In 1987, interviewers asked, "During the past 12 months, this is, since [12 month date] a year ago, did you take any multiple vitamins?"

A similar question was asked in 1992. The 1992 Field Representative's Manual provided this general instruction to interviewers, not found in other years:

Regarding what counts as a multi-vitamin, the 1992 Manual tells interviewers to:

A similar direction is found in the 1987, 2000, and 2005 Manuals.

In 2000 and 2005, interviewers asked, "During the past 12 months, did you take any multi-vitamins such as One-A-Day, Theragran, or Centrum, etc.?" Interviewers were instructed, "If multi-vitamins were already mentioned, enter '1' for yes without asking." Additionally, interviewers were told to "read if necessary" the following clarifying information:

The 2000 Manual also urged the interviewer to examine the respondent's bottle of vitamins and review the ingredients in case of doubt. Specifically:

In 2002, respondents were asked, "During the past 12 months, did you use high dose or megavitamin therapy for your own health or treatment?" Those with an affirmative response (i.e., a "Yes" response in VITYR) were shown a flash card containing a list of vitamins and minerals. They were asked, "During the past 12 months, did you take any of the following vitamins in high dose? You may choose more than one."

The other such supplements, along with the years these variables are available, are:

• Beta carotene (VITABETA, 2005)

• Calcium (VITACALC, 1987, 1992, 2000, 2002, 2005)

• Vitamin A (only) (VITAMINA, 1987, 1992, 2000)

• Vitamin C (VITAMINC, 1987, 1992, 2000, 2002)

• Vitamin E (VITAMINE, 1987, 1992, 2000, 2002, 2005)

• Selenium (VITASELN, 2002, 2005)

• Coenzyme Q-10 (VITYCOEN, 2002)

• DHEA (VITYDHE, 2002)

• Vitamins A and/or D (VITYVITAD, 2002)

• Vitamin B complex (VITYVITB, 2002)

• Zinc (VITYZINC, 2002)

In 1987, 1992, 2000, and 2005, respondents who gave an affirmative response to the question associated with VITAMULTV were asked to report the number of months they had taken multi-vitamins (VITAMULTVMO).

In 1987 only, respondents were asked to report the number of multi-vitamin pills they took per day (VITYMULTVPIL).

In 2000 and 2005, respondents who had taken a multi-vitamin during at least one month in the past 12 months were asked how often they had taken multi-vitamins in those months (VITAMULTVNO), responding as desired in terms of days per week or days per month (VITAMULTVTP).

For 2000 (but not 2005), the National Center for Health Statistics recoded data from VITAMULTVNO and VITAMULTVTP to create two new variables with consistent time units: number of days per week took a multi-vitamin during the relevant months of the past 12 months (VITAMULTVDWK), and number of days per month took a multi-vitamin during the relevant months of the past 12 months (VITAMULTVDMO). VITAMULTVDMO is also available for 1987 and 1992, when this information was collected through a direct question (i.e., "About how many days per month did you take multi-vitamins?").

For more information on the full range of variables dealing with vitamins, see VITANY and VITYR.

Respondents who were only taking an herbal supplement or who were only taking a combination of 2 vitamins or minerals may thus have been less likely to be coded with an affirmative response for VITAMULTV in 2000 than in 2005. The effect of this difference in the instructions to interviewers was probably modest, however, since, in both 2000 and 2005, the survey form reminded respondents not to count herbal supplements or combinations of just 2 vitamins or minerals as multi-vitamin supplements.

The 2002 Field Representative's Manual defines such therapy as "the use of vitamins in excess of the Recommended Daily Allowances (RDA) established by the National Academy of Sciences, Food and Nutrition Board."

Thus, an affirmative response to VITAMULTV in 2002 indicates that the sample adult took multi-vitamins in high doses. A negative response to VITAMULTV in 2002 indicates that the respondent did not take high doses of multi-vitamins, but does not indicate that he/she took no multi-vitamins at all in the past 12 months. By contrast, affirmative responses to VITAMULTV in 1987, 1992, 2000, or 2005 indicate that the respondent took any amount of multi-vitamins in the past 12 months.