http://cre.sagepub.com/Clinical Rehabilitation

http://cre.sagepub.com/content/17/7/750The online version of this article can be found at:

 DOI: 10.1191/0269215503cr673oa

2003 17: 750Clin RehabilAvital Hershkovitz, Daniel Gottlieb, Yichayaou Beloosesky and Shai BrillProgramme evaluation of a geriatric rehabilitation day hospital

  

Published by:

http://www.sagepublications.com

can be found at:Clinical RehabilitationAdditional services and information for    

  http://cre.sagepub.com/cgi/alertsEmail Alerts:

 

http://cre.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://cre.sagepub.com/content/17/7/750.refs.htmlCitations:  

What is This? 

- Jul 1, 2003Version of Record >>

at CORNELL UNIV on May 29, 2014cre.sagepub.comDownloaded from at CORNELL UNIV on May 29, 2014cre.sagepub.comDownloaded from

Clinical Rehabilitation 2003; 17: 750–755

© Arnold 2003 10.1191/0269215503cr673oa

Address for correspondence: Avital Hershkovitz, ‘Beit Rivka’Geriatric Rehabilitation Center, 4 Hachamisha St, PetachTikva 49245, Israel. e-mail: uritamar@netvision.net.il;uritamar@picknowl.com.au

Programme evaluation of a geriatric rehabilitationday hospitalAvital Hershkovitz ‘Beit Rivka’ Geriatric Rehabilitation Center Day Hospital, Petach Tikva and Sackler School ofMedicine, Tel-Aviv University, Tel Aviv, Daniel Gottlieb ‘Beit Rivka’ Geriatric Rehabilitation Center Day Hospital, PetachTikva, Yichayaou Beloosesky Department of Geriatrics, Rabin Medical Center, Golda Campus, Petach Tikva and SacklerSchool of Medicine, Tel-Aviv University, Tel Aviv and Shai Brill ’Beit Rivka’ Geriatric Rehabilitation Center Day Hospital,Petach Tikva and Sackler School of Medicine, Tel-Aviv University, Tel Aviv, Israel

Received 27th December 2001; returned for revisions 27th March 2002; revised manuscript accepted 8th June 2002.

Objective: To evaluate the rehabilitation programme in a geriatric day hospital.Design: An observational study.Setting: An urban geriatric rehabilitation day hospital.Subjects: Three hundred and �fty-three older patients admitted to arehabilitation day hospital during 2000: 163 post stroke, 113 withdeconditioning and 77 post orthopaedic surgery (hip fracture and jointreplacement).Main outcome measures: Functional Independence Measure (FIM),Nottingham Extended ADL Index, timed ‘get up and go’ test.Results: The mean discharge scores of all FIM items were signi�cantly higherthan the mean admission scores. Mean score change, however, for each itemwas less than one point. Most of the patients (70%) improved by less than5 points, regardless of main admission diagnoses. Patients with a loweradmission FIM score improved more than those with a higher admissionscore. The majority of the enrolled patients (92%) showed an improvement intheir Nottingham Extended ADL Index score. The mean (SD) score onadmission was 21 (11.9) and at discharge 31 (14.6). In all but three items(self-feeding, using a telephone and driving a car), mean scores at dischargewere signi�cantly higher than admission scores (p < 0.001). A signi�cantimprovement in timed ‘get up and go’ score was found for all patientsregardless of main diagnosis. Performance time decreased by 33% anddischarge scores for all patient groups were approximately 20 seconds.Conclusions: The notable improvement in mobility and instrumental activitiesof daily living on the one hand, and the minor improvements achieved in basicactivities of daily living on the other, suggests a need to revise treatmentgoals of day hospitals.

at CORNELL UNIV on May 29, 2014cre.sagepub.comDownloaded from

Programme evaluation for geriatric day hospital 751

Introduction

Geriatric rehabilitation day hospitals areregarded as an attractive setting for the provisionof rehabilitation services for older patients expe-riencing functional decline. They have the advan-tage of serving as a link between inpatientfacilities and the community and may also avoidunnecessary hospitalization.1,2

Day hospitals have several treatment goals:treating impairment, decreasing disabilities andhandicap, managing caregivers’ stress andimproving quality of life. Improving the patients’functional capacity (activities of daily living,ADL) is their major objective.

Many outcome measures are used to evaluatethe quality of treatment provided by day hospi-tals. Results achieved by these instrumentsdepend largely on their sensitivity to change, typeof population being examined (case-mix sever-ity), type of rehabilitation programme providedand its duration.3–9

The ambiguity in the literature as to the effec-tiveness of day hospitals results to some degreefrom the fact that in many studies, de�nitions forthe above parameters are not provided.10,11

Unfortunately, only limited discussion regardingan optimal rehabilitation programme provided byday hospitals is available, and a detailed descrip-tion of the treatment programme applied is sel-dom provided.6 Traditionally, most rehabilitationprogrammes are primarily targeted towardsimproving mobility and basic ADL. Whetherthese are appropriate treatment goals for currentday hospitals requires reconsideration.

