Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Saturday, January 13, 2007: 10:00 AM-11:45 AM
Seacliff B (Hyatt Regency San Francisco)
Validity Issues with Standardized Instruments in Community Mental Health Settings
Organizer:Catherine Greeno, PhD, University of Pittsburgh
 Screening for Anxiety and Depression in Community Mental Health: An Evaluation of a Series of Brief Screens
Catherine Greeno, PhD, Shaun M. Eack, MSW, Jonathan B. Singer, LCSW
 An Exploratory Factor Analysis of the Barriers to Treatment Participation Scale: Implications for Future Barriers Scale Development
Courtney Colonna-Pydyn, MPH, Christopher D. Gjesfjeld, MSW
 The Columbia Impairment Scale: A Confirmatory Factor Analysis
Jonathan B. Singer, LCSW, Shaun M. Eack, MSW, Kevin H. Kim, PhD
 Keeping Interventions in Mind: The Mos-Sss and Social Support Measurement in Community Settings
Christopher D. Gjesfjeld, MSW, Mary Lindsey Smith, MSW
 Do the DSM-IV Symptom Clusters Really Matter? a Confirmatory Factor Analysis of the PTSD Checklist-Civilian Version
Mary Lindsey Smith, MSW, Courtney Colonna-Pydyn, MPH
Abstract Text:
The importance of reliable and valid standardized instruments to measure need, outcomes and relevant aspects of context in community settings may be under-rated -- and the availability of such measures may be over-rated. Psychometrically sound instruments are needed to transfer evidence-based treatments into community settings; without them, it is impossible to know if poor results indicate that treatments do not work, or simply that we cannot accurately measure what happens in community settings. We have encountered serious problems with measurement in our community-based research. Standardized instruments are too long, and too cognitively complex, to work well in community mental health settings (Greeno, Colonna-Pydyn & Shumway, in press). Instruments have usually been standardized on people who are college-educated, employed full-time, and privately insured, which is not typical of community mental health consumers. Simpler versions of standardized instruments are badly needed. This symposium reports on-going psychometric analysis and presents recommendations for community researchers and providers. We conducted three closely-related studies in four community mental health centers on the mental health needs of mothers bringing their children for care. Women were recruited at the time they presented their child for treatment. Women and children completed standardized assessments. The key difference between studies was the method used to detect mental health need for the mothers. Study 1 used the Structured Clinical Interview for DSM-IV (SCID), the comprehensive and lengthy "gold standard" of research diagnosis. Study 2 used the Patient Health Questionnaire (PHQ), a version of the PRIME-MD, which is a shortened questionnaire version of the most prevalent illnesses in the DSM. Study 3 used the SCID and the PHQ. Sample sizes were 238, 291, and 50 for Studies 1, 2 and 3 respectively. The assessment battery was revised for each study, and sample sizes of symposium presentations reflect some protocol differences. Few studies have been conducted using community mental health samples. Furthermore, because about half of the women met criteria for active disorder, this sample is especially useful for testing concordance of diagnostic and screening instruments. Because the sample sizes are fairly large, psychometric tests aimed at shortening or simplifying instruments can also be conducted. This symposium presents original findings from 5 psychometric studies on 7 standardized measures commonly used in services and epidemiological research. Four are related to diagnostic screening; these include the PHQ, the Beck Depression and Anxiety Inventories, and the PTSD Checklist. Three measures address other issues, including child impairment using the Columbia Impairment Scale (CIS), Kazdin's Barriers to Treatment Scale, and the MOS Social Support Survey. Recommendations for appropriate use in community settings, and for shorter simpler instruments, where possible, are made. Each author participated across projects. Valid and reliable instruments are essential if we are to improve community mental health practice. Attaining valid and reliable instruments requires large studies, or coordination across many small studies. Social Work researchers, working across studies, and working with members of other disciplines, notably community mental health practitioners and psychometricians, can work together to create the excellent instrumentation needed for applied treatment research.

See more of Symposium

See more of Bridging Disciplinary Boundaries (January 11 - 14, 2007)