Lindenberg, Kirstin BPhty; Nitz, Jennifer C. PhD, MPhty, BPhty, FACP; Rahmann, Ann PhD, BPhty, Grad Cert Physio; Bew, Paul MPhty, BPhty, FACP
With an increasingly aging population, the need for comprehensive, cost-effective geriatric rehabilitation is paramount. In light of this, research into potential markers of functional decline and optimal rehabilitation strategies has been conducted with mixed results.1–3 A key goal of these rehabilitation strategies is to return the older adults to their prior living conditions and maintaining independence and quality of life,4 thus preventing early transfer to an aged care facility. Variables decreasing the likelihood of this goal include increasing age, hospital length of stay (LOS) and comorbidities, lower functional status before admission, and decreased mobility while hospitalized.2–5 When looking exclusively at survivors of stroke, the presence of a live-in partner, independent sitting balance, rolling ability, gait, and higher cognitive function were predictors of return to home.6,7
Few studies have has investigated rehabilitation outcomes between a number of diagnostic categories,8–10 despite orthopedic replacement and fracture, stroke, and reconditioning accounting for the highest percentages of admission to rehabilitation units in Australia.11 Of those that have, only a single outcome is measured, potentially excluding information important when deciding discharge destination.12 Many studies either exclude certain diagnostic categories13 or do not compare outcomes between categories.14
Elphick et al15 noted that primary diagnosis did not predict discharge destination, LOS, or hospital readmission. Denkinger et al16 also found diagnosis to have minimal impact on function at discharge (according to the Barthel Index and Falls Related Self Efficacy Scale) or 4-month follow-up. A primary diagnosis of debility, however, has been associated with a decreased likelihood of discharge home.9 Several studies have determined the predictive ability of admission function and discharge destination following rehabilitation in stroke populations.7,17,18 Admission Barthel Index has also been shown to be predictive of discharge home; however, this was not given according to diagnosis.15 From this, further research investigating the relationship between diagnostic category, function, and discharge destination is warranted.
The 10-Meter Walk Test (10MWT)19 and Balance Outcome Measure for Elder Rehabilitation (BOOMER)20 can be used to identify changes in mobility and balance in the geriatric population. The BOOMER demonstrates validity when compared with the Berg Balance Scale, gait speed, Modified Elderly Mobility Scale, and the Functional Independence Measure.20,21 No previous study, however, has used this suite of outcome measures to quantify the response to individually tailored rehabilitation within and between diagnostic categories. Thus, the aim of this retrospective audit was to determine whether diagnostic category matters or whether it is rehabilitation LOS, ability on the 10MWT, or BOOMER at discharge that predicts discharge destination in elderly patients undergoing rehabilitation who previously lived at home.
The study was a retrospective audit of patient files.
Patients consecutively admitted to a 50-bed inpatient geriatric rehabilitation unit for physiotherapy rehabilitation in South East Queensland, Australia, between June 2010 and March 2012 were considered for this study. Those with a primary diagnosis of orthopedic conditions, debility (participants referred to rehabilitation for general reconditioning targeting balance, mobility, and strength), stroke, and other neurological conditions (conditions other than stroke; eg, multiple sclerosis and Parkinson disease) according the Australasian Rehabilitation Outcomes Centre11 were included. Patients were excluded if their diagnosis was assessment only; congenital deformity; major multiple trauma; cardiac, respiratory, and pain syndromes; amputation (without prosthesis); burns; or developmental disability, or were younger than 60 years (Figure 1). Ethical approval was obtained through university and hospital human ethics research committees.
All participants underwent usual care physiotherapy consisting of a program individually tailored to target relevant deficits in body, structure, and function. These programs contained exercises such as sit-to-stand practice, reaching outside base of support, and stepping in all directions that incorporates elements of balance, strengthening, and functional exercise. Basic gait training, as well as more challenging gait tasks that included surface manipulation and multitasking while walking, was also added as indicated. The goal of intervention was to enable the patient to achieve maximal independence where possible commensurate with discharge to home.
