Using the Remote Food Photography Method to Assess Diets

Linda F. Benedict, Tuuri, Georgianna

Taking photos of food on an iPhone. (Photo by Tobie Blanchard)

Georgianna Tuuri

It is difficult for nutrition researchers to accurately determine what people eat and how often they eat. The most common method to find out what someone consumes is to ask them what they ate the previous day. This method, called the 24-hour recall, is relatively easy to administer and not too time-consuming, but people may forget what they have eaten the day before, and it may not represent their typical diets. Another way to estimate dietary intake is to ask people to keep written records of everything they eat or drink for several days. The length of time these records are kept can vary from a few days to several weeks. While this method is more representative of what a person normally eats, remembering and writing down all the food and beverages consumed creates a burden for a participant. Food items may be forgotten, and participants may tire of recording the information.

A new dietary assessment method, called the Remote Food Photography Method (RFPM), offers an alternative to traditional approaches. Instead of interviewing people or asking them to write down what they eat or drink, the RFPM relies on pictures taken from cameraenabled smartphones. Participants don’t have to come to a laboratory or spend time writing down the dietary information. They simply take a picture before and after eating or drinking using a smartphone. The images are then sent via a wireless network to a custom-built server at Pennington Biomedical Research Center. Food images are compared to food pictures of certain portion sizes, and the calories and important food components are calculated by registered dietitians. This semi-automatic process uses both computer automation and human input. In addition, researchers send email or text messages to help participants remember to collect food images and communicate with subjects if they have questions about particular foods eaten.

Recently, this novel dietary assessment method was used to examine if a person’s age was associated with the number of times each day that food was consumed and if age was related to the energy density, or kilocalories per gram, of a person’s diet. Fifty-four Caucasian and African-American men and women from south Louisiana used the RFPM to record their food and beverage intake. They ranged in age from 16 to 65 years of age. Before using the RFPM the participants learned about the study and practiced taking pictures and sending them to the researchers. They were asked to continue practicing their usual dietary habits and to send pictures of everything they ate or drank each time they consumed food. The participants’ daily physical activities were estimated, and they were measured for height and weight. Body mass index (BMI) values, which describe people’s weight in relation to their height, were calculated, and the daily calorie needs estimated. Participants began sending pictures using the RFPM on a Wednesday, and this day was considered a practice day. Over the next three days (Thursday through Saturday) participants continued recording and sending information. Researchers sent text messages to remind participants to take and send pictures and communicated with them about their dietary intake. The number of times each day that participants sent pictures was recorded, and the average energy density of their diets was calculated.

The three-day diet records collected using the RFPM indicated that most people ate about four times per day and the average energy density was about 1.7 kilocalories per gram of food eaten. To see if age predicted the number of eating occasions and the energy density of the diet, it was important to control for the influence of other variables such as gender, race and weight status. Using backward stepwise regression while controlling for these other influences, it was found that age did predict the number of times each day that people ate, but their age was not related to the energy density of their diets. Older participants tended to eat more frequently than younger ones. It is often assumed that adolescents snack more than adults, but the findings from this study suggest that this may not be true. Further research, with larger groups of individuals, is needed, but the RFPM offers an effective way to estimate dietary intake. The results from this study were presented at the 2013 Experimental Biology Meeting in Boston, Mass., and are published in the April 2013 edition of the Federation of American Societies for Experimental Biology Journal.

Georgianna Tuuri is an associate professor in the School of Nutrition and Food Sciences.

(This article was published in the fall 2013 issue of Louisiana Agriculture magazine.)

12/4/2013 8:25:04 PM
Rate This Article:

Have a question or comment about the information on this page?

Innovate . Educate . Improve Lives

The LSU AgCenter and the LSU College of Agriculture