The data

View the most recent version.

This study is based on results of cycle 1 of the Canadian Health Measures Survey (CHMS), which collected data from a nationally representative sample of the household population aged 6 to 79. Residents of Indian Reserves, Crown lands, institutions and certain remote regions, and full-time members of the Canadian Forces, were excluded. Data were collected at 15 sites across the country from March 2007 through February 2009.

A Statistics Canada interviewer administered a detailed in-home health questionnaire, which covered medication and natural health product use. One day to six weeks later, respondents visited a mobile examination centre for a series of physical measurements, including blood samples taken by a certified phlebotomist to measure a variety of analytes.⁷

Of the 8,772 households selected for the CHMS, 69.6% agreed to participate; 88.3% of them responded to the in-home survey, and of those, 84.9% visited the mobile examination centre. The overall response rate was 51.7%. A comprehensive consent process was employed; participation was voluntary and respondents could opt out of any part of the survey at any time.⁸ The final sample consisted of 5,604 respondents aged 6 to 79 and is representative of approximately 96.3% of the Canadian population. Details about the CHMS are available elsewhere.^7,9

Drug identification and natural health product numbers were collected from respondents during the in-home interview. This information was verified at the mobile examination centre visit, and changes in drug and supplement use were recorded.¹⁰ Consumption of folic acid supplements—alone or in a multi-vitamin—in the 30 days before the mobile examination centre visit was determined by matching drug identification and natural health product numbers to product information extracted from the Health Canada Drug Product and Licensed Natural Health Product databases.^10,11.12 This approach was also used to quantify vitamin B₁₂ supplement use.

Dietary reference intake age groups were used for this study. These age groups were collapsed into three categories to achieve adequate sample size in the analyses examining doses of folic acid-containing supplements. Socio-economic status was measured by per person household income equivalents (which grouped respondent household income into quartiles after adjusting for household size and composition) and highest level of household education (less than postsecondary graduation and postsecondary graduation).¹³Immigrant status was defined as born in Canada or not born in Canada.

A brief, non-quantitative questionnaire was used to examine usual frequency of consumption from the grains and fruit and vegetables food groups. Cereal, white bread, brown bread and pasta were combined to formulate the derived variable for grains.  Self-reported smoking status and history were used to categorize respondents as daily, former or never smokers. A physical activity index of active, moderately active or inactive was derived based on average daily energy expenditure values during self-reported leisure time activities in the three months before the in-home interview.¹⁴

Measured height and weight were used to calculate body mass index (wt(kg)/ht(m)²). The body mass index of adults, excluding pregnant women, was classified using Health Canada's guidelines.¹⁵ Children aged 6 to 17 were classified as being normal weight, overweight or obese based on definitions proposed by the International Obesity Task Force, which account for both the age and sex of the child.¹⁶

A clinical risk factor variable was created to combine self-reported diabetes mellitus, use of folic acid antagonist medication, and overweight/obesity. Folic acid antagonist medication was determined by matching drug identification numbers—from medications used in the 30 days before the mobile examination centre visit—to information in the Drug Product Database.^11,17 The variable identified three parameters: no risk factors, one risk factor, and more than one risk factor.

Blood was taken from 5,373 CHMS respondents. Red blood cell folate was analyzed using the Immulite 2000 immunoassay (Siemens Canada Ltd., Mississauga).¹⁸Red blood cell folate concentration was calculated from the measured whole-blood folate concentration adjusting for red blood cell volume, without correction for plasma folate concentration. Serum vitamin B₁₂ was also assessed using the Immulite 2000 immunoassay, a solid phase, competitive chemiluminescent enzyme immunoassay involving an automated alkaline denaturation procedure. The cut-off for marginal Vitamin B₁₂ status was 221 pmol/L or less.^2,19 The Vitros 5, 1FS (Ortho Clinical Diagnostics, Markham) was used to assess plasma homocysteine concentrations. The quantitative measurement of plasma homocysteine was performed using the VITROS Chemistry Products Homocysteine Reagent, a spectrophotometric method involving three coupled enzymatic reactions (cystathionine synthase, cystathionine lyase and lactate dehydrogenase). Normal plasma homocysteine status was as 8 µmol/L or less for respondents aged 10 or younger, and 15 µmol/L or less for older respondents.^20-22

Descriptive statistics (frequencies, percentiles) were used to characterize the population. Missing values for predictor variables were removed for individual analyses. T-tests were used to study differences between estimates. With separate logistic regression analyses that controlled for age and household income, each co-variate was examined as a correlate of folic acid-containing supplement use. All estimates were based on data weighted to represent the Canadian population. Variance estimation (95% confidence intervals) and significance testing were based on the bootstrap technique to account for the complex CHMS sampling design.²³ Analyses were conducted in SAS 9.1.3 (SAS Institute Inc., Cary, NC) and SUDAAN v.10.0 (RTI International, Research Triangle Park, NC), using DDF=11 in the SUDAAN procedure statements. Given the 11 degrees of freedom available for variance estimation, Satterthwaite-adjusted statistics were used to test the significance of each regression model's coefficients.²⁴ Significance was defined as a p-value of <0.05.

The overall response rate to the CHMS was slightly above 50%. Although the survey weights ensured that the sample was representative of the target population, bias might exist if the use of folic acid-containing supplements by non-respondents and respondents differed systematically. The number of independent variables in the regression model was limited by 11 degrees of freedom; future research that combines subsequent CHMS cycles will increase the number of degrees of freedom and permit more complex modelling. The cross-sectional nature of the survey precludes inferences about the temporal ordering of events or causality. As well, the CHMS was not designed to collect information on usual supplement dosage, so precisely how often supplements were consumed could not be determined.