Discharge destination was recorded as either home or (long-term) residentialaged care facility (RACF). Additional demographic details recorded included sex, age, diagnosis, comorbidities, any adverse events during rehabilitation, time from hospital admission to rehabilitation admission, rehabilitation LOS, and total LOS.
Performance on the 10MWT and BOOMER was recorded by the treating physiotherapist upon admission and discharge. The 10MWT involves recording the time taken for the participant to walk along a 10-m marked track, allowing an estimation of gait speed to be made. Two meters at either end of the track are allowed for acceleration and deceleration. This assessment is also quick, functional, and has established validity when used for a number of patient groups.10,19,22 The BOOMER comprises 4 commonly used objective measures, including the Timed Up and Go, Functional Reach, Step Test (ST), and static stance feet together eyes closed. Performance in each of these tests is converted to a 5-point ordinal scale (0-4), with a higher score representing better balance ability. The system of scoring on the BOOMER is presented in Appendix. The BOOMER is quick to administer, has demonstrated good internal consistency, and is sensitive to change.20 It has established validity when compared with the motor component of the Functional Independence Measure, Modified Elderly Mobility Scale, Berg Balance Scale, and gait speed.21 All participating physiotherapists had received training in the use of these tests.
APPENDIX BOOMER Scor...Image Tools
Participant characteristics, including discharge destination and performance on the 10MWT and BOOMER at admission and discharge, are presented as descriptive statistics. Also presented are the percentages of participants able to undertake each measure at admission and those able at discharge.
Pearson χ2 was undertaken to determine any difference in discharge destination between diagnostic categories. Correlation between these variables and discharge to RACF was investigated first using the Spearman correlation coefficient, with those variables with a Spearman r more than 0.3 included in the logistic regression model, depending on collinearity.
Preliminary investigations of collinearity were performed to determine the variables to be excluded from the regression. Collinearity was observed between the total BOOMER score and the component scores, which would be expected since the components determine the overall score. On the basis of the Spearman correlation coefficient, either total BOOMER score or the individual components could be included in the regression model. The dependent variable in the logistic regression was discharge destination. Logistic regression was undertaken to determine whether rehabilitation LOS, 10MWT, or total BOOMER score at discharge was able to predict discharge destination. Total BOOMER score was dichotomized to 4 or less and more than 4, as a score of 4 or less described patients who were barely mobile.
Statistical analysis was performed using STATA 10 program (StataCorp, College Station, Texas), with significance level set at P < .05.
A total of 293 patients were considered for inclusion in this study. Of them, 41 patients did not meet inclusion criteria and 4 declined to participate. Discharge data were unavailable for a further 21 patients. Overall, discharge data were collected from 227 participants. Of these participants, 43 (19%) had a diagnosis of stroke, 20 (9%) had other neurological conditions, 129 (57%) had orthopedic conditions, and 35 (15%) had debility. Age of participants ranged from 65 to 96 years, with a mean (SD) of 79 (9) years. Seventy-seven (34%) participants were men. On average, participants had 5 (2) comorbidities, irrespective of diagnostic category (Table 1). Of patients discharged from rehabilitation, 12.3% went to residential aged care, with the remainder returning home.
10MWT and BOOMER
Mean scores on the 10MWT, individual components of the BOOMER, and total BOOMER score are presented in Table 2, for admission and discharge, as well as LOS in rehabilitation. The percentage of participants able to complete these measures at admission and discharge is also shown for each diagnostic category. On admission, 84% of individuals with stroke, 80% of individuals with neurological conditions, 77% of individuals with orthopedic conditions, and 83% of individuals with debility could complete some components of the BOOMER. The step test, in particular, shows poor performance across all diagnostic categories, with only 51% of those with stroke, 55% with neurological conditions, 32% with orthopedic conditions, and 57% with debility able to perform the test at admission. Only 77% of individuals with stroke, 80% with neurological conditions, 77% with orthopedic conditions, and 83% with debility were able to perform the 10MWT. Thus, discharge data were chosen for statistical analysis, as including admission measures would likely have resulted in overestimating participants' change in performance.
An increase in the percentage of participants able to complete the measures from admission to discharge was observed in each test and for each diagnostic group, apart from static stance eyes closed in the group with debility (91%).
In regard to the whole sample, 199 (87.7%) participants were discharged home. Overall, diagnostic category (stroke, neurological conditions, orthopedic conditions, and debility) was not significantly associated with discharge destination (Pearson χ2 = 1.26, P = .74).
A total of 28 participants were discharged to RACF. As displayed in Table 3, those participants unable to perform the 10MWT at discharge went to RACF more frequently than those able to perform. Participants with a discharge BOOMER score of less than 4 and unable to perform the 10MWT also went to RACF more frequently.
The Spearman rank correlation showed individual discharge components of the BOOMER to be significantly correlated with discharge to RACF (P < .05). Correlation coefficients (r) ranged from −0.47 to 0.37 and are presented in Table 4. A discharge total BOOMER score of less than 4 (r = −0.47) and an inability to perform the 10MWT (r = 0.34) were most significantly correlated with discharge to RACF and so were included in the logistic regression along with rehabilitation LOS (r = 0.37). Since no significant difference was found between diagnostic categories, this model was investigated for the whole sample.
This model was able to explain approximately 60.7% (P = .002) of the variability in the study sample. Sensitivity of 71.4% (discharge to RACF) and specificity of 93.3% (discharge home) was observed with an overall predictive accuracy of 86.4%. Receiver operating characteristic curves are presented in Figure 2, displaying an area under the curve calculation of 97.1%.
The use of discharge data and inclusion of multiple diagnostic categories in this study provides important information for clinicians regarding the level of function required for adequate community mobility and balance following rehabilitation in a geriatric population. Patients from different diagnostic categories did not respond differently to rehabilitation. Considering the number of comorbidities was similar across diagnostic categories, it would appear unlikely for their presence to confound outcome, although this aspect was not explored. The results of the logistic regression model show that the variables rehabilitation LOS, an inability to perform the 10MWT at discharge, and discharge BOOMER score of less than 4 can predict discharge destination with 86.4% accuracy (P = .002). This indicates that those who are unable to independently mobilize, have poor balance, and are at risk for falls risk at discharge are more likely to be discharged to RACF.
Our results concurred with Sherrington et al,23 who predicted mobility-related disability following rehabilitation using 15 variables (area under curve, 0.8). Disability referred to inability to climb a flight of stairs or walk 800 m unassisted. Inability to perform a ST at discharge, longer-duration Timed Up and Go, and standing balance at discharge were associated with mobility-related disability after rehabilitation, supporting the predictors included in the model in this study. Place of residence was not identified in their study; however, the variables included in the model allude to the level of function following rehabilitation required for independence. This model was also applied to a number of diagnostic categories, including neurological conditions, musculoskeletal conditions, falls, and decreased mobility.
Using rehabilitation LOS, inability to perform the 10MWT, and BOOMER score of less than 4 at discharge in this model appears to be consistent with predictors of discharge destination when looking exclusively at stroke. This suggests that some predictors of discharge destination apply to multiple diagnostic categories. This is supported by the study of Wee et al,17 where those patients with stroke discharged to care facilities had lower mean Berg Balance Scale scores (that also measures balance and mobility) at both admission and discharge compared with those discharged home. Longer rehabilitation LOS was associated with a higher likelihood of discharge to RACF in a previous study of survivors of stroke,24 which was also observed in this study. The predictive accuracy of the model in this study is comparable to Brauer et al7 (87%), who used the Motor Assessment Scale to predict discharge destination in a sample of patients with stroke.
