(p 0.001) in their mean AMPAC scores in all three domains. In contrast, those receiving SNF care had no significant changes. Regres- sion analysis controlling for initial stroke severity, pre-stroke level of function, and other demographic variables supported this finding in two of the three AMPAC domains. Conclusions: For mobility and cognition, treatment in an IRF or HH/OR in this stroke cohort was associated with more improvement in functional outcomes at 6 months than those who went to a SNF. Further research is needed to determine how much of this difference might be due to the characteristics of the patients vs. the composition of the care they received. Key Words: Stroke; Rehabilitation; Post-acute care; HMO; Function. Poster 78 The Effects of Direct Current Brain Polarization on Motor Recovery of Lower Extremity in Stroke. Hong Min Kim (Asan Medical Center, Seoul, Republic of Korea), Sang Il Lee, Min Ho Chun. Disclosure: None disclosed. Objective: To investigate the effects of tDCS on the motor recov- ery of lower extremity in early stroke patients with hemiparesis. Design: Two-group randomized controlled trial, with pre-treatment and post treatment measures. Setting: Rehabilitation clinics. Partici- pants: Twelve, hemiparetic stroke patients are participated in this study. All subjects were able to walk independently regardless using single cane. Interventions: Nine patients were anodal tDCS group and another nine patients were sham tDCS group. The anode was posi- tioned on hot spot in the lesional motor cortex of the low extremity. Hot spot was positioned by the induction of motor evoked potentials (MEPs) in the hemiparetic tibialis anterior muscle using transcranial magnetic stimulation. tDCS was delivered for 30 minutes at 2 mA for 10 days, simultaneously both group received conventional gait and balance training. Main Outcome Measures: Before and after tDCS session, we evaluated the latency and amplitude of MEPs, gait anal- ysis, lower extremity of Fugl-Meyer assessment(FMA), Functional ambulatory category(FAC), Berg balance scale(BBS), stability in- dex(SI) and weight distribution index(WDI) of posturography. Re- sults: The latency and amplitude of MEPs, cadence, speed and stride length of gait analysis, FMA, MI, FAC, BBS, SI and WDI of postu- rography were improved in the anodal tDCS group(p  0.05), and real group had significant improvement in resting latency, activation la- tency, activation amplitude, FMA compared to sham group. Conclu- sions: tDCS on the primary motor cortex of lower extremity improved motor function and cortical excitability but did not affect gait and balance control in stroke patients. We can anticipate that tDCS may be a successful adjuvant therapeutic modality to improve the motor performance of lower extremity in stroke patients. Key Words: Rehabilitation. Poster 79 Persistence and Prevalence of Pain in Adults with Hemiparesis: Prospective Observational Study of 51 Patients. Samuel Bierner (UT Southwestern, Dallas, TX), Jose Barreto, Judith Hembree, David Garrigues, Jennifer Zahn. Disclosure: None disclosed. Objective: To determine the prevalence, persistence and character- istics of pain in adults with acquired hemiparesis. Design: Prospective cohort study of patients with hemiparesis. Setting: Conducted at a large urban teaching hospital. Patients: Subjects were recruited from PM&R clinics and by physician referral. Patients with acquired hemi- paresis were recruited regardless of the presence of pain. Interven- tions: This observational study measured the severity of stroke by the NIH Stroke Scale, and motor scores for upper and lower extremity strength; we assessed subject’s functional abilities with the Fugl- Meyer motor score (FMMS) and the Timed Up and Go (TUG) walking test. Pain was quantified using a 100-mm visual analogue scale (VAS), the NWC (number of words chosen) method for the short form of the McGill Pain Questionnaire (MPQ), the Pain Disability Questionnaire (PDQ), and a pain drawing. A two-tailed P value of less than 0.05 was considered significant. Main Outcome Measures: The primary study outcome measures were the VAS, MPQ, and PDQ measures. Results: Pain was both highly prevalent and persistent. The odds ratio was 4 [0.396 to 196.99] that pain would persist beyond one month. Motor impairment and NIHSS did not correlate with the pain rating. The PDQ was the strongest predictor of pain (p.0001). The TUG results were strongly predicted by the NIHSS (p=.0096). The NWC method of assessment of the MPQ was significantly regressed to the square root of the VAS rating (p.0001). Conclusions: Persons surviving stroke have clinically significant pain in approximately 50% of cases, persisting beyond one month. The severity of the neurological insult does not predict the likelihood of suffering pain. Key Words: Stroke; Pain, intractable; Pain measurement; Muscle spasticity; Rehabilitation. Poster 80 Physical Therapy Activities in Stroke, Knee Arthroplasty, and Traumatic Brain Injury Rehabilitation and Functional Outcomes. Jean Hsieh (National Rehabilitation Hospital, Wash- ington, DC), Gerben DeJong, Koen Putman, Randall Smout, Su- san Horn, Wenqiang Tian. Disclosure: None disclosed. Objectives: (1) Examine similarities and differences in the mix of physical therapy activities among patients from different impairment groups, and (2) examine if the same PT activities are associated with functional improvement across impairment groups. Design: Prospec- tive observational cohort study. Setting: Inpatient rehabilitation facilities. Participants: Patients with stroke, total knee arthroplasty (TKA), and traumatic brain injury (TBI). Interventions: Physical therapy activities. Main Outcome Measures: Functional Indepen- dence Measure (FIM). Results: Study group consists of 433 patients with stroke, 429 patients with TKA, and 207 patients with TBI. All three groups had similar admission motor FIM scores (between 46.4 and 50.1) but varying cognitive FIM scores (between 19.9 and 31.1). Patients with TKA spent more time on exercise (31.7 minutes/day) than the other two groups (6.2 minutes/day). Patients with TKA received the most intensive PT (65.3 minutes/day), whereas TBI group obtained the least intensive PT (38.3 minutes/day). Multivariate anal- yses showed that only two PT activities were significantly associated with better discharge motor FIM outcomes across all three groups, namely, gait training and community mobility. Three PT activities were negatively associated with discharge motor FIM outcome, namely assessment time, bed mobility, and transfers. The majority of patients were discharged to home across three groups (92%). Con- clusions: All three patient groups spent a considerable portion of their PT time in gait training relative to other activities. Both gait training and community mobility are higher level activities that are positively associated with outcomes, although all three groups spent little time in community mobility. Further research studies, such as predictive va- lidity studies, are needed to investigate if higher level or more inte- grated therapy activities are associated with better patient outcomes. Key Words: Physical therapy; Rehabilitation; Treatment outcome. Poster 81 Five-year Mortality and Related Prognostic Factors for Persons with a Stroke Initially Admitted to a Rehabilitation Centre. Liesbet De Wit (Katholieke Universiteit Leuven, Leuven, Belgium), Koen Putman, Willy De Weerdt. Disclosure: None disclosed. Background and Purpose: Few studies have investigated long- term mortality of stroke patients admitted to rehabilitation units. This study aimed to determine five-year mortality and its association with baseline characteristics and functional status at six months, for patients who received inpatient rehabilitation. Methods: In four rehabilitation centers, 532 stroke patients were recruited consecutively. Predictors were recorded on admission. Barthel Index (BI) was assessed at six months and patients were followed up for five years after stroke. The risk of dying was computed using Kaplan-Meier curves and compared 1715 2011 ACRM-ASNR ANNUAL CONFERENCE ABSTRACTS Arch Phys Med Rehabil Vol 92, October 2011