APPROXIMATELY 500,000 MEDICARE patients are

934 ORIGINAL ARTICLE A Comparative Evaluation of Inpatient Rehabilitation for Older Adults With Debility, Hip Fracture, and Myopathy Patrick Kortebein, MD, Carl V. Granger, MD, Dennis H. Sullivan, MD ABSTRACT. Kortebein P, Granger CV, Sullivan DH. A comparative evaluation of inpatient rehabilitation for older adults with debility, hip fracture, and myopathy. Arch Phys Med Rehabil 2009;90:934-8. Objective: To compare the functional outcomes and discharge location of older adults admitted to inpatient rehabilitation for debility, hip fracture, and myopathy. Design: Retrospective cohort study from 2002 to 2003 with information from the Uniform Data System for Medical Rehabilitation (UDSMR). Setting: United States inpatient rehabilitation facilities subscribing to the UDSMR. Participants: Patients 65 years or older (N 84.701) with primary diagnoses of debility (n 14,835), hip fracture (n 68,915), and myopathy (n 951). Interventions: Not applicable. Main Outcome Measures: Change in functional status, including efficiency (change in functional status divided by length of stay in days) and discharge setting. Results: The efficiency of the patients with debility (1.7 2.1) was significantly lower than that of the patients with hip fracture (1.9 1.6; P.001), but not different from the patients with myopathy (1.6 1.4; P.3). Significantly more patients with debility (68%) were discharged home than the hip fracture and myopathy groups (66% and 65%, respectively; P.001). Conclusions: Although statistical differences exist, the functional recovery and rate of discharge home of older adult patients admitted to inpatient rehabilitation with a primary debility diagnosis are essentially the same clinically as those of patients with a diagnosis of either hip fracture or myopathy. Given these findings, and given that hip fracture and myopathy are approved medical conditions according to the Centers for Medicare and Medicaid Services 75% rule, the medical condition debility warrants consideration for inclusion as a qualifying medical diagnosis under this rule. However, further research is needed to develop relatively objective criteria for the debility diagnosis, and to identify those patients with debility who are most likely to benefit from inpatient rehabilitation. From the Central Arkansas Veterans Healthcare System, Little Rock, AR (Kortebein, Sullivan); Department of Physical Medicine and Rehabilitation (Kortebein) and Donald W. Reynolds Department of Geriatrics (Kortebein, Sullivan), University of Arkansas for Medical Sciences, Little Rock, AR; Uniform Data System for Medical Rehabilitation and Department of Rehabilitation Medicine, University at Buffalo, The State University of New York, Buffalo, NY (Granger). Presented to the American Academy of Physical Medicine and Rehabilitation, September 27, 2007, Boston, MA. A commercial party having a direct financial interest in the results of the research supporting this article has conferred or will confer a financial benefit on the author or one or more of the authors. Dr. Granger is an employee of Uniform Data System for Medical Rehabilitation (UDSMR), a division of UB Foundation Activities Inc. FIM and UDSMR are trademarks of UDSMR, a division of UB Foundation Activities Inc. Reprint requests to Patrick Kortebein, MD, Assistant Professor, PM&R and Geriatrics, University of Arkansas for Medical Sciences, Geriatric Research, Education and Clinical Center (3J/149), Central Arkansas Veterans Healthcare System, 2200 Fort Roots Dr, North Little Rock, AR 72214, e-mail: pmkortebein@uams.edu. 0003-9993/09/9006-00828$36.00/0 doi:10.1016/j.apmr.2008.12.010 Key Words: Hospitalization; Muscle weakness; Recovery of function; Rehabilitation. Published by Elsevier Inc on behalf of the American Congress of Rehabilitation Medicine. APPROXIMATELY 500,000 MEDICARE patients are treated in acute IRF across the United States annually. 1 Designation as an IRF has been guided by what is commonly referred to as the 75% rule from the CMS. 2 This rule states that in order for a facility to be classified as an IRF, at least 75% of the patients during a 1-year period must have 1 of a limited number of medical diagnoses. This legislation stimulated significant controversy within the rehabilitation community after a revision in 2004 that implemented a gradual transition back to the 75% compliance threshold. 