The International Journal of Psychosocial Rehabilitation

Predictors for Functional Outcome in Older Rehabilitation Patients

Susan E. MacNeill
Rehabilitation Institute of Michigan

Peter A. Lichtenberg
Rehabilitation Institute of Michigan

Citation: Rehabilitation Psychology, November 1998 Vol. 43, No. 3, 246-257

Medical rehabilitation may serve to distinguish older adults capable of independent living from those in need of assistance. Return to independent living was evaluated in 372 older patients, all of whom lived alone before admission. For persons discharged alone, admission performance on the Functional Independence Measure was the only significant predictor of discharge self-care. For those discharged with supervision, age, memory, and depression were also significant predictors of discharge self-care. Thus, older live-alone patients may progress at different rates over rehabilitation, and psychosocial problems may foreshadow loss of independence.

Correspondence may be addressed to Susan E. MacNeill, Department of Psychology, 261 Mack Boulevard, Detroit, Michigan, 48201.

While census data indicate that the number of single-person households is increasing, nowhere is this more evident than in the elderly population, with increasing numbers of elderly, widowed people living entirely alone. In many respects, this group represents the ideal of successful aging. This elderly "live-alone" group makes up a population characterized by less cognitive impairment and fewer medical complications ( MacNeill & Lichtenberg, 1997 ; Mui & Burnette, 1994 ). Thus, most live-alone older adults function at a very high level. A subgroup of live-alone older adults, however, may be at the greatest risk for mortality and morbidity. This group of "at-risk" elderly people was highlighted in a study evaluating calls for emergency medical services ( Gurley, Lum, Sande, Bernard, & Katz, 1996 ). In reviewing 387 consecutive calls for emergency medical assistance, the mean age of recipients was 73 years, and older people living entirely alone were the most likely to be found helpless or dead in the home. This at-risk subset of older adults has also been identified as having increased needs for long-term care and greater risk for institutionalization ( Greene & Ondrich, 1990 ; Rockwood, Stolee, & McDowell, 1996 ). Thus, not all live-alone older adults are functioning at the same high level, and a subset of these people may be at risk for greater decline following medical events. This subset of older adults may have greater difficulty recovering from medical events and may progress through rehabilitation at a slower rate. They may also require additional assistance following discharge.

It is generally agreed on that geriatric patients benefit from structured rehabilitation ( Indredavik et al., 1991 ; Jongbloed, 1986 ; Kaste, Palomaki, & Sarna, 1994 ). There is limited appreciation for the diversity of this population, and poor understanding of the process whereby older people improve over the course of rehabilitation ( Kane, Chen, Blewett, & Sangl, 1996 ; Rusin, 1990 ). Age, premorbid motor functioning, cognition, and motivation have all been identified as factors predicting outcome for rehabilitation ( Henley, Petit, Todd-Prokeropek, & Tupper, 1985 ; Wade, Skilbeck, Wood, & Langton-Hewer, 1984 ; Coletta & Murphy, 1993 ; Dickerson & Fisher, 1993 ; Friedman, 1995 , Smith & Clark, 1995 ). Many chronic medical illnesses commonly seen with older adults have also been identified as predictive of worse functional outcome, longer lengths of stay, and greater disability following rehabilitation ( Campbell et al., 1994 ; Fox, Hawkes, Magaziner, Zimmerman, & Hebel, 1996 ; Fried & Guralnik, 1997 ; Nelson, Herndon, Mark, Pryor, & Califf, 1991 ; Rozzini, Frisoni, Ferrucci, & Barbisoni, 1997 ; Sullivan et al., 1996 ).

