Phase 1 - Develop an initial instrument and start to establish content validity. What do dually diagnosed patients believe are the most important, relevant, and significantly contributing constructs that we should measure in order to capture those individuals with type 1 diabetes who may exhibit symptoms of an eating disorder? What individual items will best measure these constructs?

Phase 2 - Conduct individual interviews for cognitive testing of our instrument and further examine its content validity. How well do the individual items included in our instrument capture eating disorder symptoms in patients with type 1 diabetes? Do these items accurately capture the information we seek to collect? What symptoms have we not considered? How readable and understandable are our questions?

Phase 3 - Explore the internal consistency reliability and construct validity of our instrument. Do the items in our instrument accurately capture the constructs we wish to measure? Do similar items "hang together" in a manner that is clinically and theoretically meaningful?

Phase 4 - Address convergent and discriminant validity as well as test-retest reliability. Does our instrument demonstrate appropriate convergent and divergent validity as well as test-retest reliability? Does our instrument capture eating disorder symptoms among persons with diabetes? Is our instrument more specific to the population of individuals with diabetes than traditional eating disorder instruments (i.e., does our instrument capture more information and give a more accurate presentation than the other instruments)?