Call for Papers
Dear Colleagues,
The National Institutes of Health (NIH), Chestnut Health Systems, and the University of Illinois at Chicago will host the International Conference on Outcomes Measurement (ICOM): Focus on Chronic Conditions on September 1-3, 2010. The conference will be held at the NIH Natcher Conference Center near Washington, D.C. The goal is to advance our understanding of health screening and outcome issues in chronic health conditions, and applications of modern measurement models in these and related fields. Chronic illness outcomes applications and sessions concerning new theoretical developments in measurement will be included.
SUBMISSION DEADLINE EXTENDED TILL MAY 21ST!
Tell us about your interest in speaking using the Presentation/Poster Form or the Symposium Proposal Form. Please respond as soon as possible. All submissions are due by May 21st, 2010 and must be sent to Shweta Malladi, ICOM Coordinator, at smalla2@uic.edu
Specifically, we are seeking speakers who can present papers addressing conference Objectives on the following key topics, as well as other pertinent issues:
- Scoring and clinical interpretation issues in modern measurement. We solicit presentations on how modern measurement models and applications can be made more relevant to clinicians and consumers. These presentations should illustrate practical advantages of modern measurement over methods based in classical measurement theory. Also encouraged are applications that address multi-dimensional models.
- Ensuring the comparability of health outcomes assessment across diverse population sub-groups. This topic can include explorations of how diverse populations perceive items, phrases, and words differently, i.e., are there "problem" items, phrases, words that cause response biases and score differences that lead to misinterpretation? When are the differences "real" and important in enabling more accurate interpretation? Can this information be used to construct culturally equivalent measures?
- "Interpretation" of changes. Does the meaning, difficulty, or validity of items change over time and/or context? How do we distinguish these measurement problems from "real" change? Issues include standard setting methodology, pre-post assessment, establishing "clinically meaningful category intervals", and standards of change within the context of defining minimally important differences.
- Measurement criteria for model choice. Many philosophical and practical differences exist among modern measurement models. Educating consumers and clinicians on the differences will help facilitate understanding of choosing the appropriate model depending on the research objective.
- Using person and item fit statistics to refine measures and target interventions more appropriately. In chronic health conditions, person misfit may provide insight into atypical clinical presentations rather than merely revealing malingerers or cheaters as is more common in education.
- Integrating measurement into multilevel modeling. Multi-level modeling, a.k.a. hierarchical linear modeling, makes assumptions regarding scale construction. Modern measurement can be used to assess item function of the measurement scales used in multi-level modeling as well provide estimates that are incorporated into the analytic model itself. The Rasch model is particularly appropriate for incorporation into multi-level models as item location can be interpreted in terms of "severity" along a unidimensional measured construct. Violations of scale unidimensionality, local independence, and additivity can be examined; and, scales can be modified to increase the precision of the estimated models. The measurement model is considered as Level-1, with subsequent levels addressing individuals within groups, nested within specified settings. Item parameter estimates such as item location can be directly modeled, for example, time can be modeled as a predictor of item location at Time 2, and also modeled as a function of covariates.
- Computer Adaptive Testing (CAT) Applications. CATs are based on modern measurement models and are typically designed to improve efficiency without loss of precision or validity. Use of CAT in clinical settings is complicated by the need to address relatively small size of item banks (e.g., DSM-IV symptom counts), need to measure multiple correlated dimensions, atypical clinical presentations, and heterogeneous populations.
- Unidimensional and multidimensional models. Unidimensionality has been a hallmark of psychological measurement. Are there methods that can improve our assessment of essentially multidimensional constructs?
- Other innovative applications or approaches to improving the validity and efficiency of measurement. This includes using measurement models as part of implementation/fidelity studies, economic evaluation, or in combinations with other methods of analysis. We will also entertain proposals to look at non-IRT/Rasch methods of trying to improve the validity of measurement.
- Development of clinically useful cut scores to improve change assessment. Since modern measurement provides clinically relevant and interpretable item hierarchies, these may be useful in defining levels of functioning that provide clinical benchmarks that may help define clinically significant change.
Included among the invited speakers and organizers for ICOM are: David Andrich, David Cella, Karon Cook, Ann Doucette, Carl Granger, Ron Hambleton, Thomas Hilton, Richard Lennox, Robert Massoff, Paul Pilkonis, Bryce Reeve, Steven Reise, Barth Riley, Brian Rush, Jackson Stenner, Agustin Tristan, Wim van der Linden, and Craig Velozo.
As we plan for the conference, we seek speakers who are interested in presenting an aspect of their work that deals with chronic illness screening and outcomes measurement. Tell us about your interest in speaking using the Presentation/Poster Form or the Symposium Proposal Form. Please respond as soon as possible. All submissions are due by May 1st, 2010 and must be sent to Shweta Malladi, ICOM Coordinator, at smalla2@uic.edu
Sincerely,
|
Ken Conrad, Ph.D. Program Co-Chair University of Illinois at Chicago |
Barth Riley, Ph.D. Program Co-Chair Chestnut Health Systems |
|
Thomas Hilton, Ph.D. Project Officer National Institute on Drug Abuse |
Michael Dennis, Ph.D. Chestnut Health Systems |