Thus, many factors may affect progress in rehabilitation for older adults, but these findings may be confounded by the diversity of people over 60 years of age. In fact, discharge disposition may serve to highlight subgroups of patients, separating those who were functioning at a high level prior to admission and for whom illness does not result in a permanent decline from those for whom needs for assistance were increasing prior to admission. In a prior study designed to better understand live-alone older adults, we examined predictors of discharge home alone following medical rehabilitation ( MacNeill & Lichtenberg, 1997 ). Predictors of return to living alone, including demographic variables, chronic medical illness, Functional Independence Measure performance, and cognition, were examined in a sample of 372 older medical rehabilitation patients. All patients were living entirely alone prior to admission. Using logistic regression, we identified cognition as the single most important predictor of return to living entirely alone, above and beyond physical self-care abilities.

These findings provide potential support for the premise of at least two distinct groups within the live-alone subset of older adults. In the current study, we examine factors that contribute to functional recovery over the course of rehabilitation in each of these two groups, including demographic characteristics, chronic medical illness, memory, and depression. It was hypothesized that live-alone older adults unable to return home alone would have more difficulties over the course of rehabilitation, including fewer self-care gains, increased cognitive deficits, and increased report of depression.



Participants included 372 consecutively admitted geriatric medical rehabilitation patients. This sample has been previously described in detail ( MacNeill & Lichtenberg, 1997 ). Table 1 provides general demographic information. Patients were included if they were 60 years of age or older and living entirely alone prior to admission. Admitting diagnoses included peripheral involvement (53%; arthritis, peripheral vascular disease), fracture (33%; pelvic, hip, or leg), and stroke (13%; primarily subcortical).
Measures Functional Independence Measure.

The Functional Independence Measure (FIM) was designed to assess the ability to carry out activities of daily living, including dressing, toileting, ambulation, and grooming ( Hamilton, Granger, Sherwin, Zielezny, & Tashman, 1987 ). Each of 18 self-care tasks are rated on a 7-point scale ranging from 1 ( completely dependent ) to 7 ( completely independent ). The FIM has been shown to have high interrater reliability, with an intraclass correlation coefficient of .97 for the total score ( Hamilton, Laughlin, Granger, & Kayton, 1991 ). Validity has also been demonstrated with brain-injured patients and stroke patients ( Cook, Smith, & Truman, 1994 ; Granger, Cotter, Hamilton, & Fielder, 1993 ). Rasch analysis of the FIM has provided support for two subscales measuring different aspects of disability ( Granger, Hamilton, Linacre, Heineman, & Wright, 1993 ; Linacre, Heineman, Wright, Granger, & Hamilton, 1994 ). Thirteen self-care and locomotion items measure motor disability, and five communication and cognition items measure cognitive disability.
Comorbid Medical Illness.

To evaluate impact of chronic comorbid medical illnesses (CMI) on independent functioning, medical diagnoses were coded using the Index of Comorbidity ( Charlson, Pompei, Ales, & MacKenzie, 1987 ). This index was developed on a sample of 559 medical patients, followed over one year. Based on mortality rates during that period, an index was developed that measured not only the number of comorbid diseases, but also the seriousness of these diseases. Those diseases significantly associated with mortality were identified and weights equivalent to adjusted relative risks were assigned to them. Total score on the index was calculated by coding each coexisting disease according to these weights and summing the weights for each patient; high scores reflect a high number or degree of seriousness of coexisting diseases. Charlson et al. (1987) validated the index on a cohort of 685 medical patients and found scores to be significant predictors of one-year survival, accounting for a greater proportion of the variance associated with death due to comorbid conditions than did the simple measure of number of coexisting conditions. The CMI is the only published method of controlling for comorbidity factors that was developed empirically. The relationship of the CMI to disability was demonstrated in a double cross-validation study in which CMI was the best predictor of activities of daily living recovery in medical rehabilitation ( Moore & Lichtenberg, 1996 ).
Wechsler Memory Scale–Revised

( Logical Memory Delayed Recall ).

