Skip directly to search Skip directly to A to Z list Skip directly to navigation Skip directly to site content Skip directly to page options
CDC Home

NHANES Dietary Web Tutorial

Information About Dietary Variables

As described in the Locate Variables module, NHANES dietary, food frequency, and supplement variables and information about the variables, are stored in different data and documentation files. The Locate Variables module provides lists of variables that are commonly included in dietary analyses. This page provides additional information on these variables.  Please note, however, that the most complete descriptions of these variables can be found in the Analytic Notes section of the “Docs” files.   

Dietary Recall Variables

Dietary Recall Status (dr1drsstz, dr2drstz )

The dietary recall status variable is used to indicate the quality and completeness of a survey participant’s response to the dietary recall.  This variable is important in identifying all individuals who should be included in a study.  Four possible values are provided.  However, only two of the values (1 and 4) are included in Individual Foods File records.

The values for this variable are as follows:

1 = Reliable and met the minimum criteria.  A value of 1 for this variable in the Total Nutrient Intakes File indicates that all relevant variables associated with the 24-hour recall contain a value.  Individuals who reported fasting are considered to have a reliable recall. 

2 = Not reliable or did not meet the minimum criteria.  Data on individual food consumption are not provided for these cases.  These individuals do not have any valid records in the Individual Foods File nor do they have data on total nutrient intakes in the Total Nutrient Intakes File.  These records should not be included in nutrient intake estimates.

4 = Reported consuming breast milk.  For breast-fed infants, the Individual Foods File includes a record for each report of human milk.  Because amounts of human milk are not quantified, these records contain missing values for the amount consumed and for the amounts of energy and nutrients from human milk.  Total nutrient intakes are not provided because of this missing information.  Breast-fed children are excluded from most analyses for the reasons stated above.

5 = Not done.  This code is assigned when the dietary recall did not take place due to various reasons.  These individuals have no records in the Individual Foods File.

 

Breast-fed infant (either day) (drabf )

Beginning with NHANES 2003-2004, the variable drabf is included to identify breast-fed children.  This variable eliminates the need to review and identify individuals with human milk records in the Individual Foods and Total Nutrient Intakes Files. This variable has a code of 1 if a child consumed breast milk on either day of the recall.   

 

Intake day of week (dr1day, dr2day )

This variable identifies the day that reported foods were consumed.  By using this variable, intakes and eating patterns can be compared according to day of week or by weekend days versus week days.

 

Food energy and nutrient variables (many)

Food energy and nutrient values in the Individual Foods File represent the amount of energy or nutrients provided by consumed food.  In the Total

Nutrient Intakes File, these values represent the total daily amount of nutrients consumed for each survey participant.  These totals were derived by summing the nutrient amounts from all foods listed in the Individual Foods File for a survey participant.

 

Variables in Individual Foods File only

Food/individual component number (dr1iline, dr2iline )

The Individual Foods File contains one record for each food consumed by a participant.  Each food record is assigned a line number.  The dr1iline or dr2iline variable may be used along with the respondent sequence number to uniquely identify a food record.  An example is when a respondent reports the same food more than once in a day, as shown in the table below.

 

SEQ N dr1iline Food
20000 1 Banana
20000 2 Milk
20000 3 Coffee
20000 4 Milk

                                             

USDA food code (dr1fdcd, dr2fdcd)

Every food in the Individual Foods File is identified by an 8-digit USDA food code.  Each food code is associated with a food description. There are two versions of each description: a complete, 200 character version and an abbreviated 60 character version.  Starting with the 2003-2004 release cycle, a Food Codes file that accompanies the Individual Foods File contains both the long and short descriptions.  An appendix in the file documentation provides SAS code to link the descriptions in the Food Codes File with the USDA food code in the Individual Foods File.

Beginning with the 2001-2002 release, the source of the food codes is USDA’s Food and Nutrient Database for Dietary Studies (FNDDS). 

 

Modification code (dr1mc, dr2mc )

During the food coding process, predefined recipes for some food mixtures are modified to more closely match the food as described by the respondent.  Nutrients are modified by substituting ingredients in a predefined recipe for the mixtures.  An example of a modified recipe is an egg fried in butter instead of margarine.  Each modification code is assigned a unique 6-digit identification number.

