Nutrition Care for Clients with Diabetes and An Eating Disorder

By Nancy L. King, MS, RD, CDE, and Linda M. Watts, MA, RD
Reprinted from Eating Disorders Review
January/February 2003 Volume 14, Number 1
©2003 Gürze Books

Adolescents with Type I diabetes have a higher-than-normal prevalence of eating disorders. In fact, although roughly 1% and 4%, respectively, of patients in the general adolescent and college-age populations have Type I diabetes, nearly 7% of patients with anorexia nervosa (AN) and bulimia nervosa (BN) have it. In one study, 16.9% of 89 adolescent females with Type I diabetes had “disturbed eating behavior.”1 There is also greater body dissatisfaction and an increased drive for thinness in among females with Type I diabetes. Interestingly, an increased drive for thinness has been observed in males one year after the diagnosis of Type I diabetes.2

Early Warning Signs of an Eating Disorder

How can clinicians detect early warning signs of the development of an eating disorder among clients with Type I diabetes? By maintaining an open dialogue with patients about body image, self-esteem, and a sense of self-efficacy as these relate to diabetes. For example, you might ask the client, “How do you feel about your body’s demand for food when your blood sugar drops below 60 mg/dl?” Or, “Do you feel differently about your body when you’re wearing your insulin pump?” This may give her a chance to disclose that she is struggling with eating disorder thoughts or behaviors.

Blood glucose levels that are difficult to control or erratic, or don’t seem to match food intake, stress levels, or physical activity, could be the result of disordered eating. Skipping insulin doses or scaling back on the prescribed regimen to make weight loss easier or to prevent weight gain, can also be associated with an eating disorder. This particular behavior is extremely important to monitor because insulin omission is the primary cause of recurrent diabetic ketoacidosis (a potentially life-threatening biochemical event triggered by unusually elevated blood glucose levels) in adolescents. Some other warning signs are trying to eat “perfectly,” expressing disgust when glucose levels are elevated, self-berating one’s body, excessively exercising to bring glucose levels back to normal, and binge eating.

Over time, disordered eating will cause hemoglobin A1c (glycosated hemoglobin; see also table below) levels to rise. In one study, adolescents with an eating disorder and Type I diabetes had a mean A1c of 9.4%; non-eating-disordered teens had a mean A1c of 8.6%; and subclinical eating-disordered adolescents hadmean A1c of 9.1% (the ideal is approximately 4%-6%).3 Even though these values indicate elevated blood glucose levels over time, the client’s erratic eating, and exercise, and insulin patterns can create a much higher risk for severe and unanticipated hypoglycemic events. Poor diabetic control can lead to long-term consequences among persons with diabetes and an eating disorder.

One overall approach to management of this population is to anticipate that the relationship with a client will be long term, open, process-oriented, and supportive. See yourself as a guide and advisor who “comes alongside” the client in a supportive way rather than as the authority on diabetes and eating disorders. This approach may be crucial for individuals, especially adolescents, who are struggling with an eating disorder because of underlying control issues and the daily demands and decision-making associated with diabetes management. Empowering the client to control what is in her primary “circle of influence” is a primary focus. But, it is equally important that she identifies and accepts the facets she cannot control and learns how to live with these limitations.

Important Nutrition Goals

Clients should be able to: (1) eat a wide variety of foods and manage the impact on their glucose levels; (2) eat in a way that adequately nourishes their body daily and preventively for future needs and challenges (i.e. growth, pregnancy, illness, aging); (3) identify and respond to their body’s cues of hypo- and hyperglycemia; and (4) take more responsibility for their dietary choices, physical activity, and insulin/medication.

A good beginning in nutrition education is to ask the client what she already knows about diabetes. For example, does she know someone who currently has or had diabetes? What did they observe? What concerns do they have today? Who is in her life to support her? It’s also helpful to cover the basic physiology of digestion (with diagrams and drawings), including the biochemical and mechanical aspects. This provides a natural lead-in to the glycemic impact of food and physical activity and the relevance of the body’s cues (i.e., hypo/hyperglycemia, hunger, fullness, appetite, satiety, satisfaction). Later, you can weave in the micronutrient value of foods such as calcium, iron, and folic acid, for example. The initial goals are to inform and equip the client so she can normalize glucose levels as soon as possible, or at least prevent severe hypo- and hyperglycemic episodes.

Promoting Healthy Eating

Most of the time, clients can expect to finish a meal or snack and be sustained by protein, energized by carbohydrates, and satisfied by dietary fats and sweets. The degree to which clients are ready to experience this depends on the reasons for their eating disorder. For example, if denial of pleasure or tolerance of inadequacy are the primary motivations for engaging in eating disordered behaviors, then don’t expect clients to be ready to eat in a way that brings satisfaction or satiety. But, if they are ‘tired of feeling tired’ or depressed, explore the role and the amount of carbohydrates that can make them feel energized and promote a healthier brain “marinade,” a term first used by Dr. Joel Yager.