2 This legislation also increased the number of medical conditions from 10 to 13 (appendix 1), although CMS noted that additional medical diagnoses had been recommended for inclusion. 2 In December 2007, the compliance threshold was changed to 60% by public law, although the number of medical conditions remained unchanged. 3 In response to comments that additional medical diagnostic categories should be added, CMS countered that they had not been able to identify any data to support the inclusion of other medical conditions. 2 From this document, it is also not clear what criterion CMS uses in deciding whether additional medical conditions should be incorporated into the 75% rule. However, a recent GAO report noted that medical condition alone was an insufficient criterion, and recommended that functional status should be considered in addition to medical diagnosis in identifying the types of patients appropriate for IRFs. 1 The American Academy of Physical Medicine and Rehabilitation has also recommended that a patient s functional deficits and their potential for recovery should be primary considerations in deciding which patients are appropriate for inpatient rehabilitative care. 4 One medical condition that has been suggested for inclusion in the 75% rule is debility (that is, deconditioning, or generalized weakness). 2 Approximately 5% of all Medicare patients admitted to an IRF in 2003 had a diagnosis of debility, 1 although this percentage may be even higher. 5 Recent studies indicate that older debilitated adults admitted for inpatient rehabilitation recover function just as quickly, if not more BMI CMS GAO ICD-9-CM IRB IRF LOS UDSMR List of Abbreviations body mass index Centers for Medicare and Medicaid Services Government Accountability Office International Classification of Diseases, 9th Revision Clinical Modifications institutional review board inpatient rehabilitation facilities length of stay Uniform Data System for Medical Rehabilitation

DEBILITY, HIP FRACTURE, AND MYOPATHY REHABILITATION, Kortebein 935 rapidly, than patients with more traditional rehabilitation diagnoses (eg, stroke, spinal cord dysfunction). 6-8 However, none of these studies have directly compared the outcomes of older patients with debility with those of other rehabilitation patient populations. We examined the functional recovery and discharge location of older adult patients admitted for acute inpatient rehabilitation with the primary diagnoses of debility, hip fracture, and myopathy. Hip fracture and myopathy were chosen for comparison because they seem to lead to similar functional deficits, and because both of these medical conditions are included in the CMS 75% rule (see appendix 1). We hypothesized that the functional recovery of the patients with a primary debility diagnosis would be clinically similar to that of the other 2 patient groups, and that the number of patients discharged home would be similar for all 3 patient groups. METHODS We have previously reported these data for older adult patients with a primary and secondary diagnosis of debility. 7 In this current article, data on patients with a primary diagnosis of debility only were compared with data on older adult patients ( 65y) with the diagnoses of hip fracture and myopathy. The same techniques and analyses were performed on this expanded data set and are described here in brief. As with our previous study, the UDSMR registry was the source of data, and analysis was limited to US data for the years 2002 to 2003. The UDSMR is the largest national registry of data collection from IRFs and currently collects data from approximately 70% of the rehabilitation facilities in the United States. For the years 2002 to 2003, data were available from 784 and 813 facilities, respectively. During this period, data were collected under the CMS medical rehabilitation prospective payment system. The UDSMR data are primarily used for quality assurance purposes, although these data have been reviewed by the CMS and are considered to be representative of Medicare beneficiaries. 9 Deidentified data on all patients admitted to a subscribing US inpatient rehabilitation facility during this time with a primary diagnosis of debility (ICD-9-CM code 799.3), hip fracture (ICD-9CM code 820 820.9), and myopathy (ICD- 9-CM code 359.4 359.9) were extracted from the UDSMR registry and provided to the investigators for analysis. All IRFs participating in CMS must use the Inpatient Rehabilitation Facility-Patient Assessment Instrument to submit data. As previously noted, data on patients with complex medical conditions (UDSMR impairment group 17) were not included. 