The Logical Memory subtest of the Wechsler Memory Scale–Revised ( Weschler, 1987 ) involves verbal recall of two paragraph-length narratives (Memory 1 scores) with delayed recall 30 minutes following initial presentation (Memory 2 scores). Logical Memory has demonstrated usefulness in differentiating cognitively intact and impaired patients ( Christensen, Hadzi-Pavlovic, & Jacomb, 1991 ). Normative data are available for geriatric groups up to age 97 ( Cullum, Butters, Troster, & Salmon, 1990 ; Ivnik et al., 1992 ), and have more recently been provided for elderly African Americans ( Lichtenberg, Manning, Vangel, & Ross, 1995 ). Logical Memory 1 and 2 are highly correlated. Thus we decided to use the Logical Memory 2 score in this study because it is a measure of delayed recall, and patients with dementia are likely to show significant forgetting.
Geriatric Depression Scale.

The Geriatric Depression Scale (GDS) was created specifically to detect depression in elderly people ( Brink et al., 1992 ); Yesavage et al., 1983 ). The GDS consists of 30 yes—no questions, and support has been demonstrated for reliability and validity when used with elderly individuals ( Brink et al., 1982 ; Rapp, Parisi, Walsh, & Wallace, 1988 ). Research also indicates the use of the GDS in detecting depression in medically ill elderly individuals ( Rapp, Parisi, & Walsh, 1988 ).

The Rehabilitation Institute of Michigan is a free-standing rehabilitation hospital. As part of standard protocol, all patients underwent assessment of self-care skills using the FIM within 3 days of admission and again on discharge. All FIM assessments were completed by therapists certified through Uniform Data Systems training protocol. All patients also underwent a general cognitive evaluation including administration of Logical Memory and the GDS during the first week of admission. For each patient, a multidisciplinary team met weekly to evaluate progress and set discharge goals. Discharge status was dichotomized as either returning home alone or not alone (i.e., to nursing home, with family).
Statistical Analyses

Pearson product moment correlations were used to analyze continuous variables, and point biserial correlations were used for categorical variables. Independent samples t tests were conducted for continuous variables. Multiple regression analyses were completed separately for the two discharge groups: home alone and not alone. Forced entry multiple regression analysis was used to examine contributions of demographic variables, CMI, cognition, and depression to discharge FIM scores, above and beyond admission FIM performance. Thus, admission FIM scores were entered into the equation first. This was followed by a block of demographic variables, the CMI, delayed recall, and finally, the GDS. Demographic and medical variables were entered into the regression equation before psychological variables in order to give a more stringent test of the psychological variables in predicting functional outcome. FIM Motor scores and FIM Social Cognition composite scores were examined individually.


Participants ranged in age from 60 to 99, with a mean age of 78.2 ( SD = 7.94). Overall, mean education was 9.9 years ( SD = 3.44). Seventy-seven percent were women, 59% were African American, and 41% were White. Of the 372 participants, 39% ( n = 146) were discharged entirely alone, whereas 61% ( n = 226) were discharged with supervision (see Table 1 ). Average length of stay was 19 days ( SD = 7.2). Among all patients, a mean gain of 15.3 points ( SD = 8.6) was seen on FIM Motor items over length of stay. Patients showed a mean gain of 1.2 points ( SD = 3.1) on FIM Social Cognition items.

Several significant differences were found between patients discharged home alone and not alone. Patients discharged home alone were younger, t (370) = - 3.67, p < .0001, and scored higher on both the Social Cognition and Motor components of the FIM. They also scored higher on the measure of delayed recall, t (305) = 4.35 p < .0001), and lower on the GDS, t = —2.22, p < .03 (See Tables 1 and 2 ). No significant differences were seen between groups with respect to the CMI or admitting diagnoses.

Correlational analyses revealed several important relationships between independent variables and FIM scores. Higher admission FIM scores were related to younger age, shorter lengths of stay, better delayed recall, and lower scores on the Depression scale. A similar pattern was noted for discharge FIM scores. Increased chronic medical illnesses were related to lower admission Motor performance ( Table 3 ). Given that modest correlations were found among predictor variables, the risk of collinearity is minimal.

