Eur Arch Psychiatry Clin Neurosci (2013) 263:353–364 DOI 10.1007/s00406-012-0366-z ORIGINAL PAPER The psychiatric interview: validity, structure, and subjectivity Julie Nordgaard • Louis A. Sass • Josef Parnas Received: 2 May 2012 / Accepted: 28 August 2012 / Published online: 23 September 2012 Ó The Author(s) 2012. This article is published with open access at Springerlink.com Abstract There is a glaring gap in the psychiatric liter- order to perform faithful distinctions in this particular ature concerning the nature of psychiatric symptoms and domain, we need a more adequate approach, that is, an signs, and a corresponding lack of epistemological dis- approach that is guided by phenomenologically informed cussion of psycho-diagnostic interviewing. Contemporary considerations. Our theoretical discussion draws upon clinical neuroscience heavily relies on the use of fully clinical examples derived from structured and semi-struc- structured interviews that are historically rooted in logical tured interviews. We conclude that fully structured inter- positivism and behaviorism. These theoretical approaches view is neither theoretically adequate nor practically valid marked decisively the so-called ‘‘operational revolution in in obtaining psycho-diagnostic information. Failure to psychiatry’’ leading to the creation of DSM-III. This paper address these basic issues may have contributed to the attempts to examine the theoretical assumptions that current state of malaise in the study of psychopathology. underlie the use of a fully structured psychiatric interview. We address the ontological status of pathological experi- Keywords Consciousness Á Epistemology Á Sign Á ence, the notions of symptom, sign, prototype and Gestalt, Psychiatric interview Á Subjectivity Á Symptom and the necessary second-person processes which are involved in converting the patient’s experience (originally Exposure to swans and geese plays an essential role in lived in the first-person perspective) into an ‘‘objective’’ learning to recognize ducks (third person), actionable format, used for classification, Thomas Kuhn [37] treatment, and research. Our central thesis is that psychi- atry targets the phenomena of consciousness, which, unlike somatic symptoms and signs, cannot be grasped on the Introduction analogy with material thing-like objects. We claim that in Highly structured interviews have become the gold stan- dard of diagnostic interviewing in psychiatry, primarily in J. Nordgaard (&) research but also, increasingly, in ordinary clinical work. Psychiatric Center Hvidovre, University of Copenhagen, The literature on psychiatric interviewing usually deals Broendbyostervej 160, 2605 Broendby, Denmark e-mail: [email protected] with comparisons of the relative efficacy (degrees of sen- sitivity and specificity) and reliability of particular inter- L. A. Sass view approaches. Typically, these discussions fail to GSAPP-Rutgers University, 152 Fredlinghuysen Road, address the more overarching theoretical issue, namely: Piscataway, NJ 08854, USA e-mail: [email protected] What is the epistemologically adequate manner of obtain- ing psycho-diagnostic information? Even the textbooks and J. Parnas chapters devoted to psychiatric interview tend to be mute Psychiatric Center Hvidovre and Center for Subjectivity on the theoretical underpinnings of interviewing [2]. We Research, University of Copenhagen, Njalsgade 140-142, 2300 Copenhagen, Denmark have not, in fact, found in the literature any single contri- e-mail: [email protected] bution that systematically addresses ontological and 123 354 Eur Arch Psychiatry Clin Neurosci (2013) 263:353–364 epistemological foundations (‘‘ontological’’ refers here to ‘‘Yes/no’’ answers never suffice but always require the nature of being of psychiatric symptom and sign) of the exemplifications in the patient’s own words. The psychiatric interview (although a recent paper by Stang- specifically phenomenological aspects will be articu- hellini [67] comes close). In other words, there seems to be lated in the course of the paper, but briefly put; this an important lacuna in the psychiatric literature concerning approach aims at a faithful recreation of the patient’s the interview-relevant basic concepts on the nature of subjective experience. symptom and sign (what Berrios calls the ‘‘psychiatric The issue at hand touches upon many topics from the object’’ [4, 42] and the methods used to elicit and describe philosophy of science, philosophy of mind (viz., con- them. The task of this paper is to address this lack in a way sciousness, its description, psychophysical relation, etc.), that is theoretically coherent and reflects practical clinical cognitive neuroscience, semantics and semiotics (theory of