Here - American Geriatrics Society

P OSTER
A BSTRACTS
A161
Discharge to subacute rehabilitation as a predictor of 30 day
readmission in elderly patients with acute decompensated heart
failure.
W. Qureshi, 1 F. Khalid, 1 Z. Alirhayim, 1 M. Al-Mallah. 2,3 1. Internal
Medicine, Henry Ford Health Systems, Detroit, MI; 2. Department of
Cardiology, King Abdul-Aziz Hospital, Riyadh, Saudi Arabia; 3.
Wayne State University, Detroit, MI.
Background: Acute decompensated heart failure is the admission
diagnosis in 1/10th of all the hospitalization in US. Many of these
patients may be discharged to rehabilitation centers or home with
physical therapy. It is not known if the patient’s discharge disposition
is a predictor for 30 - day readmission. The aim of the study was to determine
if there was a difference in readmissions in patients discharged
to home with physical therapy compared to subacute rehabilitation
center.
Methods: This is a single-center retrospective parallel group cohort
study. Chart reviews of all the patients from 2009 - 2010 with primary
DRG of acute decompensated heart failure was carried out.
Primary outcome was 30-day all cause readmission. Secondary outcome
was 1-year mortality. The baseline characteristic differences
were evaluated by chi-square test and t-test. Logistic regression was
used to adjust for baseline variables hypertension, diabetes, dysplipidemia,
presence of coronary artery disease, age, gender to predict the
binary outcome as readmission versus no readmission based on the
discharge disposition.
Results
Out of 1543 total admissions, 542 were discharged to either
home with home physical therapy or subacute rehabilitation center.
The mean age was 75.3 +/- 7.0 years and females were 61%. Out of
these, 324 (59.8%) were discharged to subacute rehabilitation center
and ischemic cardiomyopathy was present in 335 (61.8%) of the patients.
There were 256 (47%) 30 day readmissions to the hospital.
About 66 (12.2%) of the patients died within 1 - year. Discharge to
subacute rehabilitation was associated with higher readmission rates
(adjusted OR 1.84; 95% CI, 1.19 – 2.83) p = 0.005 while, there was no
difference in the secondary outcomes. There was no significant difference
in the all-cause mortality (adjusted OR 1.56; 95% CI, 0.84 – 2.9)
p = 0.16.
Conclusion
This study shows that even after adjustment for baseline characteristics,
discharge to subacute rehabilitation was a weak but independent
predictor of 30 – day readmission. This might be because of
early detection of an exacerbation of acute decompensated heart failure
at subacute rehabilitation center as compared to at home; however
there was no significant association of mortality with the discharge
disposition.
A162 Encore Presentation
Fighting the weekend trend: increased mortality in older adult TBI
patients admitted on weekends.
S. A. Hirani, 1,2 E. B. Schneider, 2 H. L. Hambridge, 2 E. R. Haut, 2
A. R. Carlini, 2 R. C. Castillo, 2 D. T. Efron, 2 A. H. Haider. 2 1.
University of Texas Southwestern Medical School, Dallas, TX; 2.
Surgery, Johns Hopkins School of Medicine, Baltimore, MD.
Background: Weekend admission is associated with mortality in
cardiovascular emergencies and stroke. We sought to determine
whether weekend vs. weekday differences exist for older adults with
substantial head trauma.
Methods: Data from the 2006-2008 Nationwide Inpatient Sample
were combined and head trauma admissions were isolated. Abbreviated
Injury Scale (AIS) scores were calculated for using ID-
CMAP. Individuals aged 65 to 89 years with head AIS = 3 or 4 and no
other region score > 3 were included. Individuals with missing mortality,
gender or insurance data were excluded. Wilcoxon rank sum
and Student t-tests compared LOS, demographic, and total charge
data. Chi-square tests compared gender and head injury severity. Logistic
regression modeled mortality adjusting for age, gender, injury
severity, comorbidity and insurance status.
Results: Of the 39,728 patients meeting criteria, 10,189 (25.6%)
were admitted on weekends. Mean age was similar but more weekend
admissions were female (51.5% vs. 50.0%, p = 0.011). Weekend
patients demonstrated slightly lower comorbidity (mean Charlson =
1.09 vs. 1.15, p < 0.001) and head injury severity (58.5% vs. 60.9%
AIS=4, p < 0.001). Median weekend LOS was shorter (4 vs. 5 days, p <
0.001), but total charges did not differ. Proportional mortality was
higher among weekend patients (9.4% vs. 8.4%, p = 0.001). After adjustment,
weekend patients demonstrated 17% increased odds of
mortality (O.R. 1.17, 95% CI: 1.08-1.26).
Conclusions: Older adults with traumatic brain injury admitted
on weekends are less severely injured, carry less comorbidity and
generate similar total charges compared to those admitted weekdays.
However, weekend patients demonstrated 17% greater odds of
mortality.
A163
Postoperative Cognitive Complications in the Elderly: Risks and
Outcomes.
S. J. Agarwal, M. G. Erslon, S. D. Kelley. Covidien, Mansfield, MA.
Supported By: Covidien
Background: Cognitive complications occur in elderly patients
following surgery. The burden of postoperative cognitive complications
(POCC) is difficult to assess due to differences in determining
and documenting POCC. We utilized a large administrative database
to assess the incidence, risk factors and patient outcomes associated
with POCC in elderly patients across the top 5 major diagnostic categories
(MDC).
Methods: Inpatient surgical discharges in the top five MDCs for
patients older than 50 years from the Premier Perspective® Database
(Premier, Inc.) for CY2010 were selected. POCC was identified by
ICD9 codes 780.09 and 293.0. We determined incidence and used
multivariate logistic regression to estimate the adjusted risk of POCC
based on age (reference: 50-64 years old), admitting status, general
anesthesia (GA), ICU stay, comorbidity and MDC type (reference:
circulatory system surgeries). We compared mortality, length of stay
(LOS), costs and discharge disposition between patients with and
without POCC.
Results: 576821 discharges were selected, and 6553 (1.1%) were
coded for POCC. The risk of POCC increased with age (85+ years:
311% increased risk), ICU stay (183% increased risk), non-elective
admission (94%), and GA (19%). Compared to circulatory system
surgeries, other surgeries increased risk of POCC: musculoskeletal
(158% increase), digestive system (86%), nervous system (48%) and
hepatobiliary/pancreatic (46%). Patients with POCC had increased
inpatient resource utilization and compromised outcomes (Table 1).
Conclusion: Although the incidence of POCC based upon ICD9
coding was limited, the reported cases were associated with significantly
increased inpatient mortality, facility discharge, LOS and costs
AGS 2012 ANNUAL MEETING
S71