Annotated Guide to the February 2001 SCID Revision

The following is a guide to the February 2001 SCID revision. Although it may look like a major revision (with a new typeface and even a new name "SCID-I: Structured Clinical Interview for DSM-IV-TR Axis I Disorders,") this revision is only slightly more extensive than previous SCID revisions. We have changed the name of the SCID to reflect the publication of the DSM-IV Text Revision (DSM-IV-TR) in July 2000. The DSM-IV-TR was an empirically-based revision and update of the DSM-IV text sections so, by and large, there were no changes in the diagnostic criteria. However, two changes in the DSM-IV-TR do call for changes in the SCID, both in Module E:

1. In DSM-IV-TR, the definition of the subtype "with agonist therapy" has been clarified so that the only drug class for which the term "agonist therapy" is considered appropriate is opioids (e.g., treatment with methadone or LAAM). Thus, page E.21 of the SCID has been changed so that the only drug classes in the SCID that now allow this subtype are opioids, polysubstance (since one of the substances may be opioids) and other (to allow for other substances in the future that may be treated with agonist therapy).
2. In DSM-IV-TR, the definition of Polysubstance Dependence has also been clarified. It became clear during the DSM-IV text revision process that there are two different interpretations of the concept of Polysubstance Dependence. In one interpretation, an individual is given a diagnosis of Polysubstance Dependence if there is a period of six months of indiscriminant drug use, in which there is no favorite drug that qualifies for dependence on its own. A second interpretation sees Polysubstance Dependence as analogous to "mixed personality disorder" in that if an individual is taking 3 or more different drugs and the criteria are not met for dependence on any one drug, a diagnosis of Polysubstance Dependence can be given if the sum of the positive items for all the different drugs together reaches the threshold of 3 or more items total. The SCID originally was designed to allow a diagnosis only using the first interpretation—it required the interviewer to identify a period of indiscriminant drug use, select the "POLY" column, and then check for dependence on the group of drugs taken during that period. The SCID has now been modified (page E. 18) to allow the interviewer to consider a diagnosis of Polysubstance Dependence AFTER the seven criteria for dependence are checked. If criteria are not met for any one column but can be met for 3 columns (during the same period of time), then a diagnosis of Polysubstance Dependence can be given.

Please note that, as with previous SCID revisions, we have maintained the same item field numbers (i.e., the unique number printed using a small typeface on the right side of each rating) so that data analysis programs that use these data field identifiers will continue to work. New data fields were inserted between existing data fields by using additional letters. For example, the addition of a Chronology section for Schizophrenia required that we add 8 new data fields between items C13 and C14; thus, C13a, C13b,..., C13h were inserted. Similarly, when data fields were deleted, the prior numbering was retained (i.e., there are gaps in the numbering where fields were omitted).

Here is a page-by-page summary of the major changes:

A.2, A.13 (Current and Past MDE, Mood Episodes Module), J.10 (Lifetime MDE, Optional Module): The order of the questions for Major Depressive Episode criterion A(3) has been changed so that the question about appetite change comes before the question about weight loss or weight gain.

The SCID originally put the weight loss/weight gain question first, mirroring the order of the items within the criterion. However, since the goal of the questions for criteria A(3)-A(9) is to determine whether at least 5 out of the 9 symptoms of a Major Depressive Episode occurred during the same two-week period, the weight loss/weight gain item never really made much sense—although weight loss or gain is extremely common during a Major Depressive Episode, most typically the weight change occurs over a period of weeks to months; a significant change in weight (on the order of 5% of body weight, as mentioned in the criterion) rarely occurs during the same two-week period as the other symptoms. Thus, since it makes more sense to focus on the appetite change rather than the weight change, we ask about change in appetite first. Weight loss or gain is now inquired about only if there is no appetite change.

A.8-A.9 (Melancholic Features, Mood Episodes Module): Each item in Melancholic Features now has a corresponding SCID question—no longer do we just refer the interviewer back to the corresponding items in the Major Depressive Episode. In addition, a skip instruction to "MANIC EPISODE, A.18" is added after the diagnosis of Melancholic Features is made.

Melancholic Features share several of the same symptoms with Major Depressive Episode. For this reason, previous versions of the SCID assumed that the interviewer already had the information necessary to make most of the ratings on pages A.8-A.9. However, since the time frame for Melancholic Features (i.e., the period of greatest severity during the current episode) may differ from the time frame for the current Major Depressive Episode (i.e., which was chosen to be the worst two weeks in the past month, or the most recent two weeks if the severity has been the same during the entire month), it is not uncommon for the interviewer to have to ask about the Melancholic Features for this different time frame. Furthermore, although they appear to be identical, the Melancholic Features criteria are actually more severe (i.e., "marked psychomotor agitation or retardation" vs. "psychomotor agitation or retardation," "significant anorexia..." vs. "...decrease...in appetite"). For these reasons, we have inserted questions for each of the Melancholic Features items.

Although the skip instruction on the top of page A.10 instructs the interviewer to skip to A.18 if a diagnosis of Melancholic Features has been made, some interviewers missed this instruction. Thus, we have moved this skip instruction to right below the box for Melancholic Features on page A.9.

A.10 (Atypical Features, Mood Episodes Module): As above, each item in Atypical Features now has a corresponding SCID question.

The time frame for Atypical Features is always the most recent two weeks. Although the time frame for the period of inquiry for the current Major Depressive Episode is often the same, it may have been the first two weeks of the month, if that turns out to be the worst period in the month. Thus, for those items covering information included in the Major Depressive Episode criteria set (i.e., A, B(1), and B(2)), it may be necessary to recheck the items for the most recent two weeks.

C.1-C.25 (Psychotic Disorders Module): Page numbers within this module have all changed to reflect the addition of a separate chronology section for Schizophrenia on pages C.8-C.10.

C.8-C.10 (Schizophrenia, Psychotic Disorders Module): The Chronology section has been split into two parts. Schizophrenia now has its own Chronology section on pages C.8-C.10. The other psychotic disorders covered in this module have a separate Chronology section, starting on page C.24.

The previous SCID had a single Chronology section that was intended to apply to all of the psychotic disorders, including Schizophrenia. However, the definition of "current" differs for Schizophrenia and the other psychotic disorders. For psychotic disorders, in general, the disorder is considered to be "current" if the subject has been psychotic during the past month (or alternatively for Schizoaffective Disorder, the full criteria for a mood episode have been met). For Schizophrenia, the disorder is also considered current if the subject is in the residual phase (i.e., negative or subthreshold positive symptoms). Therefore a separate three page Chronology section was added for Schizophrenia, requiring a complete page renumbering for this section.

D.1 (Bipolar I Disorder, Mood Disorders Module): The skip instruction at the top now indicates that Module D can be skipped if all mood symptoms are accounted for by a diagnosis of Schizoaffective Disorder made in Module C. Furthermore, the "mixed" subtype in the bottom box now includes the DSM-IV definition of a mixed episode for clarification.

This skip instruction was included in earlier versions of the SCID but had been inadvertently dropped. Module D is for the diagnosis of mood symptoms other than those covered under the diagnosis of Schizoaffective Disorder. Although the definition of "mixed" is included in a note in the middle column below the first criterion, the definition has been added alongside the rating of the current episode type since some interviewers missed it otherwise.

D.5 (Other Bipolar Disorders, Mood Disorders Module), Scoresheet page ii: A new subtype (i.e., type #4) for "Other Bipolar Disorder" has been added to allow the interviewer to specify that Bipolar Disorder NOS has been diagnosed due to the presence of sub-threshold Manic Episodes that do not meet criteria for Hypomanic Episodes, i.e., episodes with a duration of less than one week (thus not meeting the criteria for a Manic Episode) that cause significant impairment in social or occupational functioning (thus not meeting criteria for a Hypomanic Episode).

A little-publicized but well-known "hole" exists in the DSM-IV differential diagnosis of Manic Episode and Hypomanic Episode. The criteria for Manic Episode require a minimum 1-week duration (or less if hospitalization is required), whereas the criteria for Hypomanic Episode require that the mood symptoms not be severe enough to significantly impair functioning. Thus, a severe episode lasting 5 days would not meet the criteria for either Manic Episode or Hypomanic Episode and would require a diagnosis of Bipolar Disorder NOS. This problem was partially patched up during the April 1997 SCID revision—in that version, we fixed the skip instructions for criterion E for Hypomanic Episode on pages A.26 and A.38 (i.e., "if severe enough to require hospitalization, consider returning to A.18/A.28 and recoding as current manic episode. Otherwise, continue with A.28/A.38 and code Other Bipolar Disorder on D.5") but neglected to provide a subtype for this on page D.5. This revision includes a type #4 on page D.5 to allow the interviewer to indicate this reason for diagnosing "Other Bipolar Disorder."

E.2, E.4-E.5, E.12-E.16, E.23-E24 (Substance Use Disorders Module): The time period for the Abuse and Dependence questions has been changed to focus on the period in which the subject reported using the most, or the period in which the substance use caused the most problems (which is known from the screening sections on pages E.1 and E.10-E.11). The wording of the questions has been changed from lifetime ("have you ever") to focus on the identified time period ("did you").

In previous versions of the SCID, the Alcohol Abuse and Dependence questions (as well as the Substance Abuse and Dependence questions) were worded so that they assessed whether each symptom has ever occurred during the subjectâs lifetime (e.g., "Have you ever missed work·"). This method was problematic because it made it difficult to determine whether the symptoms clustered during the same 12-month period, as required for the diagnosis of Dependence, or whether the symptoms occurred within a 12-month period, as required for a diagnosis of Abuse. Each section has now been changed to clarify that the questions refer to the time period in which the subject reports that he or she was using the substance the most, or to that period in which the substance use caused the most problems. Furthermore, the wording of the questions has been adjusted so that they clearly apply to this period of time (e.g., "have you ever missed work·" is changed to "did you miss work·").

E.9 (Alcohol Dependence, Substance Use Disorders Module): The "With Agonist Therapy" option has been eliminated for Alcohol Dependence (i.e., benzodiazepines are not considered agonist therapy for Alcohol Dependence).

E.16 (Substance Dependence, Substance Use Disorders Module): In previous versions of the SCID, a note appeared here stating that the criterion for substance withdrawal may not apply to Cannabis and Hallucinogens/PCP. This note has been deleted based on information added to the DSM-IV-TR text indicating the possibility of withdrawal symptoms for these three classes of drug, especially for cannabis.

E.18 (Substance Dependence, Substance Use Disorders Module): Two data fields were inadvertently assigned to each rating on this page—one set of data field codes have been eliminated (but a gap is maintained to preserve the match up of data fields between SCID versions). Furthermore, below the ratings of "1," an instruction has been inserted to permit the diagnosis of Polysubstance Dependence "if the number of different dependence symptoms (taken from at least three different drug classes) and occurring during the same 12-month period adds up to at least three." This is in accordance with the clarification made in the DSM-IV text revision (see discussion above).

E.19-E.21 (Substance Dependence, Substance Use Disorders Module): The Chronology section for Substance Dependence inadvertently left out ratings for the age at onset. Furthermore, the layout of the ratings was confusing because the ratings for "With Physiological Dependence" and "Without Physiological Dependence" were embedded in the middle of the rating for criteria being met in the past month. The entire Chronology section has been reorganized as follows: The section starts with Age At Onset for each drug class in which the lifetime criteria have been met for Dependence. Next are the ratings indicating if criteria have been met in the past month. Following this, the Chronology section is divided based on the ratings made for the current month as follows. For those drug classes coded "3" (i.e., criteria met in past month), ratings are made on page E.20 indicating current type ("With Physiological Dependence" or "Without Physiological Dependence") and current severity (Mild, Moderate, Severe). For those drug classes coded "1" (i.e., lifetime criteria met but not in past month), ratings are made on page E.21 indicating type of remission (Early Full, Early Partial, Sustained Full, Sustained Partial, On Agonist Therapy, In A Controlled Environment).

E.21 (Substance Dependence, Substance Use Disorders Module): The subtype "On Agonist Therapy" has been eliminated from all drug classes except Opioids, Polysubstance, and Other (see discussion above).

E.25 (Substance Abuse, Substance Use Disorders Module): The Chronology section for Substance Abuse inadvertently left out ratings for the age at onset—these have been added for each drug class.

Alternative to Pages E.10-E.25 for Substance Dependence/Substance Abuse: The non-alcohol substance dependence and abuse section is perhaps the most complex part of the SCID because it allows for the simultaneous rating of dependence and abuse for 10 different classes of substances. While this level of information is important for studies focusing on substance use problems, many studies are only interested in determining whether dependence or abuse has been met for ANY substance. For this reason, we are including an "alternate version" of the Non-alcohol Substance Dependence and Abuse section. In this version, the interviewer decides (after the screening section) which single drug class and time period to focus on (presumably the period of heaviest use or most severe drug-related problems). In cases of multiple drug use, the interviewer first selects the drug most likely to meet criteria for dependence or abuse. If criteria are not met for the chosen drug, the interviewer then has the option of considering other drugs and other time periods until either criteria are met or else the interviewer is satisfied that there is no dependence or abuse for any drug used.