• Nutrition Physiology
• Hyperalimentation

• Nutrition Physiology
• Hyperalimentation
• Nutrition

Measuring food intake is a crucial part of the study of nutrition. Dale A. Schoeller, PhD, University of Wisconsin, Madison, Wisconsin, USA, discussed the pros and cons of the various food monitoring technologies currently being used in nutritional studies.

For outpatients, the most popular food tracking methods include weighed diet records and surveys based on recall (either based on food frequency or meal-by-meal). The problem with both of these is the accuracy and precision of the reporting. Underreporting is particularly problematic in those with higher body mass index (BMI; Figure 1) [Nielsen BM et al. J Nutr 2009] and among adolescents. Specific disadvantages include inaccuracy due to serving size error and/or missing foods, etc, that are influenced by social desirability and memory. The advantage of these record and recall techniques is that they are capable of identifying the location, time, and patterns of food intake.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Underreporting Increases With BMI

BMI=body mass index.Reproduced from Nielson BM et al. Past and current body size affect validity of reported energy intake among middle-aged Danish men. J Nutr 2009;139(12):2337–2343. With permission from the American Society for Nutrition.

Current inpatient clinical methods that avoid serving size and memory errors include using weighed trays or a form of inventory control such as a vending machine or controlled pantry. These inpatient methods are accurate, offer good dietary control, and allow for analysis of data; however, they can also be influenced by social desirability and are conducted in a nonrepresentative, artificial environment.

Better interview methods and the use of computers have improved survey approaches to dietary research. Web-based dietary recalls such as DietDay are inexpensive, offer wide accessibility, and are superior to paper food-frequency questionnaires [Arab L et al. Am J Epidemiol 2011]. The Digital Photography of Foods Method, which uses a combination of human and computer software to identify foods consumed accurately, calculating energy and nutrient intake of adults and children in cafeterias [Martin CK et al. J Hum Nutr Dietetics 2013]. A similar approach (Remote Food Photography Method [RFPM]) uses smart phones to capture images of the food selected and leftovers. These images are then transmitted in near real-time to a server for diet analysis. RFPM can be very accurate, and it offers the advantage of reduced participant burden, eliminates of the need for participants to estimate portion size, and uses automation to improve the accuracy, efficiency and cost-effectiveness of measuring food intake (Figure 2). These newer tools also reduce analysis time, improve accuracy, can identify food patterns, and allow for nutrient analysis. Among the disadvantages are the lack of precision (a 6-day average of only 30%) and the requirement for participant cooperation in and awareness of the process that may alter eating behavior.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Digital Photo Analysis of Food Intake

Reproduced from Martin CK et al. Measuring food intake with digital photography. J Hum Nutr Diet 2013; Suppl 1:72–81. With permission from John Wiley.

Less traditional methods of food monitoring include detecting swallowing by using a sound sensor located over laryngopharynx or by a bone-conduction microphone and detecting chewing through a below-the-ear strain sensor [Sazonov E et al. Physiol Meas 2008]. Methods relying on chewing sounds can successfully distinguish dry, wet, and soft foods [Amft O, Tröster G. Artif Intell Med 2008]. Wrist actigraphy can detect meal events, snacking, and within-meal patterns [Dong Y et al. Appl Psychophysiol Biofeedback 2012]. These less traditional devices are useful for identifying eating events, mealtime length, and food texture, but they cannot identify the actual food eaten or measure total energy.

Other experimental approaches include glucose monitoring to detect time of eating events, monitors for heart rate, respiration rate, skin temperature, oxygen saturation, and blood pressure are also available and may prove useful in the study of ambient and physiologic factors associated with eating behavior. Stable isotopes can be used to identify individuals having omnivore, lacto-ova and vegan diets, as well as sugar intake [Choy K et al. J Nutr 2013]. Dietary biomarkers for protein, fish oil, citrus fruit, and garlic can objectively assess dietary consumption without the bias of self-reported dietary intake errors, but assessments of their precision for most biomarkers are still in their infancy [Hedrick VE et al. Nutr J 2012].

Dr. Schoeller concluded that traditional dietary methods are inaccurate, imprecise, and differentially biased, while the technology enhanced traditional methods are more accurate and precise, but probably also differentially biased. Biosensors are less differentially biased but very imprecise. The rapidly expanding field of less traditional methods for monitoring food intake is more objective, but their accuracy, precision, and utility requires further research and development.

• © 2014 MD Conference Express®