Disordered physiology in Binge Eating and other eating disorders
Date: March 9th, 2006
Title: “The meal as a clinical assessment tool for eating disturbances in humans”
Speaker: Jennifer Nasser, Ph.D., RD
Affiliation: New York Obesity Research Center, St. Luke’s Roosevelt Hospital
Dr. Jennifer Nasser provoked an interesting discussion about the potential to use parameters of a test-meal intake to clinically evaluate patients for binge eating disorder (BED).
While still in the early stages of her work, she is planning to collaborate with several other registered dietitians and investigators to further the intriguing preliminary results. In this presentation, Dr. Nasser’s research questions were the following:
- What is the contribution of disordered physiology to binge eating behavior in obese humans?,
- How do obese humans with BED differ physiologically from obese humans without BED?, and
- How can Nutritional Assessment aid in answering these questions and contribute to better diagnosis and treatment?.
The work at hand was motivated by two key events. First, patients who have been diagnosed with BED, according to DSM-IV criteria, present with a variety of meal patterns, composition, and caloric intake. Second, the ADA recently proposed to expand the role of the RD in treatment of eating disorders, an area which has primarily been handled by the mental health field.
Currently, eating disorders such as anorexia nervosa, bulimia nervosa, binge eating disorder (and others), are diagnosed by using patient self-report and BMI, and classifying patients according to criteria established in DSM-IV.
The “self-report” nature of this classification, however, means that much of this diagnosis is subjective, and dependent heavily on patient interviewer. If additional assessment were incorporated from the American Dietetics Association, this should accord with their pre-established assessment standards that include anthropometics (A), biochemical or lab values (B), clinical signs (C), dietary intake (D), and environmental food availability interactions (E). This nutritional assessment is providing the framework for Dr. Nasser’s experimentation in using a test meal to clinically evaluate patients with BED.
In contrast to the standard use of a patient interview, test-meal intake is an objective measure, and parameters of this intake (eg., duration, energy intake) might provide useful information for more accurately diagnosing and treating BED patients. In addition, the clinical use of a test meal might be able to distinguish the physiological causes of eating disorders from the psychological causes.
The initial experimental evidence for this hypothesis came from an experiment that tested the difference in ad libitum intake (after a 12 hour fast) of a dilute (0.5 kcal/ml) liquid meal product between individuals diagnosed as meeting full DSM-IV criteria for BED, meeting some but not all criteria for BED, or meeting no BED criteria.
The data from this experiment suggest that control subjects, without BED, appeared to consume a lower amount of a test-meal, and eat for a shorter amount of time than did patients who had been diagnosed with BED. Even within BED patients, however, there was substantial variability in eating rate, total time for the meal, and amount eaten. Thus, determining additional diagnostic criteria to further classify these patients seemed like a logical extension.
Dr. Nasser discussed the characteristics of a useful clinical assessment tool for evaluating eating disturbances in the next portion of her talk.
First, the tool should be integrated into current assessment paradigms (eg. DSM-IV). There should be a low demand on both the client and the clinician, and the assessment tool should be accurate and highly correlated with clinical symptoms. The test should be easy to perform, inexpensive and cost-effective. Finally, the inter-clinician reliability should be good. Taking that into consideration, if test meal intakes are included as part of the eating disorder assessment, they would need to fit most or all of the above specifications in order to be effective.
The pilot data leading up to these hypotheses provide support that several facets of the test meal might be associated with eating disturbances. For example, BED patients might consume more food in a longer duration than control subjects, although at this point, results are not statistically significant. Further, it is possible that amount of the test meal consumed might be useful in defining what a “larger than normal amount” is.
This subjective classification is currently part of the DSM-IV criteria for establishing a binge episode, but no specific definition is offered for determining this amount. Dr. Nasser found that her data could be divided into two separate categories, patients that ate for a short time (~5.6 min) and consumed around 432 kcals, and patients that ate for a longer time (20.4 min) and consumed around 703 kcals.
Further experimentation should be done to determine the usefulness of using meal duration and intake to assess patients with BED, in particular the use of a test-meal to determine the presence of actual binge episodes. Interestingly, the patients that Dr. Nasser identified as consuming the most test meal rated the flavor of the test meal higher than those who consumed less of the test meal.
The final portion of Dr. Nasser’s talk addressed the relationships between physiological measures and test meal intake and duration. While results are preliminary, data suggest that ghrelin levels might be positively related to duration of test meal intake, but further experimentation is needed to fully test this hypothesis.
In summary, Dr. Nasser hypothesized that the test meal duration might be useful to distinguish two sub-groups of patients with BED. Her future work will further refine the notion that test meal duration, intake, physiological measures, and sensory factors might allow us to improve classification, diagnosis, and treatment of patients with eating disturbances.
Q. What is the definition of a “normal amount of food?”
A. That is going to be one of the focuses of my talk. This definition is subjective, and that is why I believe certain meal-related parameters might provide useful information about binge-eating disorder (BED) patients.
Q. What is the composition of the Boost test drink, and did you (subjects) taste test the food before the experiment?
A. Boost is the same as Ensure, and it is 50% carbohydrates. Subjects did taste test the Boost, and they had similar acceptance ratings for it.
Q. How much of the liquid diet did subjects consume?
A. Subjects consumed until they felt extremely full.
Q. In your figure entitled, “Relationship of TM Duration and Intake,” what are the controls?
A. The controls are the patients without BED.
Q. Do you believe that there is a difference in sensory specific satiety between your subjects based on how much of the test meal that they ate?
A. It could very well mean that.
Q. How long were subjects fasting before ghrelin was measured?
A. 12 hours
Q. Do you ask your subjects whether they routinely consume liquid meals as part of their diets?
A. Yes, but I don’t have that data with me.
Q. What are the anchors of the flavor rating scale?
A. 0-100 mm, but I don’t remember the exact wording of the anchors
Q. What is the test meal that you plan to use in your follow-up experiments?
A. I haven’t agreed upon that yet, but perhaps Boost.
Q. Is there any fat in Boost?
A. Boost is 68% CHO, 15% Fat and 17% Protein.
Q. Don’t you have to choose some sort of food that will actually induce a binge? Do subjects binge on liquid diets?
A. There is a high correlation between eating rates of liquids and solids, and we also have a food recall of subjects, so we can estimate how the test meal intake correlates with their usual intakes.
Q. What do you mean by saying “include a small amount of fat and protein with each meal?” What is a small amount, and what is the rationale for doing this?
A. The rationale is that protein is more satiating, so including a small amount will increase the satiating potential of the test meal. The amounts I would recommend would be based on standardized amounts for a meal.
Q. What is the evidence (in humans) that increasing the amount of fat in a meal increases the satiety? The animal data isn’t that good. Is there human data?
A. Yes, I think so. But, I am unsure how good these data are.
Q. I think the first thing that you should do is correlate intake of the liquid meal to real world intake, to find out if your test meal has any relationship to how subjects perform outside the lab?
A. That’s a good idea.