Some Problems of Validity With the Psychiatric Status Schedule as an Instrument for Case Identification and Classification in the General Population

To the Editor.—  Spitzer et al1 have criticized our article² about the Psychiatric Status Schedule (PSS) on the grounds that we were confused about the appropriate criteria with which to evaluate its adequacy for the purposes of epidemiologic research. More specifically, they argue that our reliance on internal consistency reliability was misplaced since low internal consistency is an underestimate of true reliability and therefore does not necessarily invalidate the PSS scales. Spitzer and his colleagues believe that the PSS can be used validly in the general population for "case finding," a term we will attempt to clarify before addressing their major criticism.

Uses of the PSS.—  The PSS contains nearly 500 items. Scoring these items requires a considerable amount of prior clinical sophistication. It was never our intention to use such a demanding instrument for the purpose of differentiating "cases" from "noncases," regardless of diagnostic types. There are a