The World Mental Health Composite International Diagnostic Interview
The World Mental Health Survey Initiative
World Health Organization Health and Work Performance Questionaire
NCS-R Mania notes to all users:
The mania section was revised from the earlier CIDI with the collaboration of Kathleen Merikangas and Uli Wittchen in consultation with Jules Angst. A major change was that the diagnostic stem question for euphoria (SC24) was substantially expanded to paint a more complex portrait and to make it less pathological in tone than in the earlier CIDI. The irritability stem question was also decomposed (SC25-25a).

Note that routing into the section comes from SC36 (if the respondent did not endorse and MD stem questions) or D88 (if the respondent did endorse an MD stem question). Note, too that routing differs depending on whether R endorsed SC24 or, if not, SC25a.

M1. This question is asked only if R endorsed the SC24 stem question. As in the MD section and many other diagnostic sections, we followed the lead of Lee Robins in her revised DIS in asking a global symptom-clustering question before moving into the body of the section. Note that we ask about having ANY of these changes in order not to set too high a threshold at this entry question.

M2. There are a great many complicated skip-outs in the instrument. I can't comment on the logic of each one. To illustrate, though, this particular one is part of a logic that administers an irritable depression section to all respondents who endorsed SC25, who did not go into the depression section far enough to have a symptom assessment, and who did not meet criteria for mania or hypomania.

M3-3d. As in the MD section and many other sections in the instrument, we focus R into talking about one particular episode. We ask about age and duration merely to force R into thinking a bit about the episode before turning to the symptom questions. Responses to the age and duration questions are not used in any way.

M4. This is the first of the Criterion A symptom questions for the section. Note that M4 asks about irritability, which is assessed in a different way for Rs who denied euphoria and endorsed irritability. Then they are skipped to the remaining symptom questions in the M7 series.

M5. This is the entry point into the section for Rs who denied euphoria (SC24) and endorsed irritability (SC25a).

M6-6c. These are identical to the M3a-3d questions, but asked of the people with the irritability stem rather than those with euphoria stem.

M7-7o. These are the remaining Criterion A symptom questions for mania and hypomania.

M8. Respondents who do not endorse a sufficient number of Criterion A symptoms are skipped to the next section. In most sections we set this number below the number required in the ICD and DSM systems in order to make it possible to study sub-threshold cases.

M9-9b. Most diagnostic sections contain questions similar to these. Some level of impairment is required to continue with the section. However, the threshold is set rather low in order to make sure we are capturing all clinically significant cases. The reader can refer to the diagnostic algorithms, where he will see that the impairment threshold is higher for being defined as a case. We retain respondents with sub-threshold impairment, though, as part of the larger decision to retain sub-threshold cases, as noted in the comments to M8. The reader will see a number of instances both in this and other sections where there is a similar discrepancy between the severity levels required to remain in the section and the higher severity levels required for a diagnosis.

M10-10c. This is the organic exclusion series. A similar question series can be found in many sections, although the series is considerably more elaborate than this one in some other sections. All such series contain an open-ended question. It is necessary for an experienced psychiatrist or neurologist to review responses to these questions and to make clinical judgments as to whether the physical conditions constitute legitimate rule-outs for the diagnosis in question. A useful way to reduce the burden of this task is for the researcher to run the diagnostic algorithms twice, once setting the organic exclusion questions to yes and once to no, and comparing diagnoses. The cases that emerge as cases in the latter, but not the former, are those for whom the decision about organic exclusions is critical. Only these cases need to be reviewed at M10c. Recode M10a to 1 for purposes of generating diagnoses when the clinical rater determines that the organic exclusion should not be made. Our general approach in cases of doubt has been to exclude the cases, but a case could be made for developing a special code for uncertain cases and carrying out sensitivity analyses or for having a clinician re-contact such respondents to carry out a more definitive clinical interview.

M18-18b. This age-of-onset series uses the standard three-question structure that was first developed for use in the baseline NCS. Charlie Cannell from the University of Michigan suggested this wording approach. As the reader who has carefully reviewed these notes from the first section will be aware, Charlie's name has been mentioned frequently. Cannell was probably the most influential question wording methodologist in the survey research world during the 1960s and 1970s. He had an extraordinary gift for understanding the survey interview situation and the subtleties of context that influence survey responses. We profited enormously from his consultation for many years and especially during the development of the baseline NCS interview schedule. The logic of this question series is discussed in Knauper et al. (1999)[1].