Including the presence of home support in the collected data may have strengthened predictive accuracy of this model as it has been determined a strong predictor of discharge destination in previous studies.17,18 Wee et al17 found the odds of returning home were increased when caregiver support was present. This may account for our participants who had a BOOMER score less than 4, were unable to perform the 10MWT or both but returned home. Similar to previous studies,15,16 this study did not find diagnostic category to be correlated with discharge to RACF.
The large number of participants unable to perform the 10MWT or components of the BOOMER at admission is surprising, since both measures have been validated for use in a geriatric population.19,21 It is possible that there were additional factors such as reliance on a mobility aid, preventing performance, and warrants further investigation.
The appropriateness of the ST for an orthopedic diagnosis could be raised, with only 32% and 75% able to perform this test at admission and discharge, respectively. Considering that the average total LOS for this population was 58 days, it is possible that participants still had weight-bearing limitations, preventing their ability to perform the test at both admission and discharge. Difficulty with this test item is supported by Sherrington et al,23 who noted that 36% of participants were unable to perform a 7-cm ST at discharge. Further investigation of the utility of this measure in orthopedic populations may provide more insight into these findings.
We acknowledge several limitations to this study. First, data were collected from a single rehabilitation facility that services 4 short-term care hospitals, so findings may not be generalized to the whole elderly population undergoing rehabilitation. Since consecutive admissions were included, the risk of selection bias should have been reduced, however.
Although using admission scores as predictors of discharge destination provides direction for rehabilitation, using discharge data in this study provides new information on the final functional level associated with discharge to RACF. Discharge function has been shown to be a predictor of postrehabilitation mobility-related disability,23 suggesting that follow-up would also be warranted in this population.
The small number of patients in each diagnostic category discharged to RACF (3 of the groups had 5 or less cell size) might have impacted the between-group comparison for discharge destination. Therefore, the finding that there was no difference in discharge destination among the diagnostic groups needs to be studied further.
The impact of short-term LOS on overall functional outcome is an unknown in this study. It is not clear whether short-term LOS was due to participants being medically unstable during this period, were undergoing early rehabilitation, or awaiting bed space at the rehabilitation facility.
Finally, the inclusion of participant demographics such as cognitive status, the presence of home supports, and mobility status before admission may have explained more of the variability seen in the sample.
Using the variables rehabilitation LOS, inability to perform the 10MWT, and BOOMER score of less than 4 at discharge, discharge destination can be predicted with good accuracy. These outcomes assist with determining the level of function required at discharge to return to home living across a number of diagnoses, however, presence of home support needs to be considered as well. In accordance with previous studies, longer duration in rehabilitation and increasing age were associated with a higher likelihood of discharge to RACF. It would also appear that a standardized suite of measures of functional ability may not be appropriate for all diagnostic categories undergoing rehabilitation. It would seem that, just as intervention needs to be tailored to the individual patient, the measure of their progress also should be unique.
1. Alarcón T, Bárcena A, González-Montalvo JI, Penãlosa C, Salgado A. Factors predictive of outcome on admission to an acute geriatric ward. Age Ageing. 1999;28(5):429–432.
2. Hager K, Neumann U. Rehabilitation of the elderly—influence of age, sex, main diagnosis and activities of daily living (ADL) on the elderly patients' return to their previous living conditions. Arch Gerontol Geriatr. 1997;25(1):131–139.
3. McCloskey R. Functional and self-efficacy changes of patients admitted to a geriatric rehabilitation unit. J Adv Nurs. 2004;46(2):186–193.
4. Grill E, Hermes R, Swoboda W, Uzarewicz C, Kostanjsek N, Stucki G. ICF core set for geriatric patients in early post-acute rehabilitation facilities. Disabil Rehabil. 2005;27(7/8):411–417.
5. Zisberg A, Shadmi E, Sinoff G, Gur-Yaish N, Srulovici E, Admi H. Low mobility during hospitalization and functional decline in older adults. J Am Geriatr Soc. 2011;59(2):266–273.
6. Frank M, Conzelmann M, Engelter S. Prediction of discharge destination after neurological rehabilitation in stroke patients. Euro Neurol. 2010;63(4):227–233.
