Understanding Ingredient Consumption Analysis

Understanding Ingredient Consumption Analysis
byRay A. Matulka

Oneof the keys to success is knowing your audience; this is as true for a publicspeaker as an ingredient supplier or manufacturer. Determining which foods the new ingredient will be added toand, by extension, the amount of new ingredient the average person will eat iscritical in determining both the sales potential and, more importantly, theamount that can be safely consumed. A detailed consumption analysis, based onnationally recognized food intake surveys, combined with a rigorous safetyassessment, allows the manufacturer and the regulatory agencies to determine theamount of new ingredient that could be safely consumed when added to a definedset of foods. Utilization of a consumption analysis is dependent on thepopulation surveyed and their consumption patterns, as well as understanding theinherent strengths and weaknesses of the type of survey that was used.

Estimating the amount of an ingredient consumed from thevarious sources of foods people eat is critical in establishing thesafety-in-use of food. It is not only important to consider all of the sourcesof exposure to the ingredient (e.g., if the ingredient occurs naturally in somefoods, can be consumed in the form of a dietary supplement, or as a direct foodadditive), but also to have the ability to accurately determine the quantity ofthe ingredient that is consumed by the U.S. population. For an accurate understanding of how much of an ingredientwill be consumed, one must use surveys to determine the number of people eatingthe food products and the quantity that each person may consume. However, time,finances and logistics do not realistically allow an ingredient manufacturer tosurvey every individual in an entire population. Therefore, we must rely on astatistical analysis of a defined representative sample to help us understandhow a larger population will act. In order to understand how an entirepopulation of interest, called a target population behaves, statisticiansconduct surveys to determine what foods a subset of the target population (a samplepopulation) will consume. The sample population should truly represent thetarget population, having the same relevant characteristics as the targetpopulation. In this way, generalizations from this sample population could bemade to the target population. If the sample population does not consume foodsimilar to the target population, bias will occur and the results of aconsumption analysis may be very different from what the population actuallyeats.

Once the target population and sample population are defined,the quantity of food consumed may be determined. Food, in this context, canrefer to either one or hundreds of discrete food items. Food consumption surveysallow manufacturers to obtain detailed data of the amount of individual fooditems consumed. These surveys are carefully constructed to minimize day-to-dayvariability. From this data, it is important to remove from the samplepopulation any individuals who do not eat any of the food that contains theingredient, thereby resulting in a further subdivision, an Eaters Onlypopulation. For a safety determination, regulatory agencies are only concernedwith the population that will be exposed to the ingredient. Includingrespondents in the statistical analysis who do not consume the ingredient wouldactually decrease the apparent consumption of the ingredient, when analyzed on aper person basis. This could potentially result in a hazardous situation.

To measure food consumption on a population level, varioussurveys are commonly used. These surveys include individual surveys, foodbalance sheets, total-diet studies, or household surveys. The manufacturershould have some understanding of the survey used, so that a specific questionis being accurately answered (for example, a manufacturer may want a marketanalysis of household consumption, rather than specific individual ingestion).For the safety related to individuals consuming the ingredient in question, afood consumption assessment method should measure some degree of variabilityexisting in the population. Household surveys contain inter-householdvariability, while individual surveys contain inter-individual variability. Asconsumption patterns among individuals provide the most information for theactual exposure distribution within a population, the food-based surveys mostcommonly used are based on individual responses, and therefore individualsurveys will be discussed here.

Methods commonly used by regulatory agencies, manufacturers,nutritionists and general researchers for assessing food consumption amongindividuals include 24- hour recalls, dietary records, food frequency recordsand dietary history accounts. As the name implies, respondents of 24-hour recallsurveys are asked questions in a personal interview concerning their food andbeverage consumption in the preceding 24 hours, or on the preceding day. Thismethod normally uses examples to help the respondent remember the foodsconsumed. Unfortunately, a single 24-hour recall survey does not provide anappropriate characterization of a respondents habitual food consumption;therefore, several repeat surveys must be employed for accurate consumptioninformation, increasing both the time and cost involved.