M18c-18d. We added these two questions into both the MD and M sections in NCS-R, and slightly different questions at the same place in other sections, in order to examine the effects of triggering events on first onsets and to allow us to study differences in subsequent illness course depending on whether initial episodes were triggered or not. The same notion appears in somewhat different form in the specific phobia section, where we ask people with, say, animal phobias whether their phobia started based on a triggering event (e.g., attacked by a dog) or "out of the blue". There is some reason to think that the two types of animal phobia differ in important ways, although we are aware of no previous systematic epidemiological research that has investigated this possibility.

M19c. The NCS-R added in for the first time questions of this sort in each section that try to learn about 12-month persistence of disorders. Earlier DIS and CIDI surveys, although they ask about recency, can tell us nothing about whether there was only a 3-day episode or a full 365 days of disorder in the past year. This lack of information is a major loss for researchers interested in needs assessment for service planning purposes. The reader will see similar questions in other sections of the instrument.

M20-23. Similarly, each section of the revised CIDI now includes a series of questions about lifetime course. The goal here is to fill in some detail about course that goes beyond the standard DIS and CIDI questions about age of onset and age of recency.

M21.1. This question is an extension of the logic described in the comments on M18c-18d.

M27-29. As discussed in the earlier notes on Section D, this set of questions, consisting of the Sheehan Disabilty Scales and a question about days out of role in the past 365 days, appears in each diagnostic section of the instrument including the CC section.

M30-30k. This question series is the clinical severity series for the M section. As described in the notes to the D section, we attempted to include some measure of 12-month clinical severity in each diagnostic section. We appreciate the consultation of Mauricio Tohen in selecting the scale to use in this section. The reader should note that this series of questions is self-administered by respondents who can read and interviewer-administered by respondents who cannot read.

M33-48. This is the standard disorder-specific services question series that is included in each diagnostic section.

M33a. We attempt to establish age of first contact with the treatment system in M33a in order to carry out studies of delay in first seeking treatment after disorder onset. See Kessler et al.(1998)[2], Olfson et al. (1998)[3], Christiana et al. (2000)[4], and Kessler et al. (2001)[5] for examples of this line of investigation.

M45. This is, in some ways, more of a "marketing research" question than an epidemiological question in the sense that the patient's self-report might or might not be accurate. Nonetheless, it is important to learn how many patients perceive treatment to be helpful.

M45a. And it is also important to see how many years it takes between the initial onset of treatment and the time when the patient feels that treatment became helpful.

M45b-45c. In a similar way, it is important to see how many different professionals a patient typically has to see before he receives what he considers helpful treatment. Note that it is necessary to collect the same information from patients who say they never received helpful treatment in order to carry out such an analysis in a survival framework.

M48.1. This question was added to each diagnostic section based on the ideas of Randy Nesse about possible differences in symptom profiles within disorders depending on family history.

[1] Knäuper, B., Cannell, C.F., Schwarz, N., Bruce, M.L., & Kessler, R.C. (1999). Improving accuracy of major depression age-of-onset reports in the US National Comorbidity Survey. International Journal of Methods in Psychiatric Research, 8(1): p. 39-48.

[2] Kessler, R.C., Olfson M., & Berglund P.A. (1998). Patterns and predictors of treatment contact after first onset of psychiatric disorders. American Journal of Psychiatry, 1998. 155(1): p. 62-69.

[3] Olfson, M., Kessler R. C., Berglund P. A., & Lin E. (1998). Psychiatric Disorder Onset and First Treatment Contact in the United States and Ontario. American Journal of Psychiatry, 155: p. 1415-22.

[4] Christiana, J.M., Gilman, S.E., Guardino, M., Kessler, R.C., Mickelson, K., Morselli, P.L., & Olfson, M. (2000). Duration between onset and time of obtaining initial treatment among people with anxiety and mood disorders: An international survey of members of mental health patient advocate groups. Psychological Medicine, 30(3): p. 693-703.

[5] Kessler, R.C., Aguilar-Gaxiola, S., Berglund, P.A., Caraveo-Anduaga, J.J., DeWit, D.J., Greenfield, S.F., Kolody, B, Olfson, M., & Vega, W.A. (2001). Patterns and predictors of treatment seeking after onset of a substance use disorder. Archives of General Psychiatry, 58(11): p.1065-1071.