Regression analyses revealed that admission functional status accounted for the greatest variance in prediction of discharge FIM performance, both for those discharged home alone and those not discharged home alone. For patients discharged home alone, admission FIM status was the only significant predictor of discharge functional status, for both motor and social cognition performance (see Tables 4 and 5)

For patients not discharged home alone, discharge FIM motor performance was predicted by age, memory, and depression, beyond initial FIM scores ( Table 6 ). Memory was also a significant and unique predictor of discharge FIM social cognition performance ( Table 7 ). Regression analyses thus indicated that decreased motor function was related to increased report of depressive symptoms ( r = .29, p < .0001), older age ( r = .29, p < .0001), and lower memory scores ( r = 0.25, p < .0001).



Variables predictive of recovery of functional self-care abilities were examined in 372 older medical rehabilitation patients, all of whom were previously living entirely alone prior to admission. For patients discharged home alone after rehabilitation, admission functional status was the only significant predictor of discharge status. For those not discharged home alone, several factors contributed to prediction of discharge functional status, including memory, depression, and older age. These results suggest depression and cognitive deficits place older medical rehabilitation patients at risk for loss of independence. Two potential explanations of these findings are (a) psychosocial adjustment and (b) markers of neurobehavioral compromise. In the psychosocial adjustment model, extreme reactions to medical illness may limit coping (e.g., adjustment and adaptation), which may in turn limit physical recovery. In the neurobehavioral compromise model, depression and memory loss are viewed as markers of cerebral dysfunction. This dysfunction is thus responsible for limitations in making adaptations necessary to resume independent living.

Findings clearly indicate that a certain subgroup of older live-alone medical rehabilitation patients will be dependent following rehabilitation. This group can be characterized as being older, with more memory problems and increased report of depressive symptoms. Overall, these patients exhibited worse performance with self-care tasks, both on admission and discharge, despite no differences between groups regarding presenting diagnosis or severity of comorbid medical illnesses. The presence of depression and memory loss can be seen as risks for loss of independence.

This study is the first to investigate the process of recovery among a sample of older, live-alone patients who were discharged to either independent or dependent settings. Recovery for those patients discharged to independent living was predicted only by initial level of functional skills. For patients discharged to a dependent setting, recovery was predicted by functional, demographic, emotional, and cognitive factors. Thus, problems in multiple domains were related to a lack of recovery and to dependency.

Limitations of this study bear mentioning. This study lacked a measure of social support. Although live-alone status may be considered a marker of independence, the extent of external support received by patients was not known. A measure of such supports may provide insight into the level of community assistance needed to maintain independent living. This study was also limited to an inner city, predominantly African American sample, and may be of limited generalizability to other samples of older adults. Finally, given the retrospective nature of this study, longitudinal follow-up was not possible, but would also be beneficial in tracking transitions in living situation after discharge.

Results presented in this study provide insight into factors that affect discharge from rehabilitation for older adults. Findings indicate that living entirely alone before admission does not automatically predict successful discharge to an independent setting. Other factors, including cognitive status and emotional status, may play an important role. As the general population of independent older adults continues to grow, continued exploration of these factors will be essential. Future investigations can build on this research by following dependent and independent older adults over time. Important questions to be explored include (a) what factors over the course of rehabilitation predict long-term independence? and (b) what individual and environmental variables determine long-term independence?


Brink, T., Yesavage, J., Lum, G., Heersema, P., Addey, M. & Rose, T. (1982). Screening tests for geriatric depression.( Clinical Gerontologist, 1, 37—41.)

Campbell, A. J., Busby, W. J., Robertson, M. C., Lum, C. L., Langlois, J. A. & Morgan, F. C. (1994). Disease impairment, disability and social handicap: A community based study of people aged 70 years and over.( Disability and Rehabilitation, 16, 72—79.)

Charlson, M. E., Pompei, P., Ales, K. L. & MacKenzie, C. R. (1987). A new method of classifying prognostic co-morbidity in longitudinal studies: Development and validation.( Journal of Chronic Disease, 40, 373—383.)

