible beneficial effects of folic acid or vitamin interventions on CVD risk(Reference Moens, Vrints and Claeys37). First, folic acid or vitamin supplementation significantly reduces carotid intima-media thickness in patients with heart disease(Reference Fernandez-Miranda, Yebra and Aranda38, Reference Ntaios, Savopoulos and Hatzitolios39), a history of stroke or at risk for cerebral ischaemia(Reference Potter, Hankey and Green40, Reference Till, Rohl and Jentsch41); an increased carotid intima-media thickness serves as a predictor of coronary and cerebrovascular complications(Reference Poredos42). Second, the antioxidant capacities of folic acid and its metabolite 5-methyltetrahydrofolate or the antioxidants from multivitamins may reduce oxidative stress and superoxide production(Reference Antoniades, Shirodaria and Warrick43–Reference Shirodaria, Antoniades and Lee45). Moreover, a high-dose folic acid or vitamin therapy can lower blood pressure, enhance coronary dilation in patients with coronary artery disease(Reference Tawakol, Migrino and Aziz46) and improve endothelial function in patients with metabolic syndrome, CHD or diabetes(Reference Antoniades, Shirodaria and Warrick43, Reference Shirodaria, Antoniades and Lee45, Reference Doshi, McDowell and Moat47–Reference Title, Ur and Giddens49); these improvements are independent of homocysteine lowering. In addition, the use of multivitamins reduces C-reactive protein levels, a risk factor for CVD(Reference Church, Earnest and Wood50).
Study limitations and strengths
Our study was subjected to several limitations. First, all information including folic acid or vitamin supplementation and CVD status was self-reported, and thus subjectto recall bias. Second, although we have assessed the frequency of folic acid supplementation, the levels of folic acid and vitamin intake were unknown in the present study. In addition, data on the biomarkers of folic acid and vitamin supplementation were also unknown. Moreover, data on folic acid intake from fortified grain products were not collected in the study and therefore we were unable to adjust for this factor in our analyses. Finally, the use of medications for coronary artery disease or stroke has been shown to affect the behaviours of taking folic acid or vitamin supplements among these patients(Reference Buettner, Phillips and Davis34); however, in the present study, information on medications for CHD or stroke was not collected and therefore we were unable to conduct stratified analyses by medications. Despite these limitations, the results of the present study were strengthened by the large sample size from a population-based survey, and the analyses were stratified by sex and by diseases.
Conclusions
Among adults aged ≥45 years, substantial variations exist in the prevalence and likelihood of taking folic acid or vitamin supplements by gender and by CHD status, but not by stroke status. Men with CHD had a significantly increased likelihood of taking folic acid supplements; whereas women with CHD had a significantly decreased likelihood of take folic acid or vitamin supplements. Our results provide fundamental knowledge for clinical physicians, nutritionists and dietitians on the behaviours of taking folic acid and vitamin supplementation among patients with CHD or stroke.
Acknowledgements
The authors thank the BRFSS coordinators for their work on the survey. Disclaimer: The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. Sources of funding: This research received no specific grant from any funding agency. Authors’ contributions: Each author has contributed significantly to this work. G.Z. obtained the data from BRFSS, conducted the data analyses, interpreted the data and prepared the paper. E.S.F. supervised the data analyses and contributed to the writing of the manuscript. E.S.F, C.L. and A.H.M. revised the paper for important intellectual content. All authors critically reviewed and approved the final version of the paper. Conflict of interest declaration: The authors have no financial support or potential conflicts of interest to disclose.
