Structured and Semistructured Interviews for Differential Diagnosis

Chapter 3
Structured and Semistructured Interviews for Differential Diagnosis
Fundamental Issues, Applications, and Features

Daniel L. Segal and Kadija N. Williams

Structured and semistructured interviews were developed to address the difficulties that clinicians and researchers historically have had in making accurate diagnoses of mental disorders with traditional unstructured clinical interviews. A major contributing factor to diagnostic imprecision was the lack of uniformity or standardization of questions asked of respondents to evaluate the nature and extent of their psychiatric symptoms and to arrive at a formal diagnosis. Structured and semistructured interviews solve this problem by their very nature. As such, they have become increasingly popular and effective in the mental health field, leading to vastly improved diagnostic clarity and precision. The purpose of this chapter is to provide a basic introduction to structured and semistructured interviews used to assess and diagnose psychopathology among adults. We begin with a discussion of the basic types of applications of structured and semistructured interviews followed by an exploration of their major features, advantages, and drawbacks. We conclude this chapter with a discussion of the most popular multidisorder structured and semistructured interviews used to diagnose clinical disorders and personality disorders.

With the recent publication of the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5; American Psychiatric Association [APA], 2013), many changes have occurred, especially regarding the classification and organization for many clinical disorders. One major change was the removal of the multiaxial system, which previously denoted an important distinction between clinical disorders and personality disorders on separate diagnostic axes. Notably, no changes were made to the classification and diagnostic criteria for the personality disorders. In this chapter, we describe the anticipated changes to some of the major structured and semistructured interviews as these measures become updated from their linkage to the DSM-IV-TR (APA, 2000) to the DSM-5.

Basic Issues Regarding Structured and Semistructured Interviews

The most common method among mental health professionals to evaluate and diagnose their clients is the direct clinical interview (Segal & Hersen, 2010). Such interviews, however, can vary tremendously, especially regarding the amount of structure that is imposed. Indeed, some important differences exist between less structured interviews and more structured ones. Unstructured clinical interviews are dependent on the clinician’s unique background, knowledge base, theoretical model, and interpersonal style, and thus are highly flexible. Within this unstructured approach, clinicians are entirely responsible for asking whatever questions they decide are necessary to reach a diagnostic conclusion. In fact, any type of question or topic (relevant or not) can be pursued in any way that fits the mood, preferences, training, specific interests, or philosophy of the clinician. As a consequence, one can imagine the variability across interviews from one clinician to another. On the other hand, structured interviews conform to a standardized list of questions (including follow-up questions), a uniform sequence of questioning, and systematized ratings of the client’s responses. These questions are designed to measure the specific criteria for many mental disorders as presented in the DSM. These essential elements of structured interviews serve several important purposes, most notably that their use:

  • Increases coverage of many mental disorders that otherwise might be overlooked.
  • Enhances the diagnostician’s ability to accurately determine whether particular symptoms are present or absent.
  • Reduces variability among interviewers, which improves reliability.

These features of structured interviews add much in developing clinical psychology into a true science. For example, structured interviews are subject to evaluation and statistical analysis, and they can be modified and improved based on the published literature regarding their psychometric properties.

Not all structured interviews are the same. In fact, the term structured interview is broad, and the actual amount of structure provided by an interview varies considerably. Structured interviews can be divided into one of two types: fully structured or semistructured. In a fully structured interview, questions are asked verbatim to the respondent, the wording of probes used to follow up on initial questions is specified, and interviewers are trained to not deviate from this rigid format. In a semistructured interview, although initial questions for each symptom are specified and are typically asked verbatim to the respondent, the interviewer has substantial latitude to follow up on responses. The interviewer can modify or augment the standard inquiries with individualized and contextualized probes to more accurately rate specific symptoms. The amount of structure provided in an interview clearly impacts the extent of clinical experience and judgment that are required to administer the interview appropriately: Semistructured interviews require clinically experienced examiners to administer the interview and to make diagnoses, whereas fully structured interviews can be administered by nonclinicians who receive training on the specific instrument. This latter difference makes fully structured interviews popular and economical, especially in large-scale research studies in which accurate diagnoses are essential.