As clients become more open to feeling sustained, energized, and satisfied, they begin to experiment and discover foods from the respective groups that produce these outcomes, and are able to track their glycemic responses. This framework can be effective because it puts clients in the driver’s seat. It’s not focused on what they should do (a moralizing term), but rather on what outcomes they can and want to produce. At some point, clients discover this is the same dynamic that guided their eating disorder behaviors, but now the outcomes are supporting health. This shift depends upon each client’s readiness to transition from expressing life issues through eating disordered behaviors to acknowledging, accepting, and meeting the body’s needs through a nourishing eating style.

Meal Planning Tools

Carbohydrate counting is useful for many clients with diabetes, whether they have an eating disorder or not. Carbohydrate counting is a glucose management method that was used in the Diabetes Control and Complications Trial.4 With this method, clients are taught the carbohydrate content of each food group and given carbohydrate goals for each meal and snack to match their insulin regime (e.g., 62 g of carbohydrate at breakfast). Clients are encouraged to eat a variety of foods as long as the carbohydrate total for the meal/snack is within 3 g of their goal.

However, emphasize that they do not have to aim for perfection. In the same way most of us wouldn’t stop to pick up a penny on the sidewalk, our bodies do not count pennies (“pennies” being 1 g of carbohydrate, or 5-15 kcal over or under what is anticipated). But, some individuals might need to care about the nutrition nickel, dime, or quarter. Explain to the client that one of our jobs as partners in the nutrition work is to discover what her body cares about using her self-monitoring blood glucose (SMBG) values. Of note, equating glucose monitoring with taking one’s temperature is helpful to portray a nonjudgmental, but concerned and caring response to glucose values.

Carbohydrate counting can help clients with eating disorders see foods in a more neutral way, rather than as “good or bad.” This encourages them to expand their “safe foods” list and to rediscover the pleasure in eating. The meal planning technique can empower them to make their own food choices, and by using the nutrition information on food labels, predict their blood glucose results. This tends to motivate clients to shift the responsibility of glucose management from their parents, or treatment team, to themselves.

Interestingly, carbohydrate counting can also help clients access the anger, injustice, frustration, and disappointment they feel, by making the limitations and the exceptional daily demands of their bodies more concrete. Ironically, although many clients with an eating disorder truly believe their bodies process food differently, or have different requirements than others, in the case of diabetes and eating disorders, facets of these beliefs are quite true. As clinicians, it is important to validate any accurate beliefs and the accompanying thoughts and feelings.

Then there are some clients for whom more “counting” approaches are not helpful. In this case, one can generalize portions by handful, fist-size, or hand-size as much as possible. If glucose values are predictable using this manner of meal planning, it is adequate for glucose management. The Willett pyramid may be helpful as a general nutrition guide. (It can be found at

Patients Who Purge

It is particularly challenging to work with clients who struggle with purging. Without SMBG, it may be nearly impossible to work safely and effectively with these clients. Complications such as delayed gastric emptying, coupled with purging through laxatives, vomiting or excessive exercising, can make medication adjustments difficult and can result in unstable and/or unsafe glucose levels. So, in the interest of clients’ safety, one may see them three times a week and have phone contact with them between appointments. This dual diagnosis may require making a recommendation for inpatient treatment sooner for these clients than for clients without diabetes.

Insulin Regimens

Insulin regimens are determined by the client’s ability and readiness to eat and engage in physical activity in a consistent and adequate way. With consistency and adequacy, blood glucose fluctuations occur within a narrower range. When the client experiences few serum glucose fluctuations, the risk of hypoglycemic episodes is lowered and her goal range can be moved closer to what is considered normal. But the more a client has an erratic and an inadequate eating style, the higher her serum glucose goal range will need to be above normal to prevent severe hypoglycemia. One approach is a combination of medium-/long-acting insulin, with short-acting insulin to accommodate eating according to hunger and fullness cues. Some clients who have a more consistent eating style find they prefer an insulin pump. Just like the other facets of an individual’s treatment plan, insulin is tailored and adjusted in a way that meets the client’s needs at any given point during recovery from the eating disorder.

About the Authors

Nancy L. King is a registered dietitian, certified diabetes instructor, and exercise physiologist who has been in private practice since 1987. She is a nationally known speaker and workshop and retreat leader, and author of the audiotape series Counseling Skills for Personal Wellness and co-author of Moving Away from Diets: New Ways to Heal Eating Problems and Exercise Resistance.

Linda M. Watts, MA, RD, is a member of the EDR Editorial Board and author of our Nutrition Notes column. She specializes in eating disorders and nutrition at St. Paul’s Hospital, Vancouver, Canada.


  1. Engström I, Kroon M, Arvidsson CG, et al. Eating disorders in adolescent girls with insulin-dependent diabetes mellitus: a population-based case-control study. Acta Paediatr 1999;88(2):175.
  2. Steel JM, Lloyd GG, Young RJ, et al. Changes in eating attitudes during the first year of treatment for diabetes. J Psychosom Res 1990; 34(3):313-8).
  3. Jones JM, Lawson ML, Daneman D, et al. Eating disorders in adolescent females with and without type 1 diabetes: cross sectional study. Br Med J 2000;320:1563.
  4. Diabetes Control and Complications Trial. NIH Publication No. 02-3874. October 2001.
No Comments Yet

Comments are closed