reality. meaning), linguistics, anthropology, and affective science. The goal of a psychiatric assessment is to describe the An exhaustive, scholarly review of all these issues is patient’s complaints, appearance, and existence in an obviously beyond our scope. We restrict ourselves to a few actionable psychopathological format, namely, one that indispensable and pragmatically relevant aspects of clinical results in diagnostic classification and other clinical deci- phenomenology and philosophy of mind. sions. This process includes, to a large degree, describing the patient’s experiences, originally lived in the first-person perspective, in potentially third-person terms, thus pro- The origins of structured interview viding ‘‘objective’’ data that can be shared for diagnosis, treatment, and research. We exclude from consideration ‘‘a The development of the structured interview was prompted free-style clinical interview’’ which grants the clinician by the need for improving reliability of psychiatric total liberty, thus failing to prevent limited comprehen- assessments. As is well known, the WHO-sponsored siveness (due to lack of systematic exploration of psy- US–UK diagnostic project [9] demonstrated markedly chopathology) or guard against incompetence. This type of different diagnostic habits of British and American clini- interview has been shown to be notoriously unreliable [13]. cians. It was clear from these studies that a science of For the sake of illustration, we will articulate, as a part psychiatry was not possible without strengthening the of our presentation, a contrast between two types of reliability of psychiatric assessments. The project also interviewing: demonstrated that the diagnostic differences could be 1. a fully structured psychiatric interview, performed by a minimized by using a standardized structured interview clinician psychiatrist or psychologist or even a non- and shared diagnostic criteria [9]. clinician (a student, a nurse, etc.) who has been The US–UK study served as an important impetus for the specifically trained for this purpose. This sort of ‘‘operational revolution,’’ leading to criteria-based diagno- structured interview consists of asking the patient pre- ses, ‘‘operational’’ definitions of such criteria, and a strong specified questions in fixed sequence and rating the emphasis on interrater agreement, a development vigor- responses as positive, negative, or threshold [16]. The ously spearheaded by an influential psychiatric group from fully structured interview relies on a series of Washington University in St Louis, Missouri (the so-called assumptions that we wish to explore. Neo-Kraepelinian movement). The New York Post—in a 2. a conversational, phenomenologically oriented, semi- very enthusiastic tone—described these first attempts as ‘‘a structured interview, performed by an experienced and new tool that rolled psychiatrist’s thermometer, microscope reliability-trained psychiatrist. The ‘‘structured’’ com- and X-ray machine into one’’ (quoted in [66]). The criteria ponent in the ‘‘semi-structured interview’’ consists of a of diagnostic categories, eventually summarized in the list of items (typically, an aggregate of relevant scales) DSM-III? and ICD-10, became, with the passage of time, on which the interviewer must elicit sufficient infor- the catalog of officially sanctioned symptoms and signs, mation in order to score these items after completing while the remaining psychopathological features largely interview session. Here, however, the flow of the went into oblivion [1] and are no longer mentioned in the interview is conversational. Questions are contextually major textbooks. The interview schedules are constructed to adapted and follow the train of the patient’s narrative, be as directly compatible with the diagnostic criteria as yet with a constant opportunity to ask for more detail possible, to the point that the criteria are often used as the or further examples (this includes the possibility of a interview questions. gentle interrupting and changing the direction of the Robert Spitzer, an important figure behind the DSM- interview). Spontaneity, recollection, and reflection III? project, justified the creation and use of the structured on the part of the patient are strongly encouraged. psychiatric interviews in his famous paper: ‘‘Are clinicians 123 Eur Arch Psychiatry Clin Neurosci (2013) 263:353–364 355 still necessary?’’ [66]. Interviews, he argued, could be more Historically speaking, the most comprehensive analysis reliably performed by economically affordable, naı¨ve raters of psychiatry’s theoretical foundations was offered in who would stick to the pre-formed questions than by cli- successive editions from 1913 to 1954 by a German psy- nicians who were both expensive