7 As with our previous study, several patient characteristics were evaluated, including sociodemographic variables, prehospitalization living arrangements, admission location/facility, discharge disposition/location, LOS, and source of payment. Admission and discharge diagnoses were the same in 100% of the patients. Patient function was evaluated with the FIM instrument. FIM rating measurements are recorded within 72 hours of admission and at discharge by trained therapists and rehabilitation nurses. The total-fim instrument rating may range from 18 to 126 (ie, 18 items rated on a 1 7 ordinal scale), with a higher rating indicative of a higher level of function and independence. Therapists and rehabilitation nurses must complete a standardized FIM training course administered by UDSMR every 2 years to maintain certification. The validity and reliability of the FIM instrument has previously been reported. 10,11 Data from subscribing facilities are examined for consistency, and the FIM instrument has excellent interrater reliability (.90). 10 Rehabilitation efficiency and discharge location were the primary outcome measures for this study. Rehabilitation efficiency is reported as defined by Ottenbacher et al 8 ; specifically, efficiency is FIM effectiveness (total FIM instrument discharge rating minus total FIM instrument admission rating) divided by rehabilitation LOS (total number of inpatient rehabilitation days). Thus, given comparable LOSs, an individual with a greater increase in total FIM instrument rating will have a higher rehabilitation efficiency. Discharge location was dichotomized (discharged to home vs not to home). Other outcome measures included discharge to the community (ie, patient s/ relative s or another person s home, transitional living setting, board and care setting, assisted living residence), transfer to an acute care hospital, and death. This study proposal was reviewed and approved by the IRB of the University of Arkansas for Medical Sciences. The IRB of the University at Buffalo, The State University of New York, determined that no review was required because this study analyzed deidentified data. No informed consent was required for the same reason. The investigators had full independence in analysis of this data. Data Analysis Subjects with a primary diagnosis of debility were compared with subjects in the hip fracture and myopathy diagnostic groups for each outcome. Univariate statistics (analysis of variance) were used to test for differences in continuous variables, and contingency tables ( 2 ) for analysis of categorical data. All analyses were conducted using SAS statistical software. a Statistical significance was defined as a P value of less than.05. RESULTS Study Population As noted in our previous study, the UDSMR registry includes data on a total of 957,630 inpatient rehabilitation patients for the years 2002 to 2003. Table 1 provides information on the sociodemographic characteristics of these 3 groups of patients. There were substantially fewer patients with myopathy, although in general, the demographic characteristics of all 3 groups were quite similar. The patients with myopathy were, on average, younger, with a higher percentage married, while a significantly greater percentage of the patients with hip fracture were white women. More than 93% of the patients in all 3 groups had been living at home previously, and approximately one third lived alone. Nearly 90% of the patients with debility, and even greater percentages of the patients with hip fracture and myopathy, had been transferred to rehabilitation from an acute care hospital. Compared with the debility group, the period of disability prior to rehabilitation admission was significantly shorter for the patients with hip fracture, and significantly longer for the myopathy group. Medicare was the primary insurance for virtually all of these patients, regardless of diagnostic group. Outcomes Our primary outcomes of interest were rehabilitation efficiency and discharge location. Functional data, including admission and discharge FIM ratings, FIM effectiveness, rehabilitation LOS, and rehabilitation efficiency are displayed in table 2. The functional recovery and LOS of all 3 groups was very similar, although the efficiency of the debility group (1.7 2.1) was significantly less than that of the hip fracture group (1.9 1.6), but not significantly different from the patients with myopathy (1.6 1.4). The hip fracture group started rehabilitation at a slightly lower total FIM instrument rating, but had a slightly greater FIM effectiveness, resulting in a