According to the dietary record method, respondents recordtheir actual intake of foods and beverages at the time of consumption. Theweight or volume of the food products consumed can be determined by measurementor through estimation of the weight of the food. Usually, the dietary record period lasts three to fourconsecutive days. Since quantitative and accurate information on foods andbeverages actually consumed is obtained at the time of consumption, dietaryrecords are considered the gold standard in the assessment of a persons dietat that point in his life. A shortfall with this type of analysis is thatunder-reporting may occur, which may result in an underestimation of usualdietary intake.

The food frequency method utilizes questions concerning theindividuals usual frequency of consumption (and sometimes portion size) of alist of foods and beverages. The burden to respondents of the food frequencyquestionnaire is low, and so this method is now commonly used to estimate usualdietary intake. Weaknesses of the food frequency method include incompletelisting of consumables, as well as an inability to assess cooking methods orrecipes. Dietary variability cannot be covered by a limited list. Food frequencyquestionnaires are better utilized for ranking subjects according to food consumptionor intake of a chemical or drug than for estimating the level of intake.Compensating with a longer food list tends to overestimate intake, while shorterlists underestimate individual intake.

The individuals past diet is investigated in the dietaryhistory method, in which many details about types of usually consumed foods,amounts consumed and frequencies of consumption are collected. Most dietaryhistory methods utilize a crosscheck technique, such as a food list, which iscrucial since the method requires respondents make several judgments concerningtheir usual food consumption and meal pattern. The need for crosschecks is moreapparent for individuals with an irregular eating pattern. Estimates based on dietary histories have been noted to behigher than estimates based on 24-hour recalls and dietary records (methodscovering a short period of time).

In estimating the consumers exposure to an ingredient,information is needed about the foods that potentially contain the specificingredient of interest. In theory, each of the four methods for individualconsumption estimation can be used for exposure estimation, as long as the foodsinvolved are covered in the questionnaire. One must weigh the pros and cons ofeach of the surveys available. As a food additive, the assessment of lifetimeexposure is of interest; therefore, food frequency and dietary history methodswould be the best choices. Dietary record assessments and 24-hour recalls wouldnot give an appropriate characterization of a respondents habitual foodconsumption. However, the estimation of exposure to food additives will benormally compared with safety or reference values.

In evaluating the toxicity of food additives, a commonly usedindication of safety for use is the comparison of the calculated exposureascertained from one of the survey methods with the Acceptable Daily Intake(ADI) values. The ADI is defined as an estimate of the amount of a foodadditive, expressed on a body weight basis, that can be ingested daily over alifetime without appreciable health risk. For this type of comparison, accurateintake figures are necessary. Therefore, an assessment of absolute, rather thanrelative, intake is needed and most food frequency methods are not suitable,while dietary records, dietary history and 24-hour recalls provide relativelyaccurate intake figures. Overall, the dietary history method is probably thebest method for estimates of exposures to food additives, when constraints ofmoney, time, and staff are taken into account. Dietary records and 24-hourrecalls can be used, as long as their short-term nature is taken into account.

Effective use of the knowledge gained through a complexconsumption analysis is largely dependent on being aware of the targetpopulation, the methodology used to survey the population, correctidentification of the specific foods to which the ingredient will be added, andthe quantity of ingredient that will be added to these foods. The earlier in theprocess that a consumption analysis is conducted, the sooner any potentiallimitations may be detected, providing the manufacturer with the ability toalter the types of foods that the ingredient will be added to, or the amount ofingredient added to each specific food.

Ray A. Matulka, Ph.D., is staff toxicologist at Burdock Group,with responsibility for the development of the consumption analysis andreporting. Burdock Group (www.burdockgroup.com) has offices in Vero Beach, Fla.,and Washington, and offers guidance in strategic business planning and criticaldecision making to its domestic and international clients from the foodingredient, health, and nutrition industries.