Beginning with the 2003-2004 release, the Modification Code file that accompanies the Individual Foods File includes descriptions (up to 200 characters) associated with each modification code.  An appendix contained in the file documentation provides SAS code to link the descriptions in the Modifications Codes File with the modification code in the Individual Foods File

                                  

Combination food number (dr1ccmnm, dr2ccmnm ) and combination food type ( dr1ccmtx, dr2ccmtx )

The combination food number and combination food type variables can be important in planning analyses, especially when accounting for mixed food items such as sandwiches, salads, or other mixed dishes or beverages.  Some foods reported by respondents are coded using more than one food variable, resulting in more than one record for the food.  If a food is coded as more than one item, it is called a combination.  Combinations may represent one food added to another and consumed as a unit, such as sugar added to coffee, margarine added to toast or gravy on potatoes, or components (ingredients) of foods, such as sandwiches, salads, and mixed dishes.

All food records in a combination have the same values for the combination food number and combination food type variables – the combination food number links a group of foods that are eaten together as a unit and the combination food type identifies the type of combination.

Approximately 53% of all foods in the 2003-2004 dietary intake data were coded as combination type foods.

information icon Combination food types assigned to codes for the 1999-2000 dietary data release cycle are not the same as those for latter cycles.  For example, a food code of “1” in the 1999-2000 dataset represents a baby food combination; in the 2001-2002 and 2003-2004 datasets, “1” represents a beverage with additions.

 

 

Time of eating occasion (dr1_020, dr2_020 ) and name of eating occasion ( dr1_030Z, dr2_030Z )

Each food record includes the variables denoting time and name of eating occasion at which the food was consumed.  Note that the time of eating occasion variable is displayed in the 24-hour notation HH:MM.  Additionally, codes assigned to the same eating occasion may differ from one cycle to another.

The name of eating occasion variable may include more than one relevant code for the same eating occasion, due to the inclusion of both English and Spanish terms.  For example, three separate listings are provided for the breakfast eating occasion—breakfast, desayuno, and almuerzo. 

It may be useful to look at time of eating occasion when conducting an analysis because eating occasions may be reported at any time of day.  For example, the midday meal may be called lunch or dinner, and the evening meal may be called supper or dinner.    

Extended consumption is used to denote the eating occasion when a respondent reports a food consumed over an extended period of time without distinct periods of consumption. 

 

Source of food (dr1fs, dr2fs )

The source of food variable identifies where a food was obtained, such as from a store, fast food restaurant, vending machine or school cafeteria. This variable was first included with the 2003-2004 release.

   

Variables in Total Nutrient Intakes File only

Compare food consumed yesterday to usual (dr1_300, dr2_300 )

Each respondent was asked to assess whether the amount of food eaten on the recall day was usual, much more than usual, or much less than usual.

 

Water intake variables (dr1_320, dr2_320, dr1_330, dr2_330, dr1bwatr, dr2bwatr, dr1cwatr, dr2cwatr, dr1tws, dr2tws )

Data on intake of drinking water (i.e., tap water, plain bottled water, and unsweetened carbonated water) are generally collected after the 24-hour dietary recall is completed.  Drinking water data are collected during the 24-hour recall only when it is an addition to other foods or is an ingredient in mixed beverages. 

The availability of water variables varies with the release cycle.  For the 2003-2004 release, four variables—plain, tap, bottled, and carbonated—were included.

 

 Special diet variables (drqsdiet, drqsdt1 – drqsdt8, drqsdt91 )

Beginning with the 2003-2004 release, information is included on whether the participant is currently on any kind of diet to lose weight or for another health-related reason and if so, the type of diet.  The variable drqsdiet identifies whether a participant is on a special diet.  The variables drqsdt1 through drqsdt8 and drqsdt91 identify the type of diet(s) the participant follows.  Responses to the type of diet were collected as “code all that apply,” meaning that more than one type of diet could be reported. 

 

Variables in both Individual Foods and Total Nutrient Intakes Files

Respondent sequence number (seqn )

Each survey participant is identified by a 5-digit respondent sequence number.  The seqn variable is present in most NHANES files and can be used to link data from different files for a given respondent. 