936 DEBILITY, HIP FRACTURE, AND MYOPATHY REHABILITATION, Kortebein Table 1: Baseline Demographics Characteristics PDD (n 14,835) Hip Fracture (n 68,915) Myopathy (n 951) Age (y) (mean SD) 80 7 81 7 77 7 Age range (y) 65 108 65 107 65 97 Female 9114 (61) 51,720 (75) 581 (61) White 12,755 (86) 62,428 (91) 797 (84) Married 5839 (39) 25,055 (36) 452 (48) Prehospital living setting Home 13,850 (93) 64,252 (93) 914 (96)* Assisted living 561 (4) 2881 (4) 20 (2) NH/SNF 191 (1) 837 (1) 7 (1) Other 233 (2) 945 (1) 10 (1) Prehospital living with Relatives 8275 (56) 35,944 (52) 588 (62) Alone 5290 (36) 27,007 (39) 291 (31) Other 1270 (9) 5964 (9) 72 (8) Admission type Initial 13,916 (94) 66,287 (96) 886 (93) Admitted from Hospital 13,138 (89) 66,730 (97) 897 (94) Home 1091 (7) 479 (1) 28 (3) NH/SNF 371 (2) 1411 (2) 18 (2) Onset of current disability (d) (mean SD) 15 26 8 12 19 36 Medicare insurance (%) 96 92 93 NOTE: Values are n (%) or as otherwise indicated. Abbreviations: PDD, primary diagnosis of debility; NH/SNF, nursing home/skilled nursing facility. *P.05 compared with PDD group; P.001 compared with PDD group. slightly higher efficiency. In contrast, the patients with myopathy, on average, had a minimally higher admission total FIM instrument rating, but a lower efficiency because of their lower FIM effectiveness. Discharge location data, including transfers to an acute care hospital and death, are presented in table 3. A significantly greater percentage of the debility group was discharged home, home alone, and to the community. Transfer to an acute care hospital was relatively common for all 3 groups; the debility group fell in the middle (13%), with the myopathy group significantly more commonly (19%), and the hip fracture group significantly less frequently (8%), transferred to an acute hospital. A very low percentage of patients in all 3 groups died during rehabilitation, although patients in the hip fracture group (0.2%) were significantly less likely to die than patients in the debility or myopathy groups (0.5% and 0.6%, respectively). DISCUSSION The results of this study indicate that among older adults admitted to an inpatient rehabilitation facility during 2002 or 2003, the functional recovery of patients with a primary diagnosis of debility is very similar to that of older rehabilitation patients with a diagnosis of either hip fracture or myopathy. In addition, the patients with debility were just as likely to be discharged home as the patients with hip fracture and myopathy, and mortality rates were low for all 3 groups. To our knowledge, no previous study has directly compared the outcomes of these 3 groups of patients, and only a limited number of studies have examined the rehabilitation outcomes of patients with debility. 6,7,12 Raj et al 6 reported on the functional and discharge outcomes for a group of slightly younger (mean age 72y) patients with debility undergoing inpatient rehabilitation at a single facility. In this study, a geriatrician led the primary rehabilitation team, although physiatry consultation was mandatory for each patient. In addition to being younger than our group of patients with debility, this study population had a greater percentage of nonwhites (41%) and patients previously living alone (46%), and a shorter duration of hospitalization prior to rehabilitation admission (9d). The functional recovery (ie, rehabilitation efficiency) of their debility patients (2.4 1.9; corrected from 3.0 with erratum to be published; D. Carr, personal communication, January 2009) Table 2: Functional Outcomes Outcomes PDD (n 14,835) Hip Fracture (n 68,915) Myopathy (n 951) FIM admission 67.0 17.0 64.0 16.0 70.0 16.0 FIM discharge 87.0 21.0 88.0 20.0 89.0 20.0* Effectiveness 20.0 14.0 23.0 13.0 19.0 14.0* LOS (d) 13.0 7.0 14.0 6.0 13.0 8.0 Efficiency (FIM /d) 1.7 2.1 1.9 1.6 1.6 1.4 NOTE: Values are mean SD. Effectiveness [FIM instrument discharge rating FIM instrument admission rating] (see Methods section for details); Efficiency is defined as effectiveness/los in days (see Methods section for details). Abbreviation: PDD, primary diagnosis of debility. *P.05 compared with PDD group; P.001 compared with PDD group.

DEBILITY, HIP FRACTURE, AND MYOPATHY REHABILITATION, Kortebein 937 Table 3: Discharge Location Location PDD (n 14,835) Hip Fracture (n 68,915) Myopathy (n 951) Discharge location Home 10,125 (68) 45,815 (66) 622 (65) Home alone 2226 (15) 8976 (13) 129 (14) Community 10,779 (73) 48,598 (71) 650 (68) Acute care hospital 1870 (13) 5483 (8) 185 (19) Death 71 (0.5) 144 (0.2) 6 (0.6) NOTE: Values are n (%). See Methods section for explanation of community. Abbreviation: PDD, primary diagnosis of debility. P.001 compared with PDD group. was markedly higher than that of our group of subjects with debility, although the percentage discharged home (71%) was similar, and the rate of discharge back to the acute hospital setting (21%) was substantially higher. Jain et al 12 retrospectively examined the link between BMI and functional recovery in over 1000 debilitated patients who had been admitted to a single inpatient rehabilitation facility over a 6-year period. These patients were slightly younger than our debility group, and their rehabilitation LOS tended to be slightly longer. Although the rehabilitation efficiency of the entire group was not reported, the efficiency of each BMI group was a bit lower than that of our subjects. Several studies have examined the outcomes of older adult patients with hip fracture undergoing inpatient rehabilitation. Graham et al 13 used the UDSMR database to analyze outcomes of patients with hip fracture for the year 2003, and not surprisingly, their results were the same as those reported here. Other reports have examined the rehabilitative outcomes of older patients with hip fracture in Israel, and the rehabilitation efficiency of patients in these studies is lower than that of rehabilitation patients with hip fracture in the United States. 14,15 We were unable to identify any previous studies evaluating the rehabilitative outcomes of older patients with myopathy. The functional recovery of our 3 groups of patients compares favorably with that of more traditional inpatient rehabilitation patient populations, such as those with stroke, brain injury, and spinal cord injury, but is lower than that reported for orthopedic patients. 8 Similar to the patients in this study, the average FIM instrument admission rating for traditional rehabilitation patients is generally in the 60 to 70 range, with a mean increase of approximately 20 to 30 by discharge. 8 The resultant efficiency of most rehabilitation patient populations is approximately 2, although orthopedic patients tend to have shorter LOSs and thus higher efficiency ratings. Also, as noted, the patients with debility in the study by Raj 6 had an efficiency of 2.4. This information may be helpful in identifying patients, regardless of diagnosis, who are most likely to benefit from inpatient rehabilitation. The aforementioned AAPMR guidelines do provide recommendations regarding patient characteristics of importance in assessing patient appropriateness for inpatient rehabilitation. However, it may also be reasonable to investigate the use an upper and/or lower limit cutoff for the admission FIM instrument rating, for instance. Alternatively, rehabilitation efficiency ratings might also be used to assess progress objectively while a patient is participating in inpatient rehabilitation. This type of research would be in accordance with the GAO and AAPMR recommendations that patient function, in addition to primary medical diagnosis, be considered in deciding which patients are most appropriate for inpatient rehabilitation. 1,4 The rate of discharge home for our 3 patient populations was lower than that of more typical inpatient rehabilitation populations. 8 This may be because our patient populations were older, with fewer married patients. Similar comparisons cannot be made regarding the rate of transfer to an acute hospital setting because there are insufficient data. In our patients, the hospital transfer rate was significantly higher for the patients with myopathy than the debility group, which was greater than that of the hip fracture group. While Raj 6 reported an even higher rate of acute hospital transfers (21%) for patients with debility, we were unable to identify any similar acute hospital transfer data for other rehabilitation populations during rehabilitation. However, other studies have reported similar rehospitalization rates in the months after rehabilitation discharge. 16,17 The overall mortality rate during inpatient rehabilitation in our patient groups was much lower than that reported for more traditional rehabilitation populations. 8 However, the patients with debility and myopathy were more than twice as likely to die during rehabilitation as those in the hip fracture group. Study Limitations The main limitations of this study include those inherent in a large database analysis (eg, coding and reporting errors), as well as the relatively limited number of patients with myopathy, and the potential diagnostic misclassification of the patients with debility. Because debility is not one of the approved CMS 75% rule diagnoses and has no specific diagnostic criteria, these patients may have been classified in another diagnostic category, such as myopathy. Myopathy is a logical candidate, because the clinical picture is quite similar to debility, and because myopathy is one of the diagnostic categories in the CMS 75% rule. However, our data were collected prior to active enforcement of the compliance threshold in 2004, and it has previously been documented that only 6% of IRFs were compliant with the 75% rule in 2003. 1 Given this low compliance rate, and the limited number of patients with myopathy compared with patients with debility in the UDSMR database for this period, it appears that misclassification was relatively uncommon. Additional studies might investigate this issue. In addition, the outcomes of rehabilitation patients with diagnoses similar to debility (ie, muscular wasting/disuse atrophy [ICD-9-CM code 728.2], muscle weakness [ICD-9-CM code 728.87], asthenia [ICD-9-CM code 780.79]) should be evaluated. The results of this current study indicate that the physicians admitting patients with debility during 2002 or 2003 were, in general, able to identify those patients likely to benefit from inpatient rehabilitation despite any specific objective criteria for this diagnosis. However, relatively objective diagnostic criteria will most certainly be required if debility (or deconditioning, or generalized weakness) is to be considered a distinct medical diagnostic group for consideration as a qualifying condition for the CMS 75% rule. The diagnostic criteria of Raj 6 may be a reasonable starting point. The development of criterion or a schema to identify patients with debility who derive

938 DEBILITY, HIP FRACTURE, AND MYOPATHY REHABILITATION, Kortebein greater benefit from the intensive rehabilitation of an IRF as opposed to a less intensive rehabilitation setting such as that provided in a skilled nursing facility would also be beneficial. As previously noted, further study is needed to identify risk factors for acute hospital readmission 6 and/or methods of preventing medical complications in older patients with debility participating in inpatient rehabilitation. Finally, as with all rehabilitation populations, it will be important to examine the long-term effects of inpatient rehabilitation on function and survival as has been done in the geriatric literature with evaluating postacute geriatric rehabilitation. 18,19 CONCLUSIONS From a clinical perspective, older patients with debility admitted for inpatient rehabilitation during 2002 or 2003 had similar rates of functional recovery, and were discharged home just as frequently, as older patients undergoing inpatient rehabilitation for hip fracture and myopathy. Transfers to the acute hospital setting for all 3 patient groups were fairly common, although comparable to other more traditional inpatient rehabilitation populations, and mortality rates were low for all 3 patient populations. Because the functional recovery and frequency of discharge to home of the patients with debility are comparable to those of patients with hip fracture and myopathy, as well as other more traditional rehabilitation diagnoses, it seems reasonable to consider debility as a qualifying medical condition under the CMS 75% rule. However, additional research is needed to develop relatively discrete diagnostic criteria for debility, and to identify those patients with debility most likely to benefit from inpatient rehabilitation. APPENDIX 1: CENTERS FOR MEDICARE AND MEDICAID SERVICES QUALIFYING MEDICAL CONDITIONS, 2004 (ABBREVIATED) 1. Stroke 2. Spinal cord injury 3. Congenital deformity 4. Amputation 5. Major multiple trauma 6. Femur fracture (hip fracture) 7. Brain injury 8. Neurologic disorders (including multiple sclerosis, muscular dystrophy, myopathy, Parkinson disease) 9. Burns 10. Active polyarticular rheumatoid/psoriatic arthritis and seronegative arthritides, with qualifiers 11. Systemic vasculitides with joint inflammation, with qualifiers 12. Severe or advanced osteoarthritis involving 2 or more major weight-bearing joints, with qualifiers 13. Hip or knee joint replacement, or both, with qualifiers Data from: CMS Manual System. 20 References 1. Medicare. More specific criteria needed to classify inpatient rehabilitation. Report to the Senate Committee on Finance and the House Committee on Ways and Means: Publ. no. GAO-05-366. Washington (DC): Government Accountability Office; April 2005. 2. Medicare Program. Changes to the criteria for being classified as an inpatient rehabilitation facility; final rule. Washington (DC): Dept of Health and Human Services, Centers for Medicare and Medicaid Services; May 2004. Publ No. 42 CFR Part 412; 2004. p 25752-76. 3. Public Law 110-173; Medicare, Medicaid and SCHIP Extension Act of 2007. Washington (DC): Dept of Health and Human Services; December 29, 2007. 4. American Academy of Physical Medicine and Rehabilitation. Standards for assessing medical appropriateness criteria for admitting patients to rehabilitation hospitals or units. 2006. Available at: http://www.aapmr.org/zdocs/hpl/mirc0906.pdf. Accessed February 17, 2009. 5. Kortebein P. Rehabilitation for hospital associated deconditioning. Am J Phys Med Rehabil 2009;88:66-77. 6. Raj G, Munir J, Ball L, Carr DB. An inpatient rehabilitation service for deconditioned older adults. Top Geri Rehab 2007;23: 126-37. 7. Kortebein P, Bopp MM, Granger CV, Sullivan DH. Outcomes of inpatient rehabilitation for older adults with debility. Am J Phys Med Rehabil 2008;87:118-25. 8. Ottenbacher KJ, Smith PM, Illig SB, Linn RT, Ostir GV, Granger CV. Trends in length of stay, living setting, functional outcome, and mortality following medical rehabilitation. JAMA 2004;292: 1687-95. 9. Carter GM, Relles DA, Buchanan JL, et al. Classification system for inpatient rehabilitation patients: a review and proposed revisions to the Functional Independence Measure function related groups. Washington (DC): US Department of Commerce, National Technical Information Services; 1997. 10. Ottenbacher KJ, Hsu Y, Granger CV, Fiedler RC. The reliability of the functional independence measure: a quantitative review. Arch Phys Med Rehabil 1996;77:1226-32. 11. Granger CV, Kelley-Hayes M, Johnston M, Deutsch A, Braun S, Fiedler RC. Quality and outcome measures for medical rehabilitation. In: Braddom RL, editor. Physical medicine and rehabilitation. Philadelphia: W. B. Saunders Co; 2000. p 151-64. 12. Jain NB, Al-Adawi S, Dorvlo AS, Burke DT. Association between body mass index and functional independence measure in patients with deconditioning. Am J Phys Med Rehabil 2008;87:21-5. 13. Graham JE, Chang PJ, Berges IM, Granger CV, Ottenbacher KJ. Race/ethnicity and outcomes following inpatient rehabilitation for hip fracture. J Gerontol Med Sci 2008;63A:860-6. 14. Lieberman D, Friger M, Lieberman D. Inpatient rehabilitation outcome after hip fracture surgery in elderly patients: a prospective cohort study of 946 patients. Arch Phys Med Rehabil 2006; 87:167-71. 15. Hershkovitz A, Kalandariov Z, Hermush V, Weiss R, Brill S. Factors effecting short-term rehabilitation outcomes of disabled elderly patients with proximal hip fracture. Arch Phys Med Rehabil 2007;88:916-21. 16. Ottenbacher KJ, Smith PM, Illig SB, Fiedler RC, Granger CV. Length of stay and hospital readmission for persons with disabilities. Am J Public Health 2000;90:1920-3. 17. Ottenbacher KJ, Smith PM, Illig SB, Peek MK, Fiedler RC, Granger CV. Hospital readmission of persons with hip fracture following medical rehabilitation. Arch Gerontol Geriatr 2003;36:15-22. 18. Applegate WB, Miller ST, Graney MJ, Elam JT, Burns R, Akins DE. A randomized, controlled trial of a geriatric assessment unit in a community rehabilitation hospital. N Engl J Med 1990;322:1572-8. 19. Cohen HJ, Feussner JR, Weinberger M, et al. A controlled trial of inpatient and outpatient geriatric evaluation and management. N Engl J Med 2002;346:905-12. 20. CMS Manual System. Pub. 100-04, Medicare Claims Processing, Transmittal 347, October 29, 2004. Available at: www.cms.hhs. gov/inpatientrehabfacpps/downloads/r347cp.pdf. Accessed April 16, 2009. Supplier a. SAS for Windows 9.1.2003; SAS Institute, 100 SAS Campus Dr, Cary, NC 27513.