Overview of Instruments
Use of the Instruments by Other Investigators
Questions and Answers about the Instruments
Links to the Instruments
The following instruments are the ones used in the National Comorbidity Survey (NCS) family of surveys:
The Baseline NCS was a national survey of people in the age range 15-54 carried out in 1990-92. The Baseline NCS Interview Schedule is a fully structured interview that was administered face-to-face using paper and pencil interviewing (PAPI) by trained lay interviewers.
Baseline NCS Clinical Reappraisal interviews were carried out in small disorder-specific sub-samples of the NCS after completion of the Baseline NCS Interview Schedule. These clinical interviews were based on the lifetime version of the Structured Clinical Interview for DSM-III-R (SCID). Despite its name, the SCID is a semi-structured (not fully structured) clinical interview that requires clinical judgment to administer. The Disorder-Specific Baseline NCS Clinical Reappraisal Interview Schedules were administered over the telephone by trained clinical interviewers using PAPI.
The NCS Replication (NCS-R) Survey was a national survey of people in the age range 18+ carried out in 2001-02. The NCS-R Interview Schedule is a fully structured interview that was administered face-to-face by trained lay interviewers. The NCS-R aimed both to replicate some aspects of the Baseline NCS for purposes of trending and also to expand on the Baseline NCS in important ways. As a result, the NCS-R Interview Schedule includes some questions that are identical to questions in the Baseline NCS and others that are new to the NCS-R. The NCS-R was administered using computer-assisted personal interviewing (CAPI) on laptop computers. The CAPI program was written in the Blaise software language. The hard copy instrument shown here is laid out in the form of a PAPI instrument for ease of reading. This is the version of the instrument from which the Blaise program was written.
The NCS-2 Survey was a re-interview survey of people who participated in the Baseline NCS. NCS-2 was carried out a decade after the Baseline NCS, in 2001-02. The aim was to collect information about changes in mental disorders, substance use disorders, and their predictors and consequences over the ten years between the two surveys. The NCS-2 Interview Schedule was administered using CAPI with an individually customized preload file. This means that the computer program knew many things about the respondent's reports in the Baseline NCS and used this information to customize skip patterns. As with the NCS-R interview schedule file, the hard copy instrument shown here is the version of the instrument from which the Blaise program was written.
As in the Baseline NCS, Clinical Reappraisal Interviews were carried out in sub-samples of respondents who participated in the NCS-R and NCS-2 surveys. As in the Baseline NCS, these reinterviews were administered by telephone using PAPI. Unlike in the Baseline NCS, though, where small sub-samples were administered diagnosis-specific clinical reappraisal interviews, a single clinical interview schedule was used in the NCS-R and NCS-2 that included all diagnoses that were reappraised in these surveys. As in the Baseline NCS, the clinical reappraisal interviews were based on the SCID, but in this case the updated version of the SCID for DSM-IV. Another important difference is that both lifetime and 12-month prevalence were assessed in the NCS-R and NCS-2 clinical reappraisal interviews, while only lifetime prevalence was assessed in the Baseline NCS Clinical Reappraisal interviews. However, it was impossible to administer a comprehensive version of both the lifetime and 12-month SCID to the same respondents because of respondent burden. As a result, separate lifetime and 12-month NCS-R/NCS-2 Clinical Reappraisal Interview Schedules were developed and used in this aspect of the research.
The NCS-A survey is a nationally representative survey of adolescents in the age range 13-17 carried out in 2001-02. The interview schedule used in the NCS-A is similar in many ways to the adult NCS-R interview schedule, but there are also differences both in diagnoses and in several domains of questioning about risk factors and social consequences. As in the NCS-R, the NCS-A interviews were administered using CAPI. An additional difference between NCS-R and NCS-A is that parent self-administered questionnaires were included in NCS-A.
As in the other NCS surveys, Clinical Reappraisal Interviews were carried out in sub-samples of the NCS-A. These interviews were administered over the telephone using PAPI by trained clinical interviewers. Both the adolescent and one parent were interviewed. Reconciliation interviews were also carried out, as necessary, with the adolescents when parent reports were discrepant from adolescent reports. The results of adolescent (always interviewed first), subsequent parent clinical interviews, and reconciliation interviews with adolescents were recorded in the same PAPI interview schedule. The same clinical interviewer who carried out the adolescent interview also interviewed the parent of that adolescent and, when needed, carried out the reconciliation interview with the adolescent. Only 12-month disorders were assessed in the adolescent clinical reappraisal interviews. These interviews were based on the Kiddie Schedule for Affective Disorders and Schizophrenia (K-SADS).
Finally, we include here the PAPI version of the World Health Organization's World Mental Health (WMH) Survey instrument. WMH is an initiative of WHO that is administering a modified version of the NCS-R interview schedule to nationally representative samples in approximately 28 countries around the world. Some of these countries are using laptop computers to carry out their surveys, in which case they are using the NCS-R CAPI interview schedule (with modest changes, such as not asking about AFDC, asking about income in the local monetary units rather than in dollars, etc.). The majority, though, are unable to use laptops either for financial reasons or for logistical reasons (e.g., lack of access to a stable electrical supply). As a result, we developed a PAPI equivalent of the NCS-R interview schedule. This PAPI instrument uses less complex skip logic than the NCS-R instrument because the computer controls the skip sequences in the latter and these sequences can consequently be very complex. This need to simplify the skip logic required a number of simplifications of question wording as well. However, the vast majority of the questions in the WMH PAPI interview schedule are identical to those in the NCS-R interview schedule.
These instruments are posted for informational purposes only and should not be used by other investigators without proper training. Training is no longer being provided for the Baseline NCS instrument or for the Baseline NCS Clinical Reappraisal Interview Schedule. Training for use of updated versions of the NCS-R interview schedules and the WMH PAPI interview schedule is available through the World Health Organization CIDI Training and Reference Center at the University of Michigan. Contact Beth Ellen Pennell, Director of the CIDI TRC, at firstname.lastname@example.org or 734-222-8905. The Michigan CIDI TRC also offers training in an updated PAPI version of the NCS-R interview schedule. CAPI training includes a copy of the Blaise CAPI program, while PAPI training includes a copy of the Blaise Direct Data Entry program for keypunching the interview data. Both CAPI and PAPI training includes a copy of SAS programs to generate DSM-IV and ICD-10 diagnoses.
The Baseline NCS and NCS-R data are available as a public use dataset through the Inter-university Consortium for Political and Social Research (ICPSR).
Questions about downloading or working with the public use NCS dataset should be directed to SAMHDA staff at: email@example.com.
Questions about downloading or working with the public use NCS-R dataset should be directed to ICPSR staff at: firstname.lastname@example.org.
All other questions and comments including those regarding the website please contact: NCS@hcp.med.harvard.edu.
As opportunities for training in the NCS-R interview schedule are available, only brief questions of clarification about this instrument will be answered. Questions about the NCS-R and NCS-2 interview schedule should be sent to NCS@hcp.med.harvard.edu. As the NCS-2 interview schedule was developed specifically for use in following-up the Baseline NCS sample, no arrangements have been made for training in this interview schedule or for making the CAPI program for this interview schedule available. Miriam Gibbon from New York State Psychiatric Institute trained the NCS-R and NCS-2 clinical interview staff in the SCID. Miriam Gibbon and Molly Howes modified the SCID for use in these surveys.
Baseline NCS Interview Schedule and Respondent Booklet (including Tobacco supplement)
All content © 2005 Harvard Medical School