Differences in DSM-III-R and DSM-IV diagnoses in eating disorder patients.

Two hundred eighty-eight eating disorder patients were administered the DSM-III-R Structured Clinical Interview (SCID) and the DSM-IV SCID for axis I and II. Concordance between DSM-III-R and DSM-IV was excellent for the axis I affective and anxiety disorders, bulimia nervosa, and substance abuse/dependence. It was also excellent for axis II paranoid, schizoid, borderline, and antisocial personality disorders. Agreement between the two nosological systems was lower for alcohol abuse/dependence with a kappa of.63. Kappas were also poor for the following personality disorders: schizotypal (.44), histrionic (.29), dependent (.54), obsessive-compulsive (.62) and not otherwise specified (.63). There was a substantial difference in the diagnosis of anorexia nervosa between DSM-III-R and DSM-IV. Fourteen patients were diagnosed with anorexia nervosa, binge/purge type, using DSM-IV criteria, while only six received the diagnoses of anorexia nervosa and bulimia nervosa using DSM-III-R criteria. Kappa was.49 and the percent agreement was 79%. While there are considerable areas of overlap in DSM-IV and DSM-III-R, there are also areas of substantial differences. Clinicians and researchers must be very cautious when attempting to compare data from the different nosologies.