 

Number of days of intake (drdint )

Beginning with the 2003-2004 release, the variable drdint is included to indicate the number of days of intake available for each participant.  This variable eliminates the need to review records and identify individuals in the Individual Foods and Total Nutrient Intakes Files with 1 or 2 days of intake.

 

Dietary Supplement Variables

Any dietary supplements taken (dsd010 )   

This is a derived variable.  Participants who took a product coded as a supplement in the past 30 days are coded as 1.  Supplements that were mistakenly recorded as an antacid in the medication section (RXQ) or as a prescription medication are moved to the DSQ section and are counted as supplements.  Participants who reported an antacid containing calcium or magnesium in the past 30 days that was recorded only in RXQ antacid section and did not take any dietary supplement are coded as 2.  Participants who did not take any product determined to be a dietary supplement in the past 30 days are also coded as 2.  Prescription medicines and analgesics that are mistakenly recorded in the DSQ section are moved to their appropriate sections for data release.

 

Total number of supplements taken (dsdcount )  

This is the count given to all supplements reported in the dietary supplement (DSQ) or medications (RXQ) section, as well as antacids that were reported in the DSQ section.  Antacids recorded in the antacid section do not contribute to this count.  Non-supplements that were reported in the DSQ section do not count.  Products with don’t know or refused responses, are counted as supplements because no clear evidence exists that they are not and they were reported by the participant as a supplement.

 

Any non-prescription antacids taken (dsd010an )

Indicates if any non-prescription antacids were taken. 

 

Total number of non-prescription antacids taken (DSDANCNT)

Total number of non-prescription antacids taken per respondent. 

 

Supplement ID Number (DSDSUPID) 

These are numbers assigned by the database for each product entered.  Supplement ID numbers are 10 digits long and all Supplement IDs begin with the number ‘1’. The next 3 digits (positions 2-4) are ‘888’ if the supplement was created by NCHS as a generic or default product; otherwise the digits in positions 2-4 are coded ‘000’.  The next 4 digits (positions 5-8) are assigned by the database and do not indicate anything about the product.  The last 2 digits (positions 9-10) indicate formulations of the same supplement:  the first formulation entered into the database = 00, the first reformulation = 01, the next = 02, etc.  Note that these are reformulations of the same product: different versions (e.g. liquid vs. tablet, with iron vs. without iron, regular vs. high potency) have different 4 digit numbers (positions 5-8).   When a product name was entered as “refused” or “don’t know”, the ID number is a string of 7’s or 9’s.

 

Name of Supplement (DSDSUPP)

This is the name from the supplement front label which is entered into the database. Matching what supplement the interviewers record to an actual product label is made with varying degrees of certainty. When no match could be made, then the product was given a match of 6, “unknown.” For these matches, the 10 digit supplement ID begins with a 6 and the phrase “no product information” is assigned as the name. Products with brand names that are available only in a limited region of the country are released with a generic name, not a brand name, to ensure participant confidentiality. Product names that were entered as “refused” or “don’t know” are named “7777” and “9999”, respectively.

For some entries made by interviewers, no corresponding product label could be found nor could a reasonable default product be assigned. These entries are counted as supplements because no evidence that they are not supplements, but only the words “no product information available” are used in place of a name in this public data release.

 

Was container seen (DSD010)

This variable denotes whether the dietary supplement container was seen during the household interview. Containers are seen approximately 88% of the time.  This enables interviewers to collect more complete and accurate supplement names compared to supplement containers that are not seen.  For containers that are not seen, the interviewer relies on the participants’ memory for the name of the supplement taken.

 

Matching code (DSDMTCH)

Supplements are recorded in the household interview with varying degrees of accuracy and completeness.  The matching code indicates the level of certainty with which a supplement recorded during the interview matches the actual supplement label. 

 

How Long Supplement Taken (DSD090)

This variable is released in terms of days that a supplement has been taken by multiplying years by 365; months by 30.4; and weeks by 7.

 

Days Supplement Taken, Past 30 Days (DSD103)

This variable is released as recorded by interviewer.  This information is missing for dietary supplement data that were recorded in the medication (RQX) section. 

 

Antacid reported as a dietary supplement (DSDANTA)

Indicates whether an antacid was reported in the dietary supplement (DSQ) or antacid section.  For a few participants, the same antacid was recorded in both questionnaire sections.  In these instances, the antacid was considered to be in the DSQ section and coded as 1. 

 

Quantity and dosage form of supplement taken daily (DSD122Q)

This variable is released as recorded by interviewer.  This information is missing for dietary supplement data that were recorded in the medication (RXQ) section. 

 

Supplement Information Source (DSDRCE) 

The source of each product label is recorded in the database.  These source codes are listed in Appendix 5 of the supplement file documentation. Generic and default products do not have a source code. 

 

Formulation Type (DSDTYPE)

The type of formula is recorded into the database.  These codes are listed in Appendix 7 of the supplement file documentation.

 

Serving Size Quantity (DSDSERVQ)

This is thee product dosage quantity, which is recorded from the product label supplements facts panel.  When calculating the amount of a nutrient consumed from supplements, it is important to take serving size into consideration.  For some supplements, the serving size may be more than one unit (e.g., tablet, drop, teaspoon).  In addition, the ingredient listed may be a compound (e.g. calcium carbonate), and the amount of the elemental nutrient (e.g., calcium) needs to be calculated.  Appendix 5 of the supplement file documentation contains suggestions for conversions, but analysts are advised to confirm these.

 

Serving Size Unit (DSDSERVU)

This is the serving size unit, which is recorded from the product label supplements facts panel.  The codes are listed in Appendix 10 of the supplement file documentation.

 

Alternative Serving Size (DSDSERVA)

This is listed in labels for some products.  Not all products offer an alternative serving size. 

Label may include alternative serving size (e.g., 1 dropperful = 1 mL).

 

Ingredient ID (SDSDINGID)

This is the ingredient ID created by the NCHS database for each ingredient recorded from the product label supplements facts panel.

 

Ingredient name (DSDINGR)

Ingredient names are recorded from the product label supplements facts panel.

 

Ingredient operator (DSDOPER)

This is a symbol (=, < , or >)  that comes from the product label supplements facts panel.

 

Ingredient quantity (DSDQTY)

Ingredient quantity is recorded for each ingredient listed from the product label supplements facts panel. 

 

Ingredient unit (DSDUNIT)

Ingredient unit is recorded for each ingredient listed from the product label supplements facts panel.  Appendix 11 of the supplement file documentation lists the code for these. 

 

Ingredient category (DSSDCAT)

NCHS assigns ingredient categories: Vitamin, Mineral, Botanical, Others, Amino Acid.  Appendix 8 of the supplement file documentation lists the code for these variables.

 

Number of ingredient categories in each supplement (DSDCNT)

For each supplement, this variable is the number (count) of ingredients in each ingredient category (vitamin, mineral, amino acid, botanical, other) listed in the facts box on the label, including ingredients listed within blends. 

Number of vitamins in the product (DSDCNTV)

Number of minerals in the product (DSDCNTMdsdcntm)

Number of amino acids in the product (DSDCNTA)

Number of botanicals in the product (DSDCNTB)

Number of other ingredients in the product (DSDCNTO)

 

Blend Flag (DSDBLFLAG)

This indicator variable denotes whether an ingredient is a blend or not a blend.   

 

Blend component ID (DSDBCLID)

These are ingredient ID numbers for blend ingredients.

 

Blend component name (DSDBCNAM)

These variables are the ingredient names for blend ingredients.  Blends in products will not give the actual breakdown of ingredient quantities in the blend.  The ingredients will usually just be listed, and most of the time a whole blend amount is given. 

 

NCHS assigns ingredient categories for each blend ingredient: Vitamin, Mineral, Botanical, Others, Amino Acid.  Appendix 8 of the supplement file documentation lists the code for these variables.

 

Contact Us:
  • National Center for Health Statistics
    3311 Toledo Rd
    Hyattsville, MD 20782
  • 1 (800) 232-4636
  • cdcinfo@cdc.gov
USA.gov: The U.S. Government's Official Web PortalDepartment of Health and Human Services
Centers for Disease Control and Prevention   1600 Clifton Rd. Atlanta, GA 30333, USA
800-CDC-INFO (800-232-4636) TTY: (888) 232-6348 - Contact CDC–INFO
A-Z Index
  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #