Clinical reports suggest that interpersonal problems are associated with the onset and maintenance of eating pathology, but existing measures of such problems have limited links to eating pathology. Therefore, the aim of this study was to develop an eating-specific measure of interpersonal problems. The new measure, the Interpersonal Relationships in Eating Disorders scale (IR-ED), was administered to a large community sample, a nonclinical replication sample, and a clinical group of eating disorder patients. In Study 1, the psychometric properties of the IR-ED were established, and they were tested using confirmatory analyses in Study 2. Study 3 determined the validity of the test score interpretations in a clinical sample. The final 15-item version of the IR-ED demonstrated 3 distinct factors with reliability of test scores-Food-Related Isolation; Avoidance of Body Evaluation; and Food-Related Interpersonal Tension. Study 2 demonstrated that the IR-ED comprises a common Interpersonal Problems factor and a specific group factor-Avoidance of Body Evaluation. Study 3 showed that the clinical group had higher IR-ED scores than a nonclinical group. Across the studies, Avoidance of Body Evaluation was the strongest correlate of eating pathology in this group. The IR-ED has strong psychometric properties and its test scores appear to be more valid than those of a generic measure of interpersonal problems. Avoidance of Body Evaluation is the strongest facet of such interpersonal problems, and has meaningful links to models of eating psychopathology.