S280 S290
Main Outcome Measures:
Muscle tone (AS) in upper and lower ex-
tremities, Functional Independence Measure (FIM), pain, and safety
were assessed after 6 months of treatment.
Results:
Mean (SD) AS in the affected lower limbs decreased by 0.99
(0.75) in the ITB group compared to 0.43 (0.72) in the CMM patients
(P
<
.05). Decrease of AS in upper extremities was 0.66 (0.59) versus
0.17 (0.70) in ITB and CMM groups, respectively (P
<
.05). FIM improved
in the ITB group by 2.68 (10.31) compared to a worsening in CMM arm
-2.58 (11.00) (P
¼
.054). In addition, ITB patients reported reduction in
actual, least and worst pain. Difference in the change from baseline to
month 6 between ITB and CMM was statistically significant in least and
actual pain (P
<
.05). Seven serious adverse drug reactions (SADR,
constipation, fecal impaction, epilepsy, peripheral edema, hypoten-
sion, 2 urinary retention) and 4 serious device reactions (device
dislocation, infection, catheter occlusion, intracranial hypotension)
were observed in the implanted patients (24% and 16% of patients,
respectively) versus 1 SADR (epilepsy) in the CMM group (3%). All
serious events related to the device or drug were treated and resolved.
Conclusions:
This is the first RCT clinical evidence showing superior
efficacy of ITB therapy compared to conventional oral medication in
decreasing post-stroke spasticity.
Level of Evidence:
Level I
CATEGORY: PRACTICE MANAGEMENT & LEADERSHIP
Poster 463:
The Effect of Patient Satisfaction Surveys on Physician Practice
Patterns
Joshua H. Levin, MD (Stanford University, Palo Alto, CA, United
States), David J. Kennedy, MD, Ryan Mattie, MD, Matthew Lungren,
MD MPH
Disclosures:
Joshua Levin: I Have No Relevant Financial Relationships
To Disclose
Objective:
To evaluate if patient satisfaction surveys affect the way
that physicians practice.
Design:
Survey.
Setting:
Internet based.
Participants:
Members of the Spine Intervention Society.
Interventions:
Not applicable.
Main Outcome Measures:
Voluntary physician responses to survey
questions.
Results:
74% of physicians use patient satisfaction surveys, and 14%
have an income affected by them. 32% of physicians agreed that the
surveys improve patient care, while 41% disagreed. 10% of physicians
agreed that the surveys should be used as a factor in determining a
physician’s compensation, while 75% disagreed. If a poor score would
negatively affect physicians’ incomes: 54% would refuse to accept a
new patient into his/her practice if he/she suspected the patient
might give a poor score, while 43% would discontinue the care of a
current patient; 41% would order more imaging test; 32% would
perform more invasive procedures; 31% would order more laboratory
tests; 29% would prescribe more pain medications; 27% would approve
unjustified disability paperwork; 19% would not tell an unsafe driver to
stop driving, while 15% would not report this driver to the DMV; 11%
would prescribe more antibiotics. 38% of respondents are aware of
practitioners who purposefully manipulate the system.
Conclusions:
The findings of this multi-specialty physician practice
survey suggest that routine implementation of patient satisfaction
surveys may affect the way that physicians practice medicine, and in
some circumstances lead to inappropriate care or denial of health care
services. This unintended practitioner behavior may be amplified
further when satisfaction scores are tied to physician compensation.
More thorough study is undoubtedly needed before patient satisfac-
tions scores can be used as a reliable metric in health care systems.
Level of Evidence:
Level V
CATEGORY: NEUROLOGICAL REHABILITATION
Poster 465:
Predicting Long-Term Global Outcome after
Traumatic Brain Injury (TBI): Development of a Practical
Prognostic Tool using the TBI Model Systems National Database
William C. Walker, MD (Virginia Commonwealth University, Richmond,
VA, United States), Katharine A. Stromberg, BS, Jennifer H. Marwitz,
MA, Adam P. Sima, PhD, Amma A. Agyemang, PhD, Kristin M. Graham,
PhD, CRC, Jeffrey S. Kreutzer, PhD
Disclosures:
William Walker: Research Grants - NIH R21 grant to VCU
supported this research
Objective:
Build decision tree tools to predict long-term functional
outcomes after moderate-severe closed TBI and test their accuracy in
an independent sample.
Design:
Cross-sectional analyses using flexible classification tree meth-
odology within the TBI-Model Systems (TBIMS) National Database (NDB).
Setting:
Enrollments occurred at 17 designated TBIMS inpatient
rehabilitation facilities. Follow-up assessments were conducted pri-
marily by telephone.
Participants:
All TBIMS NDB participants (adults with moderate or
severe TBI per eligibility criteria) injured between Jan 1997
e
Jan 2017
with closed TBI. Time point exclusions were death, vegetative state,
insufficient post-injury time, and unavailable outcomes. Analysis
sample sizes were 10,125 (year-1), 8,821 (year-2), and 6,165 (year-5).
Interventions:
Not applicable; Candidate predictors included de-
mographic, premorbid socioeconomic, and injury characteristics.
Main Outcome Measures:
Glasgow Outcome Scale (GOS) functional
levels at 1-, 2-, and 5-year post-injury.
Results:
The sample was 72.5% male, 9% Hispanic, and 20.9% minority
races withmedian injury age 39.0 years. In our built GOS prediction trees,
posttraumatic amnesia (PTA) duration consistently dominated branching
hierarchy, and was the lone injury variable to contribute. Lower order
variables contributing to the trees were limited to age, premorbid edu-
cation, productivity, and occupational category. Generally, patients
spending fewer days in PTA and who were younger, pre-morbidly more
productive, and more highly educated had better outcomes. Across all
prognostic groups, the best and worst Good Recovery rates were 65.7%
and 10.9% respectively, and the best and worst Severe Disability rates
were 3.9% and 64.1%. Predictability in test datasets ranged from C-sta-
tistic of 0.691 (year-1; CI: 0.675, 0.711) to 0.731 (year-2; CI: 0.724, 0.738).
Conclusions:
We developed an easy-to-use decision tree tool to provide
prognostic information on long-term functional outcomes in patients
with moderate-severe closed TBI, and demonstrated predictive accu-
racy in an independent test sample. Length of PTA, a clinical marker of
injury severity, was by far the most critical outcome determinant.
Level of Evidence:
Level II
CATEGORY: NEUROLOGICAL REHABILITATION
Poster 466:
Prospective Determination of Clinical Neurologic
Level of Injury with Early MRI Following Blunt Traumatic Spinal
Cord Injury
Lisa U. Pascual, MD, FAAPMR (Orthopaedic Trauma Inst, San Francisco,
CA, United States), J. Russell Huie, PhD, William D. Whetstone, MD,
Sanjay S. Dhall, MD, Geoffrey T. Manley, MD, PhD,
Jacqueline C. Bresnahan, PhD, Michael S. Beattie, PhD, Vineeta Singh,
MD, Adam R. Ferguson, PhD, Rachel E. Tsolinas, BA, Jason F. Talbott,
MD, PhD
Disclosures:
Lisa Pascual: I Have No Relevant Financial Relationships
To Disclose
S280
Abstracts / PM R 9 (2017) S131-S290