Prevalence

The prevalence of eating disorders in this population has been difficult to establish with some studies suggesting that there is no difference while other studies suggesting otherwise. These inconsistencies are in part due to the standard eating disorder measurements used where questions assessing dietary concern, dietary restriction, and regulation of intake may overestimate the presence of an eating disorder in someone with diabetes. Conversely, these measurements do not take into account insulin omission or manipulation as a method of purging or weight control. Young et al. recently published a systematic review in 2012 with meta-analysis to address this issue and found that despite using generic or diabetes-adapted measures, the prevalence of disordered eating behavior and eating disorders was higher in adolescents with diabetes than their peers without diabetes.

Rates of insulin manipulation alone are currently stated at 14% for teenage girls and 34% in young adult women. In the preteen and early teen population, disturbed eating behaviors are also moderately common but relatively mild and with a lower prevalence rate in comparison to the older teenage population and compared to peers without diabetes. Behaviors such as binge eating, intensive excessive exercise, and dietary restriction were more common than insulin manipulation in this younger population.

Complications and Prognosis

Following the identification of the prevalence of eating disorders in this population, several longitudinal studies were performed looking at the implications of poor glycemic control associated with insulin manipulation. Rydall et al. followed 91 young women aged 12–18 years for 4–5 years and showed that intentional insulin omission increased in prevalence from baseline to follow-up and that eating disturbances persisted if present at baseline. More recently, Colton et al. (2015) completed a 14-year prospective study showing that point prevalence for insulin omission and eating disorders in this population increased with time. In particular, insulin omission was reported by 27% of participants at the 10–14-year mark. These findings were highly predictive of poor metabolic control as evidenced by a higher hemoglobin A1C levels and increased finding of retinopathy in those with highly disordered eating at baseline. The associations between eating disorders, insulin omission, and poorer glycemic control with its resulting higher incidence of microvascular complications were replicated in other longitudinal studies in the UK, Austria, and Japan. Along with these long-term complications, malnutrition in this age group results in a deterioration in bone health and impaired pubertal development and fertility – complications that occur in eating disorders when experienced during adolescence.

The crude mortality rate for anorexia nervosa is 5% at 4–10 years of follow-up and 9% after 10-year follow-up. There is evidence of a higher mortality rate and more episodes of diabetic ketoacidosis in those with previously identified eating disorders. Nielsen et al. in 2002 demonstrated that in Denmark, mortality rates after 10-year follow-up were 2.2 per 1000 person-years for individuals with insulin-dependent diabetes without anorexia nervosa, 7.3 per 1000 person-years for individuals with anorexia nervosa without diabetes, and 34.6 per 1000 person-years for individuals with both. One of the major causes of mortality in eating disorders is cardiac complications related to electrolyte disturbances. Insulin-dependent diabetics who manipulate insulin will usually present with either hyperglycemia, dehydration with acidosis, or hypoglycemia and seizures. Electrolyte derangements such as hypokalemia, hypernatremia, hypoglycemia, hypophosphatemia, and hypomagnesemia result in QTc disturbance and variability leading to arrhythmias and the possibility of sudden death. In particular, if laxative/emetic abuse is combined with hyperglycemia, the risk of hypokalemia is increased.

Pathways of Risk

The pathways of risk can be summarized as follows:

Diabetes management involves a high degree of attention to dietary intake and activity level in addition to multiple procedures such as blood sugar checks, injections, or pump management. The degree of attention to food and its effect on blood glucose levels resembles chronic dieting behavior. During puberty, individuals with diabetes are often at a higher weight than their peers due to the metabolic properties of insulin and therefore results in a higher degree of body dissatisfaction and a lower self-esteem. Without a supportive family with good communication skills and an understanding of adolescent development and the importance of autonomy, this reinforces the role of disordered eating in providing a sense of control and accomplishment.

The 2014 ISPAD (International Society for Pediatric and Adolescent Diabetes) Clinical Practice Guidelines recommend routine screening for disordered eating and misuse of insulin especially in the setting of weight loss, failure to gain weight, or a high HbA1c. Other warning signs may include an overall deterioration in psychosocial functioning (such as school attendance), erratic clinic attendance, and evidence of lowered mood, poor concentration, poor sleep, and fatigue.

Screening

Measurement of the prevalence of eating disorders and disordered eating in diabetes has been challenging as questionnaires used in the general population may include statements that may be considered appropriate to people with diabetes but indicative of an eating disorder in the general population. These questionnaires also do not include questions on insulin omission or misuse, and this in particular could lead to underdiagnosis in this population.

Treatment Options

During acute presentation in diabetic ketoacidosis, the recognition that insulin manipulation may be a cause of the presentation necessitates a different approach to acute management. Initially the aim should be a gradual reduction in blood glucose and gradual correction of electrolyte imbalances. Following medical stabilization, it is recommended that there is a thorough psychiatric assessment reviewing risk of self-harm and suicide as well as to confirm the diagnosis of an eating disorder. Rapid discharge without adequate establishment or recognition of the role of insulin manipulation in an episode of diabetic ketoacidosis can be harmful in the long run. Early identification of insulin manipulation incorporating intervention and nutritional recovery improves long-term outcomes with regard to microvascular diseases as well as pubertal development and fertility. Bone health recovery is less clear.

Most reviews of treatment recommend a multidisciplinary approach (endocrinologist, nurse educators, dietitians, and mental health providers trained in eating disorders management) either within an inpatient, day treatment, or outpatient setting. As with the standard approach to eating disorders treatment in this age group, starting off with an outpatient family-based therapy approach and then progressing to more intensive treatment settings is usually recommended, but there is little evidence to show which approach is most successful for this patient subgroup. With adolescents, the additional challenge within a family-based approach is to fine-tune the role of parental involvement to a level that is comfortable for all involved without risking deterioration in glycemic control from over or under involvement. Within a more intensive treatment setting, autonomy in insulin management is initially removed but should be reinstated gradually under supervision. The increased use of the insulin pump has also been an area of controversy. Several small studies have demonstrated that initiation of pump therapy has resulted in reduced disordered eating behaviors in adolescents with diabetes (Markowitz et al. 2010; Pinhas-Hamiel et al. 2010), but there is limited evidence as to whether continuing or initiating pump therapy is helpful or harmful to overall treatment of those with an established eating disorder and diabetes.

There are few published studies reviewing the eating disorders prevention or treatment specifically for adolescents with diabetes. Wade et al. (2010) recommended that prevention programs should occur within a framework of media literacy, with the focus on flexibility in diabetes care, improvement of self-esteem, and problem-solving capacity.

Males

Initial studies looking at the prevalence of eating disorders in diabetics pointed to the female sex as a risk factor with very low rates of disordered eating in males. Therefore most of the studies published have looked at pathways, screening, and treatment options for females with only very few studies including males. Insulin manipulation is less common in males compared to females (rates reported between 3–5% in adolescent males versus up to 15% in adolescent females). D’Emden et al. published a review of disordered eating behavior in adolescents with diabetes including males and concluded that 25% of their sample of males had one or more disturbed eating behaviors and that excessive exercise was used as a weight control method. More research is required to establish appropriate screening methods for this population as traditional questionnaires used for diagnosis of an eating disorder may not be as sensitive in males.

Treatment

More research into treatment options for this difficult population is required. Young et al.’s systematic review proposed shifting the focus of research to more clinically relevant issues such as identification and treatment of eating disorders and insulin manipulation. In particular, a closer examination of the impact and role of insulin pumps in this population is warranted. Apart from early recognition and intervention, traditional eating disorder treatment options at present are recommended as early intervention, and treatment may prevent the development of microvascular disease in particular.