Christensen, H., Hadzi-Pavlovic, & Jacomb, P. (1991). The psychometric differentiation of dementia from normal aging: A meta-analysis.( Psychological Assessment: Journal of Consulting and Clinical Psychology, 3, 147—155.)

Coletta, E. M. & Murphy, J. B. (1993). The complications of immobility in the elderly stroke patient.( Journal of the American Board of Family Practice, 5, 389—397.)

Cook, L., Smith, D. S. & Truman, G. (1994). Using Functional Independence Measure Profiles as an index of outcome in the rehabilitation of brain-injured patients.( Archives of Physical Medicine and Rehabilitation, 75, 390—393.)

Cullum, C. M., Butters, N., Troster, A. I. & Salmon, D. P. (1990). Normal aging and forgetting rates on the Wechsler Memory Scale-Revised.( Archives of Clinical Neuropsychology, 5, 23—30.)

Dickerson, A. E. & Fisher, A. G. (1993). Age differences in functional performance.( American Journal of Occupational Therapy, 47, 686—692.)

Fox, K. M., Hawkes, W. G., Magaziner, J., Zimmerman, S. I. & Hebel, J. R. (1996). Markers of failure to thrive among older hip fracture patients.( Journal of the American Geriatric Society, 44, 371—376.)

Fried, L. P. & Guralnik, J. M. (1997). Disability in older adults: Evidence regarding significance, etiology, and risk.( Journal of the American Geriatric Society, 45, 92—100.)

Friedman, P. (1995). Stroke outcome in elderly people living alone.( Disability and Rehabilitation, 17 (2), 90—99.)

Granger, C. V., Cotter, A. C., Hamilton, B. B. & Fiedler, R. C. (1993). Functional Assessment Scales: A study of persons after stroke.( Archives of Physical Medicine and Rehabilitation, 74, 133—138.)

Granger, C. V., Hamilton, B. B., Linacre, J. M., Heinemann, A. W. & Wright, B. D. (1993). Performance profiles of the functional independence measure.( American Journal of Physical Medicine and Rehabilitation, 72, 84—89.)

Greene, V. L. & Ondrich, J. I. (1990). Risk factors for nursing home admissions and exits: A discrete-time hazard function approach.( Journal of Gerontology, 45 (6), S250—258.)

Gurley, R. J., Lum, N., Sande, M., Bernard, L. & Katz, M. H. (1996). Persons found in their homes helpless or dead.( New England Journal of Medicine, 334, 1710—1716.)

Hamilton, B. B., Granger, C. V., Sherwin, F. S., Zielezny, M. & Tashman, J. S. (1987). A uniform national data system for medical rehabilitation.(In M. J. Fuhrer (Ed)., Rehabilitation outcomes: Analysis and measurement (pp. 135—147). Baltimore: Brooks.)

Hamilton, B. B., Laughlin, J. A., Granger, C. V. & Kayton, R. M. (1991). Interrater agreement of the seven level Functional Independence Measure (FIM) [Abstract].( Archives of Physical Medicine and Rehabilitation, 72, 790.)

Henley, S., Petit, S., Todd-Prokeropek, A. & Tupper, A. M. (1985). Who goes home? Predictive factors in stroke recovery.( Journal of Neurology, Neurosurgery and Psychiatry, 48, 1—6.)

Indredavik, B., Bakke, F., Solberg, R., Rokseth, R., Haahein, L. L. & Holme, I. (1991). Benefit of a stroke unit: A randomized controlled trial.( Stroke, 22, 1026—1031.)

Ivnik, R. J., Malec, J. F., Tangalos, E. G., Petersen, R. C., Kokmen, E. & Kurland, L. T. (1992). Mayo's older American normative studies: WMS-R norms for ages 56 to 97.( The Clinical Neuropsychologist, 6 (suppl.), 49—82.)

Jongbloed, L. (1986). Prediction of function after stroke: A critical review.( Stroke, 17, 756—776.)

