Here - American Geriatrics Society

P OSTER
A BSTRACTS
The aim of this study was to determine if osteoarthritis (OA)-related
biomarker levels were associated with hand symptoms severity
and functional disability as reported by the Australian/Canadian
Hand Osteoarthritis Index (AUSCAN) among middle- and olderaged
adults.
Methods
AUSCAN total scores and biomarker measurements were
available for 661 participants from the Johnston County Osteoarthritis
Project. AUSCAN is a 15-item self-report measure of hand pain,
stiffness, and function that has been previously validated among individuals
with hand OA. Higher AUSCAN scores indicate worse pain
or function in the hand. Commercially-available kits were used to
measure levels of: serum hyaluronic acid (HA), serum carboxy-terminal
propeptide of type II collagen (sCPII), serum type II collagen
degradation product (C2C), urinary C-terminal crosslinked telopeptide
of type II collagen (CTX-II) , and urinary N-terminal crosslinked
telopeptide (NTX); serum cartilage oligomeric matrix protein
(sCOMP) was measured with an in-house sandwich enzyme linked
immunosorbent assay. The ratio of sC2C:sCPII was calculated. Spearman
correlation coefficients were computed between AUSCAN and
each biomarker. Linear regression models were used to estimate associations
between AUSCAN total score and a biomarker, adjusting
for age, race, current smoking status (yes/no), BMI, current alcohol
usage, and radiographic hip, knee, and hand OA, each defined by
Kellgren-Lawrence grade 2-4 in at least one joint in that joint group.
Results
After adjusting for possible confounders, AUSCAN total score
was positively associated with uNTX-1, sCOMP, and negatively associated
with C2C, CPII, and their ratio. The adjusted associations between
AUSCAN scores and uCTX-II and HA were not statistically
significant.
Conclusion
Even after adjusting for other factors and concomitant radiographic
hip, knee, and hand OA, the AUSCAN index for hand symptoms
and function was independently associated with biomarkers related
to type 1 collagen resorption, non-collagenous matrix protein
degradation, and decreased type II collagen synthesis. These findings
suggest that biological processes related to these biomarkers are important
in hand OA and its symptomatic consequences.
B22
Does antipsychotic dose reduction result in worsening behavior
among nursing home residents with dementia: a systematic review of
the literature.
J. Tjia, A. Kanaan, J. Donovan. Geriatrics, University of Massachusetts
Medical School, Worcester, MA.
Supported By: This study was supported by a grant from the Agency
for Healthcare Quality and Research.
Background: While federal regulations require gradual dose reduction
trials of antipsychotics prescribed for behavior management
in nursing home (NH) residents with dementia, widespread concern
about precipitating behavioral disturbances limits implementation.
We conducted a systematic review of the literature to identify clinical
trials of antipsychotic dose reductions. The study aim was to describe
dose reduction practices and evidence for risk of behavior escalation.
Methods: A comprehensive literature search was conducted in
MEDLINE, EMBASE, and International Pharmaceutical Abstracts
between January 1970 and October 2011 using the terms “antipsychotic
agent or neuroleptic agent,” “dementia,” “nursing homes,” and
“withdrawal.” One investigator reviewed abstracts for inclusion
based on: English-language, human subjects, clinical trial, nursing
home site, and reported reduction in antipsychotic medications. We
excluded review articles, commentaries and secondary analysis of
main trial results. The remaining articles were reviewed by 2 investigators
for final inclusion, resulting in 9 articles.
Results: The nine articles meeting inclusion criteria included
randomized controlled trials of both typical and atypical antipsychotics.
Study populations ranged in size from 55 to 183 NH residents
with dementia and dose reductions typically targeted patients who
were not psychotic and did not have a history of violent behavior.
Gradual dose reduction protocols typically followed a strategy of
50% dose reduction per week for 2-3 sequential weeks. Outcomes
measured included behavioral problems, cognitive function, and resumption
of antipsychotic medications. All 9 studies reported that the
majority of residents randomized to gradual dose reductions of antipsychotics
had no overall detrimental effect on functional and cognitive
status, or exacerbation of behavioral symptoms.
Conclusions: Clinical trials evaluating the withdrawal of antipsychotic
medications from NH residents with dementia do not
show evidence of rebound behavioral escalation after gradual dose
reductions.
B23
Is Fracture Risk Assessment Index (HFRAI), an Electronic Medical
Database Derived Tool, comparable to Femur neck Bone Mineral
Density (FNBMD)?
M. Albaba, S. S. Cha, P. Y. Takahashi. Mayo Clinic, Rochester, MN.
BACKGROUND
Femur neck bone mineral density (FNBMD) is the cornerstone of
hip fracture risk stratification in current clinical practice. Other risk stratification
tools use FNBMD with other selected risk factors. We have derived
and validated the Hip Fracture Risk Assessment Index (HFRAI)
(score range 0, 32) that uses electronic medical records data to predict
risk for hip fracture in community-dwelling older adults. Using appropriate
computer program,HFRAI is computed automatically to provide the
clinician with a readily available score that assesses patient’s risk for sustaining
hip fracture, without the need to manually calculate a score, perform
a physical examination or obtain a radiology test.It is unknown how
HFRAI compares to FNBMD and to other FNBMD based risk stratification
tools.The goal of this study was to compare HFRAI to FNBMD.
METHODS
This was a retrospective cohort study. All community-dwelling
subjects over 60 years of age in a primary care panel in Olmsted
County, MN on January 1st, 2005 were enrolled.
Since HFRAI uses electronic medical records data of the previous
two years, we excluded all subjects that did not have a FNBMD
tested during the two-year time frame (01/01/2003, 01/01/2005).
We created two receiver-operating curves (ROC), one ROC
using the HFRAI scores as of 01/01/2005, and the other using the lowest
recent FNBMD T-score. The primary outcome was incident hip
fracture in the subsequent four years (2005, 2008). We computed the
area under the curve (AUC) for each ROC. We used Z statistics to
compare AUC between the two ROCs. We evaluated each of
FNBMD and HFRAI for sensitivity and specificity.
RESULTS
On 01/01/2005 13,457 subjects over 60 years were empanelled in
the practice. 12,650 subjects (94%) consented to the study, among
which 1953 subjects had FNBMD-DXA (dual-energy X-ray absorptiometry)
scan within the previous two years. 83 subjects (4.3%) sustained
a hip fracture between 01/01/2005 and 12/31/2008. AUC for
HFRAI was 0.79, slightly larger than AUC for FNBMD of 0.74
(Z=1.82, p=0.09). The best combined sensitivity and specificity for
HFRAI tool was at a score of >12 (84% sensitive and 55% specific),
and for FNBMD tool was at a T-score of