Desmond Ballance, Jodie Jenkinson
Chelsea is a 13-year-old adolescent who was diagnosed one year ago with anorexia nervosa, an eating disorder characterized by self-starvation and excessive weight loss. To meet her under any circumstance that did not involve food, you probably wouldn't guess that she had an eating disorder (ED). But when you ask her about her feelings toward eating, food, and meal planning, her anxieties are evident. She finds planning a meal extremely difficult, particularly when incorporating healthy quantities of "stressful" foods like carbohydrates and fats into a meal. She relies on her parents to plan her meals and uses measuring cups to determine true portion sizes, because through her eyes, everything looks like too much food.
Nutrition counseling can help adolescents like Chelsea, who are recovering from EDs, to improve their understanding of how to plan balanced meals. Food guides and sample menus are available for teaching teenagers about healthy portion sizes, but none provides a visual reference about what a healthy and balanced meal, as well as a proper portion, should look like. Since patients with EDs have demonstrably abnormal perceptions of the size of food, a meal-visualization tool could help patients with EDs feel more comfortable about portions by helping them understand what appropriate food portions look like in the context of a balanced meal.
Nutrition-focused interactive multimedia tools are valuable for improving comprehension, but to date, no such program has been developed to help patients practice their meal-planning skills and reduce their anxieties about food portion sizes. The number of professionally designed visuals available to most dietitians for educating patients with EDs is limited. Although current visual aids do provide some reference points for patients, they are limited in their ability to help patients picture complete meals and feel comfortable with what constitutes a so-called normal-size meal.
One example of a visual aid available to registered dietitians and familiar to many Canadians is Canada's Food Guide for Healthy Living (see Figure 1) , similar to the well-recognized American Food Pyramid (see Figure 2) . This color-coded handout displays the major food groups, examples of foods found in each group, recommended daily servings, and examples of what constitutes a serving: For example, one slice of bread is equivalent to one serving from the grain-products food group. For foods with undefined shapes, imperial and metric measurements are provided: For example, one cup or 250 mL of milk is equivalent to one serving from the milk-products food group.
Although both the Canadian and American food guides provide an overview of which foods should be included in a healthy diet, the relative sizes of the food servings may be difficult for some people to visualize since they are presented in isolation, not as part of a meal. For example, two tablespoons of peanut butter constitute one serving. In the current American food guide, no visual representation is shown for peanut butter . In the Canadian food guide, two tablespoons are represented alongside a jar of peanut butter . However, without a context-appropriate size reference available for the tablespoons or the jar, the recommended single serving of peanut butter may be perceived as significantly more than it would be had it been spread on a piece of bread. At particular risk for misinterpreting such quantities are patients with EDs, whose perceptions of food sizes are abnormal. In fact, one study demonstrated that patients with EDs perceived food as 12 percent larger than the control subjects dida percentage that may be larger since the researchers did not reliably ensure that control subjects were free of EDs .
Plastic food models are another visual tool available to registered dietitians to help patients with EDs learn about portion size. These models can give the patients a sense of the amount or size of a food item that constitutes a single serving. Patients can also compare the models with other items such as a fist, the palm of a hand, or a deck of cards to ascertain their relative size for easy measurement during meals. However, extensive use of the models by patients with EDs is not recommended because of the potential for these patients to fixate on the model as the "gold standard," when in reality, servings may be slightly larger or smaller than that of the model. Furthermore, since food models are often used in moderation during patient counseling sessions and, for cost-related reasons, are not readily available for patients outside the office, the potential of food models to acclimatize patients with EDs to phobic food stimuli is limited.
Alternatives to food models are often used. Dietitians may sketch a dinner plate for the patient, dividing the plate into sections and explaining that one-half of the plate should be designated for vegetables, one-quarter for proteins, and one-quarter for carbohydrates. This type of visual imagery enables patients to imagine their own plate and, with the food sketched in, better understand how much of the plate should be covered to constitute an appropriate meal. However, these visuals lack realism and dimensionality, offering only a map of where the food will lie without any sense of food quantity. Patients with EDs need a sense of volume to relearn proper eating habits and become comfortable with the serving sizes that constitute appropriate meals.
As another visual alternative, dietitians may ask patients to compare the amount of food they eat at meals with what their family and friends eat. If motivated, the patients have the opportunity to view numerous examples of healthy meals in their own environment. However, this technique is something that can be used only outside the counseling session; patients not fully committed to their recovery could easily neglect it. Furthermore, genetic inheritance can play a moderate to substantial role in the development of an ED . According to Strober et al, immediate family members of people with anorexia nervosa were 12 times more likely to develop anorexia than people without a family history of EDs . If family members have destructive or abnormal eating behaviors, patients may find it more difficult to establish what a normal meal actually looks like and to accept what a balanced meal should be, according to their personal nutritional requirements.
While food guides and sample meal plans are among the more common types of visual media used by registered dietitians for educating patients, other types of visual tools have been explored.
Lozano et al. reported the development of a virtual environment (VE) in which patients can interact with food and practice eating . Food types in this VE included an apple, pizza, salad, water, and soda that a patient can pick up and direct toward his or her mouth, causing the geometry of the food to change, simulating a bite taken and representing a decrease in the food available as the patient "eats" the food. Sounds associated with the eating process, such as biting, chewing, and swallowing, were also incorporated into the program in an attempt to fully immerse the patient in the eating experience. In a controlled study of this VE, the authors reported greater effectiveness of the virtual eating process among patients with EDs than with other simpler interactive processes; the more realistic the environment, the stronger the feelings of anxiety among patients exposed to the food in the VE . Patients were also more motivated to practice interacting with these foods for potential integration into their real diet. Although this VE could help patients with EDs reduce their anxieties about the eating process, it did not help the patient to become comfortable putting a meal together, since only a few virtual food items were available to the user. Furthermore, since virtual reality programs are in the early stages of development and require expensive equipment and software, this technology, even with a wider range of virtual food items, is cost-prohibitive to patients with EDs and not readily available for regular use. Demonstrated improvement in patient motivation to interact with real food after immersion in the VE supports further exploration of other programs that might enable patients to interact with different food portion sizes and practice planning meals in a similarly safe setting that is easily accessible from various locations.
