Use the EAT-26 to help you determine if you need to speak to a mental health professional or a physician and get help for an eating disorder. It will take you about 2 minutes to complete. Take the EAT-40. The EAT-40 is the original version of the Eating Attitudes Test. The 40-item version was shortened to 26-items (EAT-26) based on a factor analysis. - - - - - - - - - - - - - - - - - - The EAT-26 is the most widely used screening measure that may be able to help you determine if you have an eating disorder that needs professional attention. The EAT-26 is not designed to make a diagnosis of an eating disorder or to take the place of a professional diagnosis or consultation. Please answer each question as accurately, honestly, and completely as possible. All of your results are completely confidential. - - - - - - - - - - - - - - - - - - Go to the Downloads page to download a copy of the EAT-26, as well as instructions regarding how to score and interpret the test.

RESOURCES Academy for Eating Disorders River Centre Clinic National Eating Disorder Assoc. IAEDP ANAD EDC - - - - - - - - - - - - - - - - - -

SCORING

Referral Criteria A, B and C

•	EAT-26 Test Score:

Test Items and Total Test Score:

	•	Items 1-25 are scored as follows: Always = 3; Usually = 2; Often = 1; Other answers = 0
	•	Item 26 is scored in the opposite direction (Never = 3 etc.)
	•	Total Test Score: Add item scores for a "total test score."

•

Is the total 20 or more? No Yes, make a referral

•	Behavioral Questions:

•

Did the respondent check any of the boxes as shown below? No Yes, make a referral

•	Underweight:

Determine Body Mass Index (BMI) to determine if the respondent is underweight.

Divide weight in pounds by height in inches and then by height in inches again and then multiply by 703

Example: 5’4” = 64 inches. Weight = 105 lbs. So, (105 ÷ 64=1.641) (1.64 ÷ 64=.0256) (X 703= 18.0)

Below are the BMI thresholds for defining if the respondent is underweight by age and sex.

•

Is the total 20 or more? No Yes, make a referral

What to do about Missing Data:

There is no agreed-upon procedure for dealing with missing data for the EAT-26. However, it is recommended that interpolation be used to estimate up to one missing value for each EAT-26 subscale using the median subscale item value (rounding up to a whole number) to construct the missing data point. This seems to achieve a fair balance between including questionnaires in which a small number of values may have been inadvertently omitted, while at the same time maintaining data integrety by excluding questionnaires with more missing data.