7. Brauer SG, Bew PG, Kuys SS, Lynch MR, Morrison G. Prediction of discharge destination after stroke using motor assessment scale on admission: a prospective multisite study. Arch Phys Med Rehabil. 2008;89(6):1061–1065.
8. Hasselgren L, Olsson LL, Nyberg L. Is leg muscle strength correlated with functional balance and mobility among inpatients in geriatric rehabilitation? Arch Gerontol Geriatr. 2011;52(3):e220–e225.
9. Kortbein P, Bopp MM, Granger CV, Sullivan DH. Outcomes of inpatient rehabilitation for older adults with debility. Am J Phys Med Rehabil. 2008;87(2):118–125.
10. Tilson JK, Sullivan KJ, Cen SY, et al. Meaningful gait speed improvement during the first 60 days poststroke: minimal clinically important difference. Phys Ther. 2010;90(2):196–208.
11. University of Wooloongong. The AROC Annual Report: the state of rehabilitation in Australia in 2011. http://www.ahsri.uow.edu.au/aroc
. Published 2011. Accessed February 19, 2013.
12. Morrison G, Lee H-L, Kuys SS, Clarke J, Bew P, Haines TP. Changes in falls risk factors for geriatric diagnostic groups across inpatient, outpatient and domiciliary rehabilitation settings. Disabil Rehabil. 2011;33(11):900–907.
13. Hauer K, Rost B, Rütschle K, et al. Exercise training for rehabilitation and secondary prevention of falls in geriatric patients with a history of injurious falls. J Am Geriatr Soc. 2001;49(1):10–20.
14. Potter JM, Evans AL, Duncan G. Gait speed and activities of daily living function in geriatric patients. Arch Phys Med Rehabil. 1995;76(11):997–999.
15. Elphick HL, Mankad K, Madan S, Parker C, Liddle BJ. The determinants of successful in-hospital rehabilitation in people aged 90 years and older. Gerontology. 2007;53(2):116–120.
16. Denkinger MD, Igl W, Jamour M, et al. Does functional change predict the course of improvement in geriatric inpatient rehabilitation? Clin Rehabil. 2010;24(5):463–470.
17. Wee Jy, Wong H, Palepu A. Validation of the Berg Balance Scale as a predictor of length of stay and discharge destination in stroke rehabilitation. Arch Phys Med Rehabil. 2003;84(5):731–735.
18. Agarwal V, McRae MP, Bhardwaj A, Teasell RW. A model to aid in the prediction of discharge location for stroke rehabilitation patients. Arch Phys Med Rehabil. 2003;84(11):1703–1709.
19. Peters DMF, Stacy L, Krotish DE. Assessing the reliability and validity of a shorter walk test compared with the 10 Meter Walk Test for measurement of gait speed in healthy, older adults. J Geriatr Phys Ther. 2013;36(1):24–30.
20. Haines T, Kuys SS, Morrison G, Clarke J, Bew P, McPhail S. Development and validation of the balance outcome measurement for elder rehabilitation. Arch Phys Med Rehabil. 2007;88(12):1614–1621.
21. Kuys SS, Morrison G, Bew PG, Clarke J, Haines TP. Further validation of the Balance Outcome Measure for Elder Rehabilitation. Arch Phys Med Rehabil. 2011;92(1):101–105.
22. Perera S, Mody SH, Woodman RC, Studenski SA. Meaningful change and responsiveness in common physical performance measures in older adults. J Am Geriatr Soc. 2006;54(5):743–749.
23. Sherrington C, Lord SR, Close JCT, et al. Mobility-related disability three months after aged care rehabilitation can be predicted with a simple tool: an observational study. J Physiother. 2010;56(2):121–127.
24. Elwood D, Rashbaum I, Bonder J, et al. Length of stay in rehabilitation is associated with admission neurological deficit and discharge destination. PM R. 2009;1(2):147–151.
discharge destination; older adults; outcomes; rehabilitation