Structured and semistructured interviews have been created to assist with the differential diagnosis of all major clinical disorders (formerly defined on Axis I of the DSM-IV) and all standard personality disorders (formerly defined on Axis II of the DSM-IV). Interviews used for psychiatric diagnosis are typically aligned with the DSM system and, therefore, assess the formal diagnostic criteria specified in the manual. However, structured interviews exist beyond those designed for DSM differential diagnosis. Other structured interviews are narrower in focus; for example, to assess a specific problem or form of psychopathology (e.g., eating disorders, substance abuse, borderline personality disorder features) in great depth. An excellent resource for information about a host of specialized interviews is provided by Rogers (2001). Our focus now turns to a discussion of some common functions of structured and semistructured interviews.

Applications of Structured and Semistructured Interviews

Whereas structured and semistructured interviews are used in many different venues and for many different purposes, their application falls into three broad areas: research, clinical practice, and clinical training.

Research. The research domain is the most common application, in which structured or semistructured interviews are used to formally diagnose participants for inclusion into a study so that etiology, comorbidity, and treatment approaches (among other topics) can be explored for a particular diagnosis or group of diagnoses. Sound empirical research on mental disorders certainly requires that individuals assigned a diagnosis truly meet full criteria for that diagnosis. Another research application for structured interviews is to provide a standardized method for assessing change in one’s clinical status over time. As noted by Rogers (2003), these types of longitudinal comparisons are essential for establishing outcome criteria, which is vital to diagnostic validity.

Clinical Practice. In clinical settings, administration of a structured or semistructured interview may be used as part of a comprehensive and standardized intake evaluation. Routine and complete administration of a structured interview is increasingly common in psychology training clinics, but doing so requires considerable training for clinicians and time for full administration. A variation on this theme is that sections of a structured interview may be administered subsequent to a traditional unstructured interview to clarify and confirm the diagnostic impressions. Widiger and Samuel (2005) provide another thoughtful alternative especially regarding the assessment of personality disorders in clinical practice. They recommend the strategy of administering an objective self-report inventory, which is followed by a semistructured interview that focuses on the personality disorders that received elevated scores from the testing. This strategy is responsive to time constraints in clinical practice but also allows for collection of standardized, systematic, and objective data from the structured interview. Finally, we wish to emphasize that in clinical settings, structured interviews should not take the place of traditional clinical interviews. Both can be performed, although at different times and for different purposes. The combination of the two approaches, integrated flexibly to meet the needs of the individual clinician and his or her clients, reflects the best of the scientist-practitioner model in which the science and art of assessment are both valued and valuable (Rogers, 2003).

Clinical Training. Use of structured or semistructured interviews for training mental health professionals is an increasingly popular and ideal application, because interviewers have the opportunity to learn (through repeated administrations) specific questions and follow-up probes used to elicit information and evaluate specific diagnostic criteria provided by the DSM. Modeling the questions, sequence, and flow from a structured interview can be an invaluable source of training for beginning clinicians.

Advantages and Disadvantages of Structured and Semistructured Interviews

No assessment device in the mental health field is perfect, and structured and semistructured interviews are no exception to this truism. In this section, the strengths and weaknesses of structured interviews are discussed. Our intention is to give readers an appreciation of the major issues to be considered when deciding whether to use the structured interview approach to assessment. A brief summary of the advantages and disadvantages is presented in Table 3.1.

Table 3.1 Advantages and Disadvantages of Structured and Semistructured Interviews

Advantages Disadvantages
Increased Reliability: Because questions are standardized, structured interviews decrease variability among interviewers, which enhances interrater reliability. Structured interviews also increase the reliability of assessment for a client’s symptoms across time, as well as the reliability between client report and collateral information. May Hinder Rapport: Use of structured interviews may damage rapport because they are problem-centered, not person-centered, and poorly trained interviewers may neglect to use their basic clinical skills during the assessment.
Increased Validity: Structured interviews assure that diagnostic criteria are covered systematically and completely. This is important because it serves to increase the validity of diagnosis. Limited by the Validity of the Classification System Itself: Structured interviews used for diagnosis are inherently tied to diagnostic systems. Thus, they are only as valid as the systems upon which they are based. Furthermore, it is difficult to establish the validity of particular structured interviews because there is no gold standard in psychiatric diagnosis.
Utility as Training Tools: Structured interviews are excellent training tools for clinicians in training because structured interviews promote the learning of specific diagnostic questions and probes used by experienced clinical interviewers. In addition, nonclinicians can easily be trained to administer fully structured interviews, which can be cost effective in both research and clinical settings. The Trade-Off of Breadth Versus Depth: Structured interviews are limited because they cannot cover all disorders or topic areas. When choosing a structured interview, one must evaluate carefully the tradeoffs of breadth versus depth of assessment.

Advantage: Increased Reliability

Perhaps the most important advantage of structured interviews centers on their ability to increase diagnostic reliability (reliability defined in this context refers to consistency or agreement about diagnoses assigned by different raters; Coolidge & Segal, 2010a). By systemizing and standardizing the questions interviewers ask, and the way answers to those questions are recorded and interpreted, structured interviews decrease the amount of information variance in diagnostic evaluations (e.g., Rogers, 2001). That is, structured interviews decrease the chances that two different interviewers will elicit different information from the same client, which may result in different diagnoses. Thus, interrater reliability, or the likelihood that two different interviewers examining the same individual will arrive at the same diagnosis, is greatly increased.

Increased interrater reliability has broad implications in clinical and research settings. Because many psychological and psychopharmacological treatments are intimately tied to specific diagnoses, it is imperative that those diagnoses be accurate (Segal & Coolidge, 2001). Thus, if different clinicians interviewing the same client arrive at different diagnostic conclusions, it would be challenging to make a definitive decision about treatment. Similarly, accurate diagnosis is also essential for many types of clinical research, for example, studies that address causes and treatments of specific forms of psychopathology (Segal & Coolidge). Imagine a study examining different treatments for bipolar disorder. In such a study, it would be imperative to be certain that those individuals in the treatment groups actually have accurate diagnoses of bipolar disorder. Researchers must be able to accurately and definitively diagnose participants with the disorder being studied before researchers can even begin to examine theories of etiology or the effectiveness of treatment for that particular mental disorder.

In addition to increasing interrater reliability, structured interviews increase the likelihood that the diagnosis is reliable across time and across different sources of information (Rogers, 2001). In many clinical and research settings, individuals are in fact assessed on different occasions. Making multiple assessments could be dangerous if an interviewer evaluates a client in a different manner with different questions on different occasions. The client’s presentation may be substantially altered because the manner in which the client is asked about those symptoms has changed instead of the client’s symptoms or diagnosis being different. Using a standardized interview for multiple assessments helps ensure that if a client’s presentation has changed, it is because his or her symptoms are actually different, not because of variance in interviews (Rogers). Likewise, in many settings, clinicians conduct collateral interviews with significant people in the client’s life to glean a broader picture of the client’s symptoms, problems, and experiences. Using a structured interview for both a client and a collateral source will likely increase the chances that discrepancies between the client and collateral informant are real, rather than a consequence of different interviewing styles (Rogers).

Advantage: Increased Validity

Validity of psychiatric diagnosis refers to the meaningfulness or usefulness of the diagnosis (Coolidge & Segal, 2010b). A required prerequisite for validity is reliability. Thus, by virtue of the fact that structured interviews greatly increase reliability of diagnosis, they also increase the likelihood that the diagnosis is valid. Structured interviews also improve the validity of diagnoses in other ways. The systematic construction of structured interviews lends a methodological validity to these types of assessments compared to unstructured approaches. Because structured interviews are designed to thoroughly and accurately assess well-defined diagnostic criteria, they are often better assessments of those criteria than unstructured interviews (Rogers, 2001). According to Rogers, clinicians who use unstructured interviews sometimes diagnose too quickly, narrow their diagnostic options too early, and miss comorbid diagnoses. Because structured interviews essentially force clinicians to assess all of the specified criteria for a broad range of diagnoses, they offer a more thorough and valid assessment of many disorders compared to unstructured interviews.

In our experience, it is common for beginning clinicians who are performing an unstructured clinical interview to gather information about the presence or absence of only a few common mental disorders. Coverage of other disorders may be neglected during an unstructured interview if the interviewer is unfamiliar with the specific criteria of some disorders. Some unstructured interviews may also provide limited information about whether comorbid psychopathology exists or about the severity of the psychopathology. Because they incorporate systematic ratings, structured and semistructured interviews easily provide information that allows for the determination of the level of severity and the level of impairment associated with a particular diagnosis. Structured interviews provide the same information about comorbid disorders as well.

Advantage: Utility as Training Tools

Structured interviews can be invaluable training tools for beginning mental health professionals as well as experienced clinicians who desire to enhance their diagnostic skills. Use of structured interviews in the training context may help clinicians develop or enhance their understanding of the flow, format, and questions inherent in a comprehensive diagnostic interview. With repeated administrations, much of a structured interview can be internalized by the clinician. In addition, use of structured interviews for training may reduce anxiety, especially among neophyte clinicians, because the format and flow of the interview is laid out clearly outlined for the interviewer. This type of structure can be helpful and calming for beginning clinicians, who may be initially overwhelmed by the diagnostic process and its inherent complexity.

Structured interviews can also be a useful means of training those who make preliminary mental health assessments, for example, intake staff at hospitals, so that clients are thoroughly and accurately evaluated in preparation for treatment planning. In the case of nonclinician interviewers, fully structured interviews are advisable because they minimize the amount of clinical judgment needed for accurate administration. Use of these trained paraprofessionals can make large-scale research studies cost effective.

Disadvantage: May Hinder Rapport

Despite the advantages of structured interviews, their application is not without controversy. The most common criticism of structured interviews is that their use may damage rapport or the therapeutic alliance (Segal, Maxfield, & Coolidge, 2008), which is widely viewed as an essential component of effective psychotherapy. Attaining a reliable and accurate diagnosis of a client achieves a hollow victory if the process prevents the therapeutic alliance from forming, or in a more dramatic example of clinical failure, the client does not return for continued treatment. The well-known joke poking fun at medicine, “the operation was a success but the patient died,” might be recast in terms of structured interviews as “the diagnosis was impeccable but the client never came back for another session.”

How exactly might structured interviews damage rapport? Perhaps most importantly, structured interviews may impede the connection between client and clinician because interviews are problem-centered rather than person-centered. There is a danger that interviewers may get so concerned with the protocol of their interview that they fail to demonstrate the warmth, empathy, and genuine regard necessary to form a therapeutic alliance. Indeed, the standardization of the interview may play out as “routinization” (Rogers, 2003). In addition, interviewers who are overly focused on the questions that they must “get through” in an interview may, as a consequence, miss important behavioral cues or other information that could prove essential to the case.

Proponents of structured interviews note that the problem of rapport-building during a structured interview can be overcome with training, experience, and flexibility (Rogers, 2003). We concur and emphasize the observation that “rapid inquiries or monotonous questioning represents clear misuses of structured interviews” (Rogers, 2003, p. 22). If interviewers make an effort to use their basic clinical skills, structured interviews can and should be conducted in such a way that establishes rapport and enhances understanding of the client. To ensure that this is the case, however, interviewers must be aware of the potential negative effects of structured interviews on rapport-building and make the nurturance of the therapeutic alliance a prominent goal during an interview, even when they are also focused on following the protocol. It behooves those who use structured interviews to engage their respondents in a meaningful way during the interview and to avoid a rote-like interviewing style that may alienate. On the other hand, not all clients have a negative perception of a structured interview that must be intentionally overcome. Some clients actually like the structured approach to assessment because it is perceived as thorough and detailed, and in these cases, rapport is easily attained.