Kane, R. L., Chen, Q., Blewett, L. A. & Sangl, J. (1996). Do rehabilitative nursing homes improve the outcomes of care?( Journal of the American Geriatric Society, 44, 545—554.)

Kaste, M., Palomaki, H. & Sarna, S. (1995). Where and how should elderly stroke patients be treated? A randomized trial.( Stroke, 26, 249—253.)

Linacre, J. M., Heinemann, A. W., Wright, B. D., Granger, C. V. & Hamilton, B. B. (1994). The structure and stability of the Functional Independence Measure.( Archives of Physical Medicine and Rehabilitation, 75, 127—32.)

Lichtenberg, P. A., Manning, C. A., Vangel, S. J. & Ross, T. P. (1995). Normative and ecological validity data in older urban medical patients: A program of neuropsychological research.( Advances in Medical Psychotherapy, 8, 121—136.)

MacNeill, S. E. & Lichtenberg, P. A. (1997). Home alone: The role of cognition in return to independent living.( Archives of Physical Medicine and Rehabilitation, 78, 755—758.)

Moore, C. A. & Lichtenberg, P. A. (1996). Neuropsychological prediction of independent functioning in a geriatric sample: A double cross validational study.( Rehabilitation Psychology, 41, 115—130.)

Mui, A. C. & Burnette, J. D. (1994). A comparative profile of frail elderly persons living alone and those living with others.( Journal of Gerontological Social Work, 21, 5—26.)

Nelson, C. L., Herndon, J. E., Bark, D. B., Pryor, D. B. & Califf, R. M. (1991). Relation of clinical and angiographic factors to functional capacity as measured by the Duke Activity Status
Index.( American Journal of Cardiology, 68, 973—975.)

Rapp, S. R., Parisi, S. A. & Walsh, D. A. (1988). Psychological dysfunction and physical health among elderly medical inpatients.( Journal of Consulting and Clinical Psychology, 56, 851—855.)

Rapp, S. R., Parisi, S. A., Walsh, D. A. & Wallace, C. E. (1988). Detecting depression in elderly medical inpatients.( Journal of Consulting and Clinical Psychology, 56, 509—513.)

Rockwood, K., Stolee, P. & McDowell, I. (1996). Factors associated with institutionalization of older people in Canada: Testing a multifactorial definition of frailty.( Journal of the American Geriatric Society, 44, 578—582.)

Rozzini, R., Frisoni, G. B., Ferrucci, L. & Barbisoni, P. (1997). Who are the older patients failing to recover mobility after rehabilitation?( Journal of the American Geriatric Society, 45, 250—251.)

Rusin, M. J. (1990). Stroke rehabilitation: A geropsychological perspective.( Archives of Physical Medicine and Rehabilitation, 71, 914—922.)

Smith, D. S. & Clark, M. S. (1995). Competence and performance in activities of daily living of patients following rehabilitation from stroke.( Disability and Stroke, 17, 15—23.)

Sullivan, M., LaCroix, A., Baum, C., Resnick, A., Pabiniak, C., Grothaus, L., Katon, W. &

Wagner, E. (1996). Coronary disease severity and functional impairments: How strong is the relation?( Journal of the American Geriatric Society, 44, 1461—1465.)

Wade, D. T., Skilbeck, C. E., Wood, V. A. & Langton-Hewer, R. (1984). Long-term survival after stroke.( Age and Ageing, 13, 76—82.)

Wechsler, D. (1987). The Wechsler Memory Scale—Revised. (New York: Psychological Corporation)

Yesavage, J., Brink, T., Rose, T., Lum, O., Huang, V., Adez, M. & Leirer, V. (1983). Development and validation of a Geriatric Depression Screening Scale: A preliminary report. (Journal of Psychiatric Research, 17, 37—49.)

Copyright © 2000, Southern Development Group, S.A.  All Rights Reserved.
A Private Non-Profit Agency for the good of all, published in the UK & Honduras