DESCRIPTION
HISTORY AND SCOPE
PSYCHOMETRIC PROPERTIES
The Diagnostic Interview Schedule (DIS) Version IV is a fully structured questionnaire designed to ascertain the presence or absence of major psychiatric disorders as outlined in the Diagnostic and Statistical Manual of Mental Disorders (4th ed. [DSM-IV]; American Psychiatric Association [APA], 1994). Earlier versions of the DIS have been used since 1980 to reflect earlier versions of the DSM. These include versions of the DIS designed to reflect DSM-III (APA, 1980) and DSM-III-R (APA, 1987). Like the earlier versions, the DIS Version IV attempts to mimic a clinical interview by using questions to determine whether psychiatric symptoms endorsed by a respondent are clinically significant and are not explained by medical conditions or substance use.
The DIS must be administered by trained interviewers, but these interviewers do not have to be clinicians. Due to the fully specified nature of the DIS, nonclinicians may administer the DIS with adequate reliability and validity. Thus, in many situations, the expense and complication of using clinicians to interview patients is not necessary when using the DIS.
The DIS is based on the logic and background of DSM-IV. This means that the strengths and weaknesses of such a diagnostic system are inherent in the DIS. The strength is that the diagnoses in DSM-IV have been developed based on nosological data and consensus among experts. The approach in DSM-IV is fundamentally grounded in a bio-psycho-social approach to psychopathology, which does not include inference about causation of symptoms (except in such cases as post-traumatic stress disorder or the substance-induced conditions in which the etiological agent is a specific external phenomenon). On the other hand, the weakness of strict adherence to a particular diagnostic system is that over time, research may demonstrate that other ways of classifying persons with psychiatric symptoms may be more effective. Thus, data collected using a particular diagnostic system may become obsolete over the course of a longitudinal study.
HISTORY AND SCOPE
The DIS was first developed in 1978 at the request of the National Institute of Mental Health (NIMH). At that time, the NIMH Division of Biometry and Epidemiology was beginning to organize its Epidemiological Catchment Area (ECA) Program (Robins & Regier, 1991) and needed a comprehensive diagnostic instrument for a large-scale, multicenter epidemiological study that could be administered either by lay interviewers or by clinicians.
Because the DSM-III, published by the American Psychiatric Association in 1980, was to be the official diagnostic system for the country, DSM-III criteria were to be the basis for prevalence counts. To make the selected DSM-III diagnoses, a diagnostic interview had to be able to identify on a lifetime basis the presence and clinical significance of DSM-III criteria, the frequency and severity of symptoms, their temporal clustering, whether symptoms occurred in the absence of circumstances under which they would be part of a normal emotional response, whether symptoms occurred in the absence of physical illnesses or conditions that could account for them, and whether the presence of other psychiatric disorders might preempt the disorder of interest. In 1978, no interviews used in surveys of the general population performed all these tasks in a standard replicable fashion, and the one interview that came closest was the Renard Diagnostic Instrument (RDI) developed at Washington University in St. Louis.
The RDI had been written to operationalize the Washington University Department of Psychiatry interview, which was a list of symptoms serving the Feighner criteria, criteria developed at Washington University to make 14 major psychiatric diagnoses. Operationalizing these symptoms with explicit questions was facilitated by the participation of experienced psychiatrists in the department and transcriptions of their recorded uses of the departmental interview. The developers of the RDI were given primary responsibility for developing the new instrument for the ECA study.
Questions and probes from the RDI and its coding scheme were used and the RDI was adapted to make distinctions between current and past diagnoses and to add questions needed to make diagnoses according to DSM-III criteria. The first version of the DIS was the result of these adaptations and modifications. The second version,DIS-II, was a revision produced when researchers from Columbia University, who were also leading the construction of DSM-III, became coauthors.
Like clinical psychiatric interviews, the DIS distinguishes significant symptoms from the ordinary worries and concerns of daily life by setting requirements for clinical significance, and distinguishes psychiatric symptoms from symptoms caused by physical illness or the side effects of drugs or alcohol.
The DIS was unique at the time it was developed in that it could make diagnoses without requiring clinical personnel for either interviewing or scoring responses. Its questions can be asked and coded by lay interviewers according to clearly stated rules. The coded responses are entered directly into a computer where the diagnosis is made according to the explicit rules in the diagnostic systems served.