Disadvantage: Limited by the Validity of the Classification System Itself

Earlier, we noted that proponents of structured interviews claim structured interviews may render more valid diagnoses in general. The assumption inherent in this argument is that the DSM diagnostic criteria are inherently valid, which is a debatable point. One should recognize that DSM diagnostic criteria were developed to operationalize theoretical constructs (e.g., depression, panic disorder, schizophrenia) so there is no absolute basis on which criteria were created. Furthermore, mental disorders are social constructions, which implies that they evolve over time as societies evolve.

Although successive editions of the DSM have been better grounded in empirical research, and the criteria for some disorders (e.g., major depression) have solid research support, other disorders (e.g., most of the personality disorders) and their criteria have not been examined as consistently or as completely, therefore leaving questions about their validity (Widiger & Trull, 2007). This point is also bolstered by the fact that the criteria for some disorders have changed significantly from one edition to another in the evolution of the DSM (Coolidge & Segal, 1998; Segal, 2010). Furthermore, criteria for many disorders in the DSM are impacted by cultural and subcultural variations in the respondent (see Chapter 4 in this book), as well as by the age of the respondent. Indeed, the diagnostic criteria for many mental disorders do not fit the context of later life and, therefore, some criteria do not adequately capture the presentation of the disorders among many older adults (e.g., Segal, Coolidge, & Rosowsky, 2006; Segal, Qualls, & Smyer, 2011). Thus, certain criteria may be valid only for a particular group of individuals, at a particular point in time, at a particular age. The primary method clinicians currently use to conceptualize diagnoses (the DSM), while improving, is far from perfect. Because the DSM generally does poorly in attending to these issues of age and diversity, interviews based on poor-fitting diagnostic criteria are similarly limited.

In addition to potential problems with DSM diagnostic criteria, another issue regarding structured interviews is that it is challenging to establish firmly the validity of any particular structured interview. The quandary is that our best means of establishing the validity of a structured interview is to compare diagnoses obtained from such interviews to diagnoses obtained by expert clinicians or by other structured interviews. This is inherently problematic because we cannot be certain that diagnoses by experts or other structured interviews are in fact valid in the first place (Segal et al., 2008).

Disadvantage: The Trade-Off of Breadth Versus Depth

A final criticism of structured interviews centers on the fact that no one structured interview can be all things in all situations, covering all disorders and eventualities. For example, if a structured interview has been designed to cover an entire diagnostic system (like the DSM, which identifies several hundred specific mental disorders), then inquiries about each disorder must be limited to a few inclusion criteria. In this case, the fidelity of the official diagnostic criteria has been compromised for the sake of a comprehensive interview. If the fidelity of the criteria is not compromised, then the structured interview becomes unwieldy in terms of time and effort required for its full administration. Most structured interviews attempt some kind of compromise between these two points of tension.

Thus, regarding breadth versus depth of approach, users of structured interviews are forced to make a choice about what is most useful in a given situation. Both choices have their limitations. If clinicians or researchers decide to use an interview that provides great breadth of information, they ensure that a wide range of disorders and a great many different areas of a respondent’s life are assessed. However, one may not have the depth of information needed to fully conceptualize a case. On the other hand, deciding to use an interview focused on a few specific areas will provide clinicians and researchers with a wealth of information about those specific areas, but it may result in missing information that could lead to additional diagnoses or a different case conceptualization. Thus, it is essential to understand that when choosing a particular structured or semistructured interview, there are often tradeoffs regarding breadth and depth of information.

Weighing Advantages and Disadvantages

Our examination of the strengths and limitations of structured interviews highlights the importance of carefully contemplating what is needed in a particular clinical or research situation before choosing a structured interview. Structured interviews can be invaluable tools in both clinical and research work; however, it is essential that one does not use such tools without accounting for some of the problems inherent in their use. Rogers (2001) voiced the helpful perspective that it would be unwise to view the interviewing process as an either/or proposition (i.e., unstructured vs. structured interview). In certain situations, unstructured interviews may meet the objectives of a particular clinical inquiry more efficiently than a structured interview. For example, in a crisis situation, flexibility on the part of the clinician is required to meet the pressing demands of this fluid and potentially volatile interaction. However, in other cases, greater assurances that the diagnostic conclusions are valid and meaningful would take priority, for example, in clinical research or in the delivery of clearly defined psychotherapeutic intervention protocols. As noted earlier, the integration of a nonstandardized or clinical interview with a structured or semistructured interview may also be an excellent option for clinicians and researchers.

Finally, despite some potential limitations to the use of structured and semistructured interviews, their use has clearly revolutionized the diagnostic process, vastly improving diagnostic reliability and validity. Such interviews have greatly improved clinical and research endeavors by providing a more standardized, scientific, and quantitative approach to the evaluation of specific symptoms and mental disorders. As such, it is likely that the use of structured and semistructured interviews will increase in the coming decades.

Structured and Semistructured Interviews for Differential Diagnosis

In this section, we examine several popular structured and semistructured interviews. These interviews can be divided into those that focus on either clinical disorders or personality disorders. As noted earlier, although the DSM-5 no longer makes a distinction between clinical disorders and personality disorders (with the replacement of the multiaxial coding system with a nonaxial coding system), the distinction is still relevant to the current crop of structured and semistructured interviews that were developed with this difference in mind. Instruments that focus on clinical disorders include the Anxiety Disorders Interview Schedule for DSM-IV, Diagnostic Interview Schedule for DSM-IV, the Schedule for Affective Disorders and Schizophrenia, and the Structured Clinical Interview for DSM-IV Axis I Disorders. Instruments that measure personality disorders include the Diagnostic Interview for DSM Personality Disorders, the International Personality Disorder Examination, the Structured Clinical Interview for DSM-IV Axis II Personality Disorders, and the Structured Interview for DSM-IV Personality.

Where possible, we describe forthcoming updates to some of these instruments as they are revised to conform to the current DSM-5 system. A general overview of each instrument is provided in Table 3.2. Each instrument assesses a variety of mental disorders and, therefore, can assist in the important task of differential diagnosis (i.e., a systematic way of discriminating among numerous possible disorders to identify specific ones for which the client meets the diagnostic threshold). Each interview also allows for an assessment of many comorbid mental disorders. The instruments presented in this chapter do not represent an exhaustive list of structured and semistructured interviews, but they are among the most common and well-validated ones. Interested readers are referred to Rogers (2001) and Summerfeldt, Kloosterman, and Antony (2010) for coverage of instruments not reviewed in this chapter.

Table 3.2 Comparison of Major Diagnostic Interviews

Only gold members can continue reading. Log In or Register to continue

Jun 10, 2016 | Posted by in PSYCHOLOGY | Comments Off on Structured and Semistructured Interviews for Differential Diagnosis
Premium Wordpress Themes by UFO Themes
Name Time Required Format Comment
Anxiety Disorders Interview Schedule for DSM-IV (Brown, DiNardo, & Barlow, 1994a) 45 to 60 minutes Semistructured, interviewer administered Provides in-depth assessment of anxiety disorders and other frequently comorbid conditions (e.g., mood disorders, substance abuse). Designed to be administered by trained mental health professionals with training in administration. Available in separate current and lifetime versions.
Diagnostic Interview Schedule for DSM-IV (Robins et al., 2000) 90 to 150 minutes