Another initiative focused on the use of visual tools to help patients learn how to eat is the Mandometer an ED program involving a computer-based device that measures food intake and the perception of satiety and focuses on changing a patient's eating habits and behaviors. The apparatus is composed of a scale connected to a computer that continuously weighs the food throughout the meal and displays two curves on the monitorone, the recommended eating rate and the other, the patients' satiety. Patients follow the curves during their meal, periodically inputting their perceived level of fullness so that the computer can adjust the eating rate. Overtime patients learn how to eat and perceive satiety. Studies have reported successful treatment of patients with anorexia nervosa and up to two years of remission after treatment with the Mandometer . While this tool helps patients become accustomed to sitting down to a complete meal, it offers no information on what constitutes a healthy meal. It also does not help patients become comfortable at making decisions about appropriate food portions, a skill that is essential to everyday living. Similar to VEs, the technology for the Mandometer is extremely new and relies on sophisticated hardware and software that is currently unavailable to the majority of patients with EDs. The positive effects on the treatment of EDs with the Mandometer, however, must be evaluated cautiously since these authors are the original developers of the program. To date, no independent studies of the effectiveness of the Mandometer for treating ED patients have been completed.
In response to the absence of an effective meal-visualization tool, MyMeal was designed as a Web-based, user-tailored, interactive meal-visualization tool, intended to offer teens the opportunity to become comfortable with portion size, practice creating balanced meals, and better understand what a normal meal should look like on a plate. The prototype allows users to create personalized daily meal plans by selecting food images and dragging them onto the appropriate tableware (see Figure 3). The user can also access feedback about the number of servings from each food group he or she has included in his or her daily meal plan, compare it with the recommended range of servings for a single day, and make adjustments to personalized menus as necessary. When the user's daily meal plan is complete, he or she can save it to a specific calendar date or dates, and revisit these saved meals for future referencing or editing as desired. Qualitative and quantitative evaluations of the MyMeal program still need to be completed, but initial informal assessments of the program by health care professionals caring for ED patients felt the concept had strong potential and supported continued investigation.
In summary, the major problems with the existing visuals for educating ED patients about meal planning are limitations in providing an adequate representation of food portion size within the context of a complete meal, in an accessible and cost-effective manner that enables ED patients to practice meal planning. Our preliminary investigation into the application of an interactive meal visualization tool for assisting adolescents with EDs to improve their understanding of food portion sizes and balanced meals has been encouraging. Ultimately, interactive tools such as MyMeal may offer adolescents like Chelsea a highly interactive, non-threatening way of reducing anxiety associated with eating and meal planning.
1. Health Canada. "Eating Well Canada's Food Guide"; http://www.hc-sc.gc.ca/fn-an/alt_formats/hpfb-dgpsa/pdf/food-guide-aliment/fg_rainbow-arc_en_ciel_ga_e.pdf/
2. USDA (United States Department of Agriculture). 2009. "Inside the Pyramid"; http://www.mypyramid.gov/pyramid/index.html/
4. Bulik, C.M. "Role of Genetics in Anorexia Nervosa, Bulimia Nervosa and Binge Eating Disorder." In Clinical Handbook of Eating Disorders: An Integrated Approach, ed. T.D. Brewerton. New York, NY: Marcel Dekker, 2004.
5. Strober, M., Freeman, R., Lampert, C., Diamond, J., and Kaye, W. "Controlled Family Study of Anorexia Nervosa and Bulimia Nervosa: Evidence of Shared Liability and Transmission of Partial Syndromes. American Journal of Psychiatry 157, 3 (2000): 393401.
6. Lozano, J.A., Alcaniz, M., Gil, J.A., Moserrat, C., Juan, M.C., Grau, V., and Varvaro, H. "Virtual Food in Virtual Environments for the Treatment of Eating Disorders." Studies in Health Technology and Informatics 85 (2002): 268273.
7. Bergh, C., Brodin, U., Lindberg, G., and Södersten, P. "Randomized Controlled Trial of a Treatment for Anorexia and Bulimia Nervosa." Proceedings of the National Academy of Sciences of the United States of America 99,14 (2002): 94869491.
Desmond Ballance is a project manager with Lifelearn Inc. in Guelph, Ontario, Canada, where she designs and develops interactive online learning initiatives for the veterinary and human medicine communities. With an M.Sc. in biomedical communications from the University of Toronto and an M.Sc. from the Department of Animal Science at the University of Manitoba, she looks forward to blending her interests in animals, science, education and art into a fulfilling, if not eclectic, career. Ballance is currently working with a Toronto-based research team on MyMeal, to advance its development and potential as a clinical tool for patients with eating disorders.
Jodie Jenkinson is a full-time faculty member in biomedical communications at the University of Toronto, where she teaches community-centered design research, information visualization, and Web-based health and science design. Her research interests include information visualization and perceptual design theory, and the evaluation of interactive health and science education tools. Jenkinson has extensive experience in the development and evaluation of educational tools for both the professional and lay audience. She is the author and instructional designer of The Prostate Centre, a Web-based guide for newly-diagnosed oncology patients and co-author of BreastMatters, a breast cancer awareness program for breast screening patients who are at high risk of developing breast cancer.
©2010 ACM 1072-5220/10/0300 $10.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc.