Validation of a 24-hour Caffeine Intake Recall

The objective of this study was to evaluate the accuracy of the CIR-24 compared to caffeine concentration biomarkers in urine and a caffeinated beverage intake frequency screener (CBQ) among a young adult population.

A 24-hour Caffeine Intake Recall (CIR-24) was developed to estimate caffeine consumption in the previous 24 hours. The online, self-administered CIR-24 was developed to measure dietary intake of caffeine from food, beverages, and supplements. The tool was based on the Automated Self-Administered 24-hour Dietary Assessment Tool (ASA24), a web-based, self-administered tool for collecting dietary recalls and records. ASA24 uses a multiple-pass method adapted from the Automated Multiple-Pass Method (AMPM) used in NHANES in the US and the Canadian Community Health Survey in Canada. The same general question structure was used in this custom online survey, but the items queried were limited to foods, beverages and supplements that may contain caffeine. The tool was created for the Canadian context by listing only Canadian products, changing container sizes to reflect the Canadian marketplace, and including a French version. The food, beverage, and supplement categories and sub-categories can be found in Supplemental Table 1. Specific items listed were based on ASA24 and the Canadian Nutrient File, with additional items identified through the USDA National Nutrient Databases and internet searches. Respondents were asked about the consumption of items from each of four main categories (beverages, foods, energy products with added caffeine, and supplements) in the prior day. Based on each food or beverage reported, the participant receives tailored probes based on their responses, to identify the specific items that may have contained caffeine (e.g., if carbonated beverage consumption was reported, the participant is asked to specify the type of carbonated beverage and brand consumed). Amounts of each item consumed were ascertained based on the particular category: food amounts were estimated by the number of items/pieces consumed, or by a volume amount (depending on the item); beverage amounts were estimated using images of container types and sizes (adapted from the ASA24) (or volume for powders, liquids, and concentrates); and supplements were estimated by unit (or volume for powders and liquids). The CIR-24 tool is available at http://davidhammond.ca/wp-content/uploads/2015/11/2014-CED-Technical-Report.pdf.A database was compiled of the caffeine content of all food and beverage items in the Canadian Nutrient File listed as containing caffeine (with some additional items from the USDA National Nutrient Database for Standard Reference), plus energy drinks, shots, and products (sourced from the nutrition information on products purchased in previous studies, as well as internet searches), and supplements listed as containing caffeine in Health Canada's Licensed Natural Health Products Database (http://www.hc-sc.gc.ca/dhp-mps/prodnatur/applications/licen-prod/lnhpd-bdpsnh-eng.php) or identified through an online pharmacy (www.well.ca). More precise and brand-specific values were added for popular products such as coffee, using information provided by manufacturers (e.g., Starbucks, Tim Horton's, McDonald's) and/or websites that aggregate nutrition information (http://www.cspinet.org/new/cafchart.htm).

A paper version of a caffeinated beverage questionnaire (CBQ) from the Fred Hutchison Cancer Centre, adapted slightly for use in Canada (e.g., mL measures were included in addition to fluid ounces), was also administered.The CBQ included 13 beverage categories, including caffeinated and decaffeinated coffee (brewed and instant) and tea, energy drinks and highly caffeinated sodas, regular colas and root beer, and caffeine-free colas and root beer. Nine frequency categories were included: Never or less than once per month; 1-3 per month; 1 per week; 2-4 per week; 5-6 per week; 1 per day; 2-3 per day; 4-5 per day; 6+ per day. A reference for a 'medium' size for each type of beverage category was provided; respondents stated whether each drink size was small, medium or large in comparison to the reference amount. Responses to the CBQ were used to calculate average daily caffeine consumption for each respondent. First, the number of annual servings consumed for each questionnaire item was calculated by multiplying the reported frequency by the reported portion size. A 'small' size was multiplied by a serving ratio of 0.5, and a 'large' size was multiplied by 1.5. Next, the average annual volume of each beverage consumed was divided by 365 to estimate the daily intake of each beverage. The caffeine database that accompanies the tool was used to calculate the amount of caffeine consumed in milligrams. In cases in which frequency data for items were missing, we assumed that none was consumed, and when a frequency was entered but no portion size was indicated, the 'medium' size was assumed.

For each participant, all urine samples were combined, and the total volume of the entire 24-hour sample was recorded. A sample from each participant was aliquoted into a 2 mL cryovial, and immediately frozen at -80°C. Samples were moved for 8 days to a -20°C freezer due to logistical issues, and were later returned to -80°C. At the end of the study period, all samples were transported in insulated containers with dry ice to the CDC Nutritional Biomarkers Branch in Atlanta, GA for analysis.

Urine concentrations for caffeine (1,3,7-trimethylxanthine) and 8 caffeine metabolites (1,7-dimethylxanthine (Paraxanthine or 17X); 1,3-dimethylxanthine (theophylline, 13X); 1,3,7-trimethyluric acid (137U); 1,3-dimethyluric acid (13U); 1,7-dimethyluric acid (17U); 1-methyluric acid (1U); 1-methylxanthine (1X); and 5-acetylamino-6-amino-3-methyluracil (AAMU)) were quantified by use of a high performance liquid chromatography-polarity switching electrospray ionization-tandem quadrupole mass spectrometry with stable isotope labeled internal standards based on a method reported previously.(26) Existing studies indicate that these metabolites have moderate correlation with caffeine intake (14). The limits of detection were: 0.05 μmol/L for 1U; 0.01 μmol/L for theophylline; 0.02 μmol/L for 13U and 17U; 0.005 μmol/L 137U; 0.03 μmol/L for 1X; 0.003 mmol/L for caffeine; 0.006 μmol/L for 17X; and, 0.1 μmol/L for AAMU. Samples that had amounts below the limits of detection were excluded for that particular metabolite.