Identifying Individuals with Eating Disorders Using Health Administrative Data

Area of Science:

• Public Health
• Epidemiology
• Health Services Research

Background:

• Eating disorders represent a significant public health concern with a substantial burden.
• Current clinically relevant data sources for eating disorders are limited, hindering accurate burden assessment.
• A lack of comprehensive data impedes public health research on eating disorders compared to other mental health conditions.

Purpose of the Study:

• To test the feasibility of creating large, clinically relevant eating disorder cohorts using health administrative data.
• To assess the utility of administrative data in measuring the burden of eating disorders.
• To address the existing evidence gap in eating disorder public health research.

Main Methods:

• Development of three distinct eating disorder prevalence cohorts using health administrative data from Ontario, Canada (1990-2014).
• Cohort 1: Hospitalization with eating disorder as primary diagnosis.
• Cohort 2: Hospitalization with eating disorder as any diagnosis.
• Cohort 3: Cohort 2 plus emergency department visits with an eating disorder diagnosis.

Main Results:

• Generated cohorts of 7,268 (Cohort 1), 13,197 (Cohort 2), and 17,373 (Cohort 3) patients.
• Increased cohort size correlated with a higher proportion of diagnoses for bulimia nervosa and eating disorder not otherwise specified.
• Demographic and clinical differences within cohorts were minimal compared to differences with the general Ontario population.

Conclusions:

• Health administrative data are a feasible resource for measuring the clinically relevant burden of eating disorders.
• The developed cohorts demonstrated significant differences in diagnostic composition.
• Utilizing administrative data can effectively bridge the evidence gap in eating disorder research and public health surveillance.