S181 S290
Poster 150:
Identifying Risk Factors for Acute Care Transfer
During Inpatient Rehabilitation
Cameron J. Collier, MD (University of Texas Southwestern, Dallas, TX,
United States), Benjamin Shahabi-Azad, MD
Disclosures:
Cameron J. Collier: I Have No Relevant Financial Re-
lationships To Disclose
Objective:
To identify clinical variables present on admission to
inpatient rehabilitation (IPR) that may indicate risk for transfer to
acute care services.
Design:
Case-control study.
Setting:
A 36-bed inpatient rehabilitation unit.
Participants:
862 patients admitted between July 1st 2015 - June 30th
2016 and subsequently discharged to community (control group) or
acutely transferred (case group).
Interventions:
Not applicable.
Main Outcome Measures:
Age, admission Functional Independence
Measures (FIM) score, date of admission, and length of stay were
available for both groups. Clinical variables including laboratory data,
vitals, blood product given within previous 72 hours, intravenous, oral,
and intramuscular antibiotics, and pressure ulcers present on admis-
sion were collected by chart review within the acute transfer group.
Results:
Descriptive statistics and logistic regression using XLSTAT
software was used for data analysis. Of 862 patients admitted to IPR,
733 (85.0%) were discharged to community, and 129 (14.9%) were
transferred to acute care. Admission FIM scores were a predictor for
acute transfer (p
<
.0001, OR 1.02). The most common diagnosis
prompting transfer to acute care was sepsis (86%) and respiratory
failure (13%). In subgroup analysis of acute transfers, 13% had received
blood product within 72 hours prior to admission, 31% had pressure
ulcers on admission, and 35% were receiving IV antibiotics on admission
to IPR. Receiving IV antibiotics was predictive of acute transfer within
72 hours of admission (p
¼
.017, OR 3.42).
Conclusions:
Among the patients admitted to IPR, lower admission
FIM scores were statistically significant but a weak predictor of acute
transfer when compared to control. Within the control group,
receiving IV antibiotics on admission was predictive of transfer within
72 hours. This suggests that clinical variables on admission may help
identify patients at risk for medical complications and acute transfer
during IPR.
Level of Evidence:
Level III
Poster 151:
Comparing Opioid & Benzodiazepine use in Acute
Inpatient Rehabilitation: A Quality Improvement
Project
Stacey A. Bennis, MD (McGaw MC of NW Univ NW Med Schl/RIC),
Kristen T. McCormick, DO, MS, Chris Butler, PharmD, BCPS,
James W. Atchison, DO, FAAPMR
Disclosures:
Stacey Bennis: I Have No Relevant Financial Relationships
To Disclose
Objective:
To assess concomitant usage opioid and benzodiazepine
medications (scheduled and/or as-needed) among patients in an acute
inpatient rehabilitation facility setting at the time of admission
compared to the time of discharge.
Design:
Retrospective Cohort Study.
Setting:
Acute Inpatient Rehabilitation Facility (AIRF).
Participants:
All patients above the age of 18 years old admitted to an
academic AIRF between 2/1/16 and 3/31/16 (n
¼
439) were included in
the study. Patients with chronic implanted pain pumps were excluded.
Interventions:
Not applicable.
Main Outcome Measures:
The primary outcome measure was to
determine the percent reduction in combined opioid/benzodiazepine
use from admission to discharge. Secondary outcome measures
included percent reduction in opioid usage from admission to
discharge, percent reduction in benzodiazepine usage from admission
to discharge, and reduction in moderate/high daily Morphine Milligram
Equivalents (MME) dosages from admission to discharge (defined as
>
50MME/day by the CDC “Guideline for Prescribing Opioids for Chronic
Pain”.
Results:
Of the 439 patients analyzed, 149 were admitted on opioids
(33.9%), 80 on benzodiazepines (18.2%), and 57 patients (13.0% of all
patients, 38.3% in the opioid subgroup) were being treated with
combined opioid/benzodiazepine therapy. At time of admission, 73
patients (16.6%) were on “unsafe” daily MME dosages (
>
50 MME/day)
as defined by the CDC guidelines. At the time of discharge, 19 patients
remained on dual opioid/benzodiazepine therapy (67% reduction), 112
patients remained on opioid monotherapy (25% reduction), and 43
patients remained on benzodiazepine monotherapy (46% reduction).
At the time of discharge, 45 patients (10.3% overall, 6.3% reduction)
were on “unsafe” daily MME dosages.
Conclusions:
This retrospective cohort study demonstrates trends for
a reduction in concomitant opioid/benzodiazepine use (67%), reduc-
tion in opioid and benzodiazepine monotherapy (25% and 46%,
respectively), and reduction in “unsafe” daily MME dosages (6.3%).
Inpatient physiatrists appear to be reducing concomitant opioid/
benzodiazepine use, benzodiazepine & opioid monotherapy treat-
ments, with a goal of reduce daily MME requirements.
Level of Evidence:
Level IV
Poster 152:
Improving Fall Rates Below the National Average in an
Acute Rehabilitation Hospital Using Modified
Screening Tools: A Retrospective Analysis
Raman Sharma, MD (Burke Rehabilitation Hospital, White Plains, NY,
United States), Marie Spencer, PhD, RN, CRRN, Barry Jordan, MD,
MPH, Karen M. Pechman, MD, Anne F. Ambrose, MD
Disclosures:
Raman Sharma, MD: I Have No Relevant Financial Re-
lationships To Disclose
Objective:
Hospital-related falls are a major cause of additional pain
and suffering, morbidity, mortality, length-of-stay and cost. At this
150-bed free-standing acute rehabilitation hospital, intense multi-
disciplinary efforts have reduced the fall rate significantly over the
past 5 years to below the national average. In this retrospective study
we sought to understand the factors responsible for the improvement.
Our hypothesis is that the modifications made to the screening tool
making it more sensitive, thus allowing more timely interventions has
been the major factor in driving the change.
Design:
This is a retrospective chart review of all admissions from
2011 to 2015. Data analysis was also done using information gathered
with E-Rehab tool. Information collected included patient de-
mographics, admission diagnosis, medical co-morbidities, functional
status, fall screening tools, data collection and analysis methods. Fall
related details were obtained from incident reports. IRB approval was
obtained.
Setting:
The setting is a 150-bed free-standing acute rehabilitation
hospital.
Participants:
All inpatient units (SCI, TBI, Stroke, Ortho, CardioPulm).
Interventions:
Modifications to screening tools to include fall-related
medications, seizures, confusion, movement disorders and orthostatic
hypotension.
Main Outcome Measures:
Falls reports.
Results:
Falls were defined using the standard AGS criteria. In 2011,
patient demographics were similar to national data (2011 CMI- 1.2088;
National CMI- 1.2784): fall rate was 5.43% (NDNQI participating hos-
pitals- 6.13%). At this time, Burke utilized the Hendrich II Fall Risk
Model, a standard tool used in most hospitals. In 2015, the patient
Burke CMI was 1.3099 (National- 1.3169). However our fall rate fell to
3.1% (NDNQI hospitals- 4.1%).
S181
Abstracts / PM R 9 (2017) S131-S290