The Dangers of Undiagnosed Eating Disorders in Patients with Type 1 Diabetes

Patients with type 1 diabetes and eating disorders have a greater risk of diabetic ketoacidosis and all-cause mortality. 

Eating disorders are severe illnesses associated with abnormal eating patterns secondary to pre-occupations with body weight and body image. Some examples of eating disorders include anorexia nervosa, bulimia nervosa, and binge eating disorder. Various studies noted a higher prevalence of an eating disorder in younger patients with type 1 diabetes mellitus (T1DM), with one meta-analysis estimating about a 2.5-fold higher prevalence than those without T1DM. One potential clinical pearl for patients with both conditions is the underutilization of insulin due to fear of weight gain. As a result, this can lead to poor glycemic control and a higher risk of comorbidities and mortality.

A recent cohort study looked at the risk of diabetic ketoacidosis (DKA) and all-cause mortality in patients with T1DM and an eating disorder. A study sample consisting of adolescents and young adults from the ages of 10-39 was collected from the Ontario Diabetes Database in January 2014 and followed for six years. Investigators screened the database for people with T1DM using an algorithm with a specificity of 99.5%. They also screened people for an eating disorder, defined as an acute care visit, psychiatric hospitalization, or emergency department visit where the primary or secondary diagnosis was recorded as an eating disorder using ICD-9 and ICD-10 codes. The primary outcomes were all-cause mortality and DKA. For statistical analysis, the risk of all-cause mortality was measured using a Cox proportional hazards regression while a Fine-Gray model calculated DKA incidence. In addition, the authors compared one person with a history of an eating disorder to ten people of similar age and gender without the independent variable. The authors also conducted the mortality analysis on people without T1DM to evaluate the prevalence and risk associated with both conditions. Lastly, they performed a sensitivity analysis by restricting the look-back period for hospitalizations and emergency department visits to 5 years.

Out of 20,035 people with T1DM, 168 (0.8%) had an eating disorder. Typically, patients diagnosed with an eating disorder were female (92.9%), older (~27 years old), and had a long history of T1DM (48.8% with a duration of ≥15 years). Baseline characteristics were similar between the two groups except that fewer people in the experimental group lived in a rural area or had a diabetes duration between 5 to 10 years. The results showed DKA incidence was 112.5 per 1000 person-years in those with both T1DM and an eating disorder and 30.8 per 1000 person-years in those without an eating disorder. Overall, those with both conditions had about a three-fold increase in the risk of DKA (HR 3.30 [2.58 to 3.23]; p <0.0001), with females having a slightly higher risk of DKA as compared to males (HR 3.41 vs. 2.47, respectively). The authors also found that those with an eating disorder and T1DM had about a six-fold higher risk of all-cause mortality (HR 5.8 [3.04-11.08]; p < 0.0001). Mortality in the experimental group was 16 per 1000 person-years and 2.5 per 1000 person-years in the control group. When looking at people without T1DM, the prevalence of an eating disorder was much lower (0.2%). Additionally, the risk of all-cause mortality (HR 5.83; p < 0.0001) was similar to that found in the primary analysis. Income nor rurality did not impact the risk of mortality or DKA. Finally, the sensitivity analysis with a look-back period of 5 years also reported a higher risk of DKA (p<0.0001) and all-cause mortality (p<0.0001) in those with T1DM and an eating disorder.

Based on the results from this study, patients with T1DM and a history of an eating disorder have a significantly higher risk of mortality and DKA. Of equal importance, patients with T1DM are also more likely to develop an eating disorder than those without T1DM. As both conditions are more likely to appear at younger ages, patients should be screened for signs and symptoms of eating disorders for early diagnosis to prevent comorbidities and mortality. Per Scheuing et al., higher baseline A1c levels and elevated risk of other complications like retinopathy can be expected in this subgroup of patients. In future studies, it would be interesting to see how glycemic control improves before and after therapy for an eating disorder. 

Practice Pearls

• Patients with an eating disorder and T1DM are associated with a three-fold increase in DKA incidence and a six-fold increase in all-cause mortality.
• Both conditions tend to appear early in life, and patients with T1DM are more likely to develop an eating disorder.
• Early screening of eating disorders is crucial for patients with T1DM to prevent comorbidities and mortality. 

Gibbings, Nicole K et al. “Diabetic Ketoacidosis and Mortality in People With Type 1 Diabetes and Eating Disorders.” Diabetes care vol. 44,8, Aug. 2021: 1783-1787.

Young, V et al. “Eating Problems in Adolescents with Type 1 Diabetes: A Systematic Review With Meta-analysis.” Diabetic medicine: a journal of the British Diabetic Association vol. 30,2, 22 Aug. 2012: 189-98.

Scheuing, Nicole et al. “Clinical Characteristics and Outcome of 467 Patients With a Clinically Recognized Eating Disorder Identified Among 52,215 Patients With Type 1 Diabetes: a Multicenter German/Austrian Study.” Diabetes care vol. 37,6, June 2014: 1581-9.

Andy Dao, PharmD Candidate, University of South Florida Taneja College of Pharmacy