The aim of the present study was to evaluatethe rehabilitation programme provided by theday hospital unit of ‘Beit Rivka’ Geriatric Reha-bilitation Center, using three validated outcomemeasures, and to reconsider rehabilitation goalsfor day hospitals.

Methods

The study population consisted of 353 patientsadmitted to a day hospital unit of a geriatric reha-bilitation center (‘Beit Rivka’, Petach Tikva,Israel) during the year 2000. One hundred andninety-one patients were referred from acute

inpatient and rehabilitation settings, and 162patients from the community. The mean (SD)age was 71 (9.8) years for males and 72 (9.8) forfemales. Male to female ratio was 183:170. Themain diagnoses of the admitted patients werestroke – 163 patients; deconditioning (patientswith numerous other conditions making themmore dependent) – 113 and orthopaedic surgery(hip fracture and joint replacement) – 77. Func-tional level was determined as dependent (29patients), partially dependent (244 patients) andindependent (80 patients) by the FunctionalIndependence Measure (FIM).12–14

On admission and at discharge from the dayhospital the patients were assessed by the fol-lowing measures: (a) FIM; (b) NottinghamExtended ADL Index,15 an instrumental activi-ties of daily living (ADL) measure, composed of22 self-reported questions regarding four dimen-sions: mobility, kitchen skills, domestic skills andleisure activities, with a score from 0 to 3 (fol-lowing Wade’s suggestion16) given according tothe degree of assistance needed to ful�l the tasks;(c) timed ‘get up and go’,17 which assesses bal-ance and risk of falling. The patient is observedand timed while rising from an armchair, walking3 m, turning, walking back and sitting downagain. The time score was found to correlate withfunctional capacity measured by the BarthelIndex.17,18 Patients with normal neurologicalfunction are expected to perform this test in lessthan 10 seconds. Patients who perform the test inless than 20 seconds are expected to be indepen-dently mobile and to have a gait speed of at least0.5 m/s (the minimum gait velocity required tocross a street). Patients who perform this test in30 seconds or more require assistance in manybasic and mobility tasks.17

Rehabilitation care included: (a) 30 minutes ofindividual physical therapy (improving transfer-ring, walking on smooth and rough surfaces,climbing stairs, improving equilibrium, musclestrength, and joint range of motion, etc.); (b) 30minutes of individual or group occupational ther-apy (basic ADL and instrumental ADL, cogni-tive evaluation and stimulation, use of physicaltechniques for the paralysed hand and bracesadjustment, greenhouse gardening and safetyeducation); (c) 60 minutes of group exercise, tar-geted to improve muscle strength, joint �exibility

at CORNELL UNIV on May 29, 2014cre.sagepub.comDownloaded from

752 A Hershkovitz et al.

evaluation. Fifty patients (14%) left rehabilita-tion (either due to transportation dif�culties orpreference for home treatment), most attendedless than �ve sessions. Twenty-seven (8%) werereturning patients (a second admission to dayhospital during year 2000) and were not includedin this study. Thirteen patients (4%) were read-mitted to inpatient wards due to deterioration intheir medical status. Data are missing for 25patients (7%). There was no difference betweenthe studied group and patients who were notincluded in the study in relation to age and gen-der, p = 0.282, 0.310 respectively. However, therewere more deconditioned and dependent patientsin the group of patients who were not includedin the study compared with more stroke and par-tially dependent patients in the studied popula-tion (p < 0.001).

The mean (SD) admission FIM score of thestudied population was 98 (10.2) (median was 101with a wide range of 49–116). Mean (SD) admis-sion score of stroke patients was 96 (12.3); decon-ditioned patients 97 (10.7) and orthopaedicpatients 100 (8.2). Mean (SD) discharge score ofthe studied population was 101 (8.9), with amedian of 104 and range of 49–117. Mean (SD)discharge score of stroke patients was 100 (11.6);deconditioned patients 100 (8.8) and orthopaedicpatients 103 (7.5). In all 18 FIM items, the meandischarge scores were signi�cantly higher thanthose obtained on admission. Although statisti-cally signi�cant, the mean change in score foreach item was less than one point. Of all FIMitems, only one (lower body dressing) manifesteda notable change in score (>0.5 point). The analy-sis of D-FIM distribution revealed that 70% ofthe patients improved by less than 5 points. Nosigni�cant difference in the distribution of D-FIMamong patients’ groups by main admission diag-nosis (stroke, deconditioning and orthopaedicsurgery) was found (p = 0.695).

In order to identify characteristics that maypredict improvement in basic ADL, all patientswere divided into two groups based on theirD-FIM score (patients with increased FIM score(D-FIM > 0); and patients without any progres-sion in FIM score (D-FIM £ 0) ). No differencesin age, gender and main diagnosis were foundbetween the two groups. Nevertheless, mostpatients in the �rst group had a signi�cantly

and range of motion; and (d) 30 minutes of activ-ity in a �tness room, mostly aerobic training(walking