The DIS faithfully turns the DSM diagnostic criteria into questions. For example, in the diagnostic items from major depressive episode, DSM-IV requires that each symptom be present “during the same 2 week period.” To turn “markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day” into a question that a respondent may answer, the following item is used in the DIS: “Have you ever had a period of at least two weeks when you lost interest in most things or got no pleasure from things which would usually have made you happy?”
Since its first use in the ECA study, the DIS has been used across a very wide range of projects and was adapted by the World Health Organization to create the Composite International Diagnostic Interview (CIDI). Newer versions of the DIS have been produced to take into account revisions to the APA's diagnostic manual (i.e., DSM-III-R and DSM-IV).
The most recent edition of the DIS interview is Version IV. This version preserves many of the original features of the DIS, but it also adds new features.
The DIS-IV was developed to account for changes in the DSM from the DSM-III-R to the DSM-IV. Like earlier versions of the DIS, DIS-IV has the following assets:
- It is economical to use because it does not require clinically experienced examiners to administer the interview or to make diagnoses.
- It offers a lifetime history of symptoms. In addition, it ascertains when symptoms of a disorder first appeared and were most recently experienced and asks whether a doctor was ever consulted about the symptoms.
- With the exception of a few open-ended questions, answers to the interview are completely precoded for prompt diagnostic assessment.
- Reliability of questions and diagnosis is high because questions and probes are almost entirely specified, making it possible to train interviewers to behave in very similar ways.
- It is acceptable to both patients and members of the general population. Although it contains questions about sex, drinking, drug use, and police trouble, subjects rarely (less than 1%) refuse to answer any of these questions.
In contrast to an overall structure that is consistent with earlier versions, the DIS-IV has been revised to implement many ideas that emerged in the course of field experience.
This experience was both in the ECA study and in a large number of studies in many settings, cultures, and languages. The design of the revision has also profited from experience with the field trials and studies using the CIDI, which was originally based on the DIS and uses the same strategies, and from the development and implementation of the CIDI Substance Abuse Module. It has profited from the Alcohol Use Disorder and Associated Disabilities Interview Schedule (Hasin, Carpenter, McCloud, Smith, & Grant, 1997) and from work on the Diagnostic Interview Schedule for Children (DISC). Most of all, it has profited from the advice and criticisms of the DIS’s many users.
All questions have been reconsidered in terms of how closely they served DSM-IV criteria, and refashioned where necessary to improve understandability and translatability for use in other countries and in culturally diverse subpopulations of the U.S. There have also been changes in design, which are described next.1. Current Syndrome. The DIS-IV ascertains whether each disorder has been present in the last 12 months. It had been noted that data collected with previous versions of the DIS showed a large proportion of those who ever met criteria for a disorder as current cases. At least in part, that was because a disorder was counted as current if any of its symptoms had been present in the interval specified as representing “current” – 1 month, 6 months, or 1 year. DIS-IV still records how recently any symptom has been present but also determines whether a complete syndrome was present at any time in the last year.
2. Expanded Diagnostic Coverage. In addition to diagnoses available in previous versions, DIS-IV now makes additional diagnoses that typically arise in childhood. The addition of these childhood disorders was prompted by the observation in the ECA and other studies using the DIS that many cases of adult disorders were reported to have begun in childhood. These new modules include attention deficit hyperactivity disorder, oppositional defiant disorder, and separation anxiety disorder.
The diagnosis of dementia in both previous and current versions of the DIS is made only as a current disorder. Previously, it was based solely on failing the Mini-Mental State Exam (MMSE). Follow-up studies of epidemiological samples of the elderly who were initially negative on the MMSE have shown that making even a few MMSE errors predicts deterioration in clinical status over the next few years for many subjects. To improve the interview's sensitivity to mild dementia, DIS-IV adds items that operationalize the Blessed assessment of dementia, including some tasks difficult enough so that completing them without error serves as a reliable indicator that subclinical dementia is absent.3. Reducing False Negatives in Panic Disorder. The DIS has not skipped subjects out of sections assessing major depressive episodes and manic episodes even when they initially denied the requisite mood symptoms. Instead, the remaining symptoms of a typical depressive or manic episode were asked about, and only if they also lacked such episodes were they skipped out of the section. If they qualified for the symptoms of an episode other than mood, they were given a second opportunity to report that the requisite mood was present during the episode.
The new DIS added a similar design for panic disorder in response to clinicians’ observations that in anxiety clinics some persons currently in treatment for panic disorder score negative for the diagnosis. The reason seemed to be that over time, as these patients learned that their symptoms did not indicate a heart attack or imminent death, the somatic symptoms became more salient than the fears and worries with which they were initially associated. Thus, patients might deny the initial question dealing with fear during panic attacks and subsequently be skipped out prematurely.
The solution is to ask persons who deny having had “an attack of feeling very frightened, anxious or uneasy or as though something terrible was about to happen” whether they have ever had an attack in which they suddenly had any of the four most common somatic symptoms: shortness of breath, palpitations, dizziness, or chest pain. If they have had any of these, they are asked about other somatic symptoms and other criteria, making it possible for them to qualify for a panic attack without admitting fear or anxiety.4. Course of Specific Disorders. The DIS-IV ascertains whether there has been a year or longer free of disorder between the first and most recent symptoms and the ages during which the disorder was absent. This dating of periods of remission may be particularly useful in describing whether multiple disorders are sequential or concurrent. Previously the DIS could only indicate that two illnesses occurred at some time in the life of the same person, but now it is possible to show that two conditions actually overlap.
When a respondent reports symptoms of more than one disorder, knowing the order in which they first appeared allows considering the earlier disorder as a possible risk factor for the later one. When respondents report two or more disorders beginning in the same age bracket (i.e., childhood, the teens, the twenties, or the thirties), they are now asked the order in which they first appeared. This allows distinguishing primary from secondary disorders and corrects inferences about the order of appearance based entirely on the age recalled as the onset age.5. Impairment. DSM-IV specifies impairment as a criterion for most disorders. Thus, the DIS-IV operationalizes impairment with respect to each disorder by asking whether the symptoms created problems with family, friends, job, school, or in other situations, and whether the problems lasted a month or more and were severe. For disorders with any symptom in the current year, impairment in the current 12-month period is also ascertained.
6. Treatment Utilization and Perceived Treatment Need. The DIS-IV routinely enters into the computer information about medical consultation. In addition, at the end of each diagnostic section, respondents are asked whether they ever talked to any health professional about the symptoms of the disorder. Those respondents whose symptoms have occurred in the current 12-month period are also asked whether they have wanted to talk to a health professional about their problems in the last year and whether they actually did so. These items were added as a measure of treatment utilization and perceived need for treatment.
7. Additional Question Labels. Labels in the left margin beside each question link that question to the specific criterion or item in the DSM diagnostic manual the question was designed to serve. This allows people with or without computer skills to judge the face validity of the question by comparing it to the relevant text in the diagnostic manual.
DIS-IV adds labels for constructed variables used to assess age of onset, remission, treatment, and impairment. The label is the name of the variable constructed by the accompanying computer program and saved for later data analysis.8. Increased Precision in Dating the Most Recent Symptom. DIS-IV asks for the particular month in which the most recent symptom was last present.
9. Broadening and Dating Risk Factors. All epidemiological studies have found powerful associations between disorder and demographic characteristics. To assist in deciding whether standard demographic variables such as education, marital status, or parenthood might be risk factors for disorder, DIS-IV obtains the ages at which the respondent left school, married, divorced, was widowed, or had a first child. This allows determining whether these events preceded or followed the onset of a specific disorder. Other factors found to be risk factors in longitudinal research have been added, including number of siblings (family size), living apart from one or both biological parents before age 15, and parents’ or parent surrogates’ final educational level (to measure social status in childhood).
10. Expanded Health Behavior and Social Indicators. Several questions asked in earlier versions of the DIS because they were needed for diagnosis turned out to be of great interest in their own right as social indicators and indicators of health behaviors that have general public health significance. For example, researchers were able to say what proportion of the total population and various subpopulations had attempted suicide, had been arrested, had beaten spouses and children, had used weapons in a fight, had failed to provide child support, had multiple marriages, had cohabited, or had histories of promiscuity or drug injection that put them at risk for AIDS. The DIS-IV continues to obtain these indicators and adds a few others.
11. More Efficient Ordering of Diagnostic Modules. Tobacco dependence has been moved from the beginning of the interview to a location adjacent to other substance use questions. This allows a smooth flow from questions about the history of chronic physical illness to the symptoms of somatization disorder and pain disorder. Specific phobia, the most common of the anxiety disorders, is now the first anxiety disorder introduced. This is followed by panic attack, which has been placed just before agoraphobia, the disorder with which it has the most intimate connection. Questions about sexual dysfunction needed for somatization disorder have been moved from the end of the interview into the somatization section.
12. A Single Diagnostic System. Previous versions of the DIS served multiple diagnostic systems. For instance, when the DIS was modified to serve DSM-III-R, questions serving DSM-III were preserved to provide continuity with the earlier version. To serve multiple systems, some compromises had been necessary with respect to how closely questions could match specific criteria.
Because DSM-IV, when it was published, was expected to be the standard for some time, it was felt that the DIS-IV should serve DSM-IV criteria faithfully. In the diagnostic sections, questions that were left from previous editions of the DSM and not needed for DSM-IV were dropped and questions that did not match DSM-IV criteria precisely were rewritten. While the DIS-IV could be used to approximate diagnoses for the earlier DSM systems and for the International Statistical Classification of Diseases and Related Health Problems (10th ed.; ICD-10), its focus is clearly on DSM-IV. Thus, the DIS-IV diagnostic computer programs are available only for DSM-IV.13. Reducing Interviewer Burden. The DIS requires that interviewers review a list of items that had been answered positively when respondents were asked to date the first and most recent symptoms of a disorder and to consider whether symptoms cluster to form an episode. In previous editions of the DIS, interviewers performed this review by flipping through symptom questions and reading underlined phrases to refer to positive symptoms. With the addition in DIS-IV of impairment and treatment questions requiring reminding the respondent of positive symptoms, this task became even more burdensome.
DIS-IV adds tally sheets for disorders with long symptom lists and circling of box titles for depressive and manic episodes. These provide phrases to use in recapping symptoms and relieve the interviewer of scanning all completed questions. Furthermore, the computer version of the DIS-IV automates the recapping of endorsed items.14. Assessing Criteria. The ability to say whether specific criteria are present is a valuable addition to ascertaining the presence or absence of a diagnosis. It allows explaining the utilization of services by persons who do not meet full diagnostic criteria but do have some distressing symptoms. It is also useful in longitudinal studies because it allows distinguishing incident cases created by acquisition of one or two additional symptoms by persons already well on their way toward resembling true cases from de novo incident cases. The latter are more likely to be explained by interim life events or physical illness.
DIS-IV does not sacrifice this important feature of earlier versions. However, it is often the case that the presence of any one of several responses suffices to show an item or criterion to be present. DIS-IV now saves time by asking questions by which the criteria and items listed in DSM-IV can be manifested only until a positive response assures that the item or criterion is positive. Once an item or criterion is known to be positive, the other ways in which it might also be positive are not assessed. This change saves time while still assessing the presence of every criterion and item specified in DSM-IV.15. Vignettes. To quickly get at the core of certain diagnostic sections, DIS-IV uses vignettes in which the most common symptoms of a depressive and manic episode are described. Respondents are asked whether they had such an episode, and if they had more than one, to pick the episode with the largest number of the symptoms mentioned. The respondent is then asked about the duration and frequency of each symptom during the selected episode.
A vignette is similarly used to describe the common symptoms of post-traumatic stress disorder (PTSD) when there have been multiple traumas, so that the respondent can choose the traumatic event to which he or she reacted with the greatest number of symptoms (i.e., the event most likely to meet diagnostic criteria). DIS-III-R had assessed respondents’ reactions to up to three traumatic events to guard against missing the one that would qualify as causing PTSD. The vignette method achieves that goal but requires assessing reactions to only a single traumatic event.
Vignettes are also used to describe the common symptoms of attention deficit hyperactivity, oppositional defiant disorder, and separation anxiety disorders of childhood. The reason for doing so is that many of the symptoms of these disorders (such as talking a lot, fidgeting, objecting to one's parents going out, losing one's temper) are nearly universally present in children. It is only when the symptoms are especially numerous, severe, and frequent that they have clinical significance. Adults find it easy to recognize their childhood selves in these vignettes. When they do, the DIS-IV asks about each specific symptom. Those who say they did not fit the picture painted by the vignette are not questioned further.16. Optional Shortening Strategies. The modular structure of the DIS has always allowed dropping diagnoses in which a study has no interest. Each diagnostic section is independent, except where one diagnosis preempts another. DIS-IV offers a second option for shortening the interview: the use of a Screener version.
Psychometric properties of the DIS and related instruments have been studied extensively – including test-retest reliability studies, test-comparison studies, longitudinal studies and factor analytic studies (e.g., Hasin & Grant, 1987a, 1987b; Helzer et al., 1985; Hesselbrock, Stabenau, Hesselbrock, Mirkin, & Meyer, 1982; Robins, Helzer, Croughan, & Ratcliff, 1981; Rogler, Malgady, & Tryon, 1992; Semler et al., 1987; Vandiver & Sher, 1991; Wittchen et al., 1989).
The current version of the DIS was tested for reliability and validity in a study among substance abusers (Dascalu, Compton, Horton, & Cottler, 2001; Horton, Compton, & Cottler, 1998). The sample for this study was recruited from current and previous patients of substance abuse and psychiatric treatment sites to provide a broad range of diagnoses with varying severity. Trained nonclinician interviewers administered the DIS-IV in a blinded manner at test and retest, and reliability of lifetime disorders was measured by the kappa statistic (Bishop, Fienberg, & Holland, 1975; Cohen, 1960) among the 165 subjects. Results are shown in Tables 1 and 2 and demonstrate that substance abuse and dependence disorders had fair to excellent reliability (kappa .53 to .86); suicidal ideation and attempts had excellent reliability (kappa .76 and .80, respectively); and depression, mania, PTSD, panic disorder, phobic disorder, obsessive- compulsive disorder, antisocial personality, conduct disorder, and oppositional defiant disorder had fair to good reliability (kappa .40 to .67). Disorders with poor reliability were generalized anxiety disorder (kappa .33), attention deficit disorder (kappa .33), and specific phobia (kappa .25). For attention deficit and generalized anxiety, the symptoms had a higher reliability than the full disorder. This indicates that the symptom clusters have adequate reliability but the age of onset and impairment criteria are less reliable. These results are consistent with the literature on reliability of psychiatric disorders among drug abusers, and based on these results, we conclude that DIS-IV psychiatric disorders, except for specific phobia, have adequate reliability among substance users. Because most psychiatric disorders are less reliable among substance abusers than among nonsubstance abusers (Bryant, Rounsaville, Spitzer, & Williams, 1992), these tests of reliability may show the lower limit of reliability compared to non-substance-using populations.TABLE 1 Test-Retest Agreement on DSM-IV Substance Abuse and Dependence Diagnoses from the Reliability of the DIS-IV among Drug Users Study
Diagnosis
Kappa (95% CI)
Alcohol
Dependence
.67 (.54-.79)
Abuse*
.74 (.60-.87)
Amphetamine
Dependence
.67 (.48-.85)
Abuse*
.77 (.62-.93)
Cannabis
Dependence
.60 (.45-.85)
Abuse*
.60 (.46-.93)
Cocaine
Dependence
.53 (.35-.70)
Abuse*
.56 (.39-.73)
Hallucinogen
Dependence
.59 (.33-.84)
Abuse*
.61 (.39-.84)
Opiate
Dependence
.69 (.50-.89)
Abuse*
.53 (.31-.75)
Phencyclidine (PCP)
Dependence
.69 (.42-.96)
Abuse*
.86 (.68-1.0)
Sedative
Dependence
.59 (.36-.82)
Abuse*
.50 (.26-.74)
* Abuse calculated without regard to whether dependence was present. From Horton et al., 1998.
In a study of co-occurring psychiatric illnesses among substance abusers being admitted to treatment (Compton, 2001; Compton & Horton, 2001), the computerized version of the DIS took approximately 75 minutes (Compton, personal communication, 2001).
Validity of the DIS has been tested in a subsample from the same study by comparing diagnoses obtained with the DIS to those obtained using the WHO Schedules for Clinical Assessment in Neuropsychiatry (SCAN; Wing et al., 1990). Of the 100 subjects in this diagnostic concordance sample, 46 were from the St. Louis public drug treatment Central Intake unit; the remainder were patients previously treated in inpatient drug and psychiatric programs. Overall comparison of DIS and SCAN indicated fair to good agreement for substance use disorders (kappa .45 to .71). For co-occurring major depression and social phobia fair to good agreement was found (kappa .41 to .55). For schizophrenia, agreement was marginal (kappa .39). For panic disorder, agoraphobia, and specific phobia, agreement was poor but statistically significantly greater than chance (p < .05). Agreement between the SCAN and DIS diagnoses is nearly as good as the agreement between the SCAN and clinical diagnoses (p < .05) determined by the SCAN interviewers themselves. This indicates acceptable agreement between clinical and nonclinical interviewing techniques. These results are consistent with other comparisons of clinician and nonclinician diagnostic assessments (e.g., Hasin & Grant, 1987a, 1987b; Helzer et al., 1985).TABLE 2 Reliability of Selected DSM-IV Psychiatric Diagnoses and Symptoms among Substance Users from the Reliability of the DIS-IV among Drug Users Study
Diagnosis
Kappa (95% CI)
Major Depressive Episode
.67 (.55-.80)
Suicidal ideation
.76 (.66-.86)
Suicide attempts
.80 (.70-.90)
Manic Episode
.49 (.29-.68)
Elevated mood
.40 (.22-.59)
3 + positive manic symptoms
.45 (.26-.65)
Schizophrenia
.48 (.35-.61)
Hallucinations
.44 (.26-.62)
Delusions
.61 (.46-.75)
Generalized Anxiety
.35 (.14-.56)
Difficulty controlling worry
.43 (.24-.61)
Excessive worry
.41 (.22-.60)
Panic Disorder
.52 (.27-.77)
Panic attacks
.54 (.40-.68)
Post-Traumatic Stress Disorder
.46 (.29-.62)
Exposure to trauma
.61 (.33-.89)
Any Phobia
.42 (.24-.59)
Agoraphobia
.41 (.14-.68)
Social phobia
.56 (.35-.77)
Specific phobia
.25 (.02-.47)
Antisocial Personality Disorder
.49 (.27-.71)
Adult antisocial symptoms
.44 (.28-.61)
Conduct Disorder*
.51 (.33-.68)
Oppositional Defiant Disorder**
.60 (.47-.73)
Attention Deficit Hyperactivity Disorder
.33 (.11-.55)
Attention deficit symptoms
.63 (.47-.79)
Attention deficit impairment
.56 (.38-.75)
Attention deficit before age 7
.32 (.07-.56)
Hyperactivity symptoms
.45 (.27-.63)
Hyperactivity impairment
.42 (.20-.63)
Hyperactivity before age 7
.25 (.03-.46)
* Calculated without exclusion for antisocial personality.
** Calculated without exclusion for conduct disorder.
From Horton et al., 1998.
The strengths and weaknesses of the diagnostic manual are reflected in the instrument. Thus validity of the diagnoses derived from the DIS is generally limited to the validity of the DSM constructs themselves. If future research shows that additional symptoms are relevant for particular conditions, the DIS may not be able to accurately reflect these symptoms. On the other hand, the DIS routinely assesses the full range of DSM criteria for each endorsed diagnosis (i.e., no early skipouts). Therefore, new constellations of symptom profiles can be generated with DIS data. Such work may allow the DIS to be relatively robust with regard to changes in diagnostic systems over time.
The DIS has not been designed to take the place of clinical diagnosis, which requires a degree of clinical judgment not possible with nonclinician interviewers. Therefore, results from the DIS should be considered approximations of clinical diagnoses, and medical decisions based on DIS results require clinical confirmation. On the other hand, for clinical settings where full evaluations are not feasible, the DIS can be used to screen persons for additional psychiatric conditions not routinely evaluated. Positive cases should be referred for evaluation and possible intervention.
The DIS has been used in many different cultural settings. Versions of the DIS have been translated into over a dozen languages and have been used in large-scale epidemiological projects across the globe. Examples of translation and use of the DIS in disparate settings are studies in Taiwan, Korea, and Puerto Rico (Canino et al., 1987; Hwu, Yeh, & Chang, 1989; Lee et al., 1990a, 1990b). The instrument has also been adapted for use in American Indian populations and has been applied in several specific cross-cultural studies (e.g., Compton et al., 1991; Helzer & Canino, 1992; Hwu & Compton, 1994).
First and foremost, because the DIS is closely linked to the DSM system of diagnosis, applying the DIS in disparate cultures depends on the applicability of the DSM in those cultures. In most international settings, the DSM has gained widespread acceptance as the standard diagnostic system. Specific examples of psychopathology may vary from setting to setting, but the overall diagnostic groupings are well established and consistent (Helzer & Canino, 1992; Mezzich, Fabrega, Mezzich, & Coffman, 1985).
Translation and adaptation of the DIS into different languages requires extensive work to assure the conceptual equivalence of the symptom questions. Such conceptual equivalence may be even more important than literal equivalence. Even before any formal psychometric testing is undertaken, both bilingual and monolingual experts and respondents must review the translated instrument to make sure of its applicability.
As in all research involving exploration of health experiences, some respondents may experience emotional discomfort when answering certain questions in the DIS. Training of interviewers includes consideration of such difficult interviewing situations along with ways to address these problems. If any particular questions make people uncomfortable, the question should be skipped. Despite this warning, refusal to answer particular questions and interview breakoff because of discomfort is quite rare (< 1%).
A specific concern in the depression section of the DIS is how to handle respondents who express current suicidal ideation. We suggest that each study develop its own data and safety monitoring plan for handling such situations based on available local resources. In general, for cases in which there is a clear potential for immediate danger, the interviewer is instructed to respond with an active intervention (i.e., have mental health authorities assess the respondent).
Future enhancements of the C DIS-IV will be to develop a web-based interactive version of the interview. The advantage of such an administration method is that data from remote sites can be stored in one central location and updates to the interview can be done for all users.REFERENCES
American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: Author.
American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., rev.). Washington DC: Author.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington DC: Author.
Bishop, Y.M., Fienberg, S., & Holland, P. (1975). Discrete multivariate analysis. Cambridge: MIT Press.
Bryant, K.J., Rounsaville, B., Spitzer, R.L., & Williams, J.B. (1992). Reliability of dual diagnosis-substance dependence and psychiatric disorders. Journal of Nervous and Mental Disease, 180, 251-257.
Canino, G.J., Bird, H.R., Shrout, P.E., Rubio-Stipec, M., Bravo, M., Martinez, R., et al. (1987). The prevalence of specific psychiatric disorders in Puerto Rico. Archives of General Psychiatry, 44, 727-735.
Canino, G.J., & Bravo, M. (1994). The adaptation and testing of diagnostic and outcome measures for cross-cultural research. International Review of Psychiatry, 6, 281-286.
Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurements, 20, 37-46.
Compton, W.M. (2001, December). Improving treatment services for substance abusers with co-occurring depression. Paper presented at the annual meeting of the American Academy of Addiction Psychiatry, Amelia Island, FL.
Compton, W.M., Helzer, J.E., Hwu, H.G., Yeh, E.K., McEvoy, L., Tipp, J.E., et al. (1991). New methods in cross-cultural psychiatry: Comparing rates of psychiatric illness in Taiwan to rates in the United States. American Journal of Psychiatry, 148, 1697- 1704.
Compton, W.M., & Horton, J.C. (2001, March). Case management to improve treatment engagement and outcomes for substance abusers with comorbid depression. Paper presented at the annual meeting of the American Psychopathological Association, New York, NY.
Dascalu, M., Compton, W.M., Horton, J.C., & Cottler, L.B. (2001). Validity of DIS-IV in diagnosing depression and other psychiatric disorders among substance users. Drug and Alcohol Dependence, 63, 37.
Hasin, D.S., Carpenter, K.M., McCloud, S., Smith, M., & Grant, B.F. (1997). The alcohol use disorder and associated disabilities interview schedule (AUDADIS): Reliability of alcohol and drug modules in a clinical sample. Drug and Alcohol Dependence, 44, 133-141.
Hasin, D.S., & Grant, B.F. (1987a). Diagnosing depressive disorders in patients with alcohol and drug problems: A comparison of the SADS-L and DIS. Journal of Psychiatric Research, 21, 301-3 11.
Hasin, D.S., & Grant, B.F. (1987b). Psychiatric diagnosis of patients with substance abuse problems: A comparison of two procedures, the DIS and SADS-L. Alcoholism, drug abuse/dependence, anxiety disorders and antisocial personality disorder. Journal of Psychiatric Research, 21, 7-22.
Helzer, J.E., & Canino, G. (Eds.). (1992). Alcoholism in North America, Europe and Asia. New York: Oxford University Press.
Helzer, JE., Robins, L.N., McEvoy, L.T., Spitznagel, E.L., Stolzrnan, R.K., Farmer, A., et al. (1985). A comparison of clinical and diagnostic interview schedule diagnoses. Physician reexamination of lay-interview cases in the general population. Archives of General Psychiatry, 42, 657-666.
Hesselbrock, V., Stabenau, J., Hesselbrock, M., Mirkin, &Meyer, R. (1982). A comparison of two interview schedules: The Schedule for Affective Disorders and Schizophrenia-Lifetime and the National Institute for Mental Health Diagnostic Interview Schedule. Archives of General Psychiatry, 39, 674-677.
Horton, J., Compton, W.M., & Cottler, L.B. (1998). Assessing psychiatric disorders among drug users: Reliability of the revised DIS-IV. In L. Harris (Ed.), NIDA Research Monograph-Problems of Drug Dependence. Washington, DC: NIH Publication No. 99-4395.
Hwu, H-G., & Compton, W.M. (1994). Comparison of major epidemiological surveys using the Diagnostic Interview Schedule. International Review of Psychiatry, 6, 309-327.
Hwu, H-G., Yeh, E.K., & Chang, L.Y. (1989). Prevalence of psychiatric disorders in Taiwan defined by the Chinese diagnostic interview schedule. Acta Psychiatrica Scandinavica, 79, 136-174.
Lee, C.K., Kwak, Y.S., Yamamoto, J., Rhee, H., Kim, Y.S., Han, J.H., et al. (1990a). Psychiatric epidemiology in Korea. Part I: Gender and age differences in Seoul. Journal of Nervous and Mental Disease, 178, 242-246.
Lee, C.K., Kwak, Y.S., Yamamoto, J., Rhee, H., Kim, Y.S., Han, J.H., et al. (1990b). Psychiatric epidemiology in Korea Part II: Urban and rural differences. Journal of Nervous and Mental Disease, 178, 247-252.
Mezzich, J.E., Fabrega, H., Mezzich, A.D., & Coffman, G.A. (1985). International experience with the DSM-III. Journal of Nervous and Mental Disease, 173, 738-741.
Robins, L.N., Helzer, J.E., Croughan, J., & Ratcliff, K.S. (1981). National Institute of Mental Health Diagnostic Interview Schedule: Its history, characteristics, and validity. Archives of General Psychiatry, 38, 381-389.
Robins, L.N., & Regier, D.A. (Eds.). (1991). Psychiatric disorders in America: The epidemiologic catchment area study. New York: Free Press, 1991.
Rogler, L.H., Malgady, R.G., & Tryon, W.W. (1992). Evaluation of mental health issues of memory in the Diagnostic Interview Schedule. Journal of Nervous and Mental Disease, 180, 215- 222 (discussion, pp. 223-226).
Semler, G., Wittchen, H.U., Joschke, K., Zaudig, M., von Geiso, T., Kaiser, S., et al. (1987). Test-retest reliability of a standardized psychiatric interview (DIS-CIDI). European Archives of Psychiatry and Neurological Sciences, 236, 214-222. Vandiver, T., & Sher, K.J. (1991). Temporal stability of the Diagnostic Interview Schedule. Psychological Assessment, 3, 277-281.
Wing, J.K., Babor, T., Brugha, T., et al. (1990). SCAN-Schedules for Clinical Assessment in Neuropsychiatry. Archives of General Psychiatry, 47, 589-593.
Wittchen, H.U., Burke, J.D., Semler, G., Pfister, H., Von Cranach, M., & Zaudig, M. (1989). Recall and dating of psychiatric symptoms: Test-retest reliability of time-related symptom questions in a standardized psychiatric interview. Archives of General Psychiatry, 46, 437-443.
Washington University School of
Medicine
40 N. Kingshighway, Suite 4
St. Louis, MO 63108
(314) 286-2252, FAX: (314) 286-2265