Style D square 42

1
Developing Patient Reported Outcomes Measurement
Information Systems for Spinal Cord Injury (SCI)
and Traumatic Brain Injury (TBI) David Tulsky,
Ph.D.1, Pamela A. Kisala, M.A.1, David Victorson,
Ph.D.2, Noelle Carlozzi, Ph.D1 1-University of
Michigan, Ann Arbor, MI, 2-Northwestern
University, Evanston, IL
4. Results TBI
1. Introduction
3. Results SCI
2. Methods

• In 2002, the National Institutes of Health (NIH)
  developed a "roadmap" for medical research in the
  21st century and identified that the lack of
  Dynamic Assessment of Patient Reported Outcomes
  for individuals with chronic disease as a
  significant gap in biomedical research.
• As a result, the NIH developed a series of large
  measurement initiatives designed to re-engineer
  the research process. These include the PROMIS
  (www.nihpromis.org), Neuro-QOL (www.neuroqol.org),
  and NIH Toolbox for Neurological and Behavioral
  Functioning (www.nihtoolbox.org). The goal of
  these efforts is to provide researchers with
  state-of-the-art measures that can be used as
  common data elements throughout biomedical
  research across different populations.
• Traumatic injuries such as Traumatic Brain Injury
  (TBI) and traumatic Spinal Cord Injury (SCI)
  result in a sudden and devastating constellation
  of physiological changes, cognitive deficits,
  secondary complications, and impaired physical
  activities, and a host of psycho-social issues
  and limitations.
• Generic measurement tools may not be sensitive
  nor have specificity for use in TBI and SCI
  clinical trials research.
• There is a need for development of targeted
  outcome measurement tools for TBI and SCI
  populations that are linked with large scale NIH
  initiatives (e.g., PROMIS/Neuro-QOL).
• The current work aims to develop new targeted,
  patient-reported outcome (PRO) measures of
  health-related quality of life (HRQOL) for TBI
  and SCI populations that are linked with the
  Neuro-QOL and PROMIS measurement systems.

• Focus Groups
• We conducted an extensive series (n47 groups) of
  stakeholder focus groups modeled on the
  semi-structured Neuro-QOL focus group
  methodology. Stakeholders were individuals with
  SCI (24 groups, total n136)or TBI (7 groups,
  n34), clinicians/providers (12 groups, n99),
  and caregivers (4 groups, n17).
• Qualitative Analyses
• Two independent investigators reviewed
  transcripts and developed a hierarchical taxonomy
  (codebook). All transcripts were selectively
  coded by two independent raters using a consensus
  coding methodology3. Relative frequency of
  mention of each topic area served to guide
  subdomain selection and item development.
• Item Development
• PROMIS / Neuro-QOL methodology was used to
  evaluate and include relevant Neuro-QOL item
  banks and a significant number of PROMIS items in
  the SCI and TBI scales to allow for cross-disease
  comparison. We also developed new items based
  upon focus group feedback , expert input, and
  literature review that were not already covered
  by Neuro-QOL and/or PROMIS.
• Large-scale Calibration Testing
• The resulting SCI-QOL and TBI-QOL measures are
  currently being tested in a national, multisite
  sample of n600 individuals with SCI and n600
  individuals with TBI, respectively.
• Main Outcome Measures
• TBI specific multidimensional Quality of Life
  measure (TBI-QOL) SCI specific Physical
  Functioning/Activity Limitation measure
  (SCI-CAT) SCI multidimensional Quality of Life
  Measure (SCI-QOL) Neuro-QOL Global (Social,
  Emotional, Cognitive, Physical) and Targeted
  (Fatigue, Personality Change) QOL Scales PROMIS
• Next Steps
• Upon completion of data collection, factor
  analyses will be conducted on all items banks.
  Item response theory (IRT) analyses will be
  conducted on all item banks, and computerized
  adaptive test (CAT) versions of each bank will be
  developed and made publicly available on the
  PROMIS Assessment Center SM.

PROMIS items imbedded in Neuro-QOL
PROMIS items imbedded in Neuro-QOL
5. Conclusions
6. References

• There are significant HRQOL issues relevant for
  individuals with TBI and SCI that have not been
  incorporated into large measurement scales like
  PROMIS and Neuro-QOL.
• Expanding the content areas with relevant domains
  can enhance HRQOL measurement for clinical trials
  in these important patient groups.
• Incorporating verbatim items from Neuro-QOL and
  PROMIS can allow for comparisons across studies
  and across patient group populations.
• The SCI-QOL and TBI-QOL outcome measures have the
  potential to become gold standard measures in
  clinical trials research.

• White, G. W., Suchowierska, M., Campbell, M..
  (2004) Developing and systematically implementing
  participatory action research..
  Arch.Phys.Med.Rehabil ,85(4 Suppl 2), S3-12.
• Glaser BG Strauss AL. (1967) The Discovery of
  Grounded Theory Strategies for Qualitative
  Research. New Brunswick AldineTransaction.
• Kisala, P., and Tulsky, D. (In press).
  Opportunities for CAT applications in medical
  rehabilitiation Development of targeted item
  banks. Journal of Applied Measurement.

Acknowledgements
Supported by grants H133N060022, H133G070138, and
H133A070037 from the National Institute of
Disability and Rehabilitation Research (NIDRR)
and grant 5R01HD05469 from the National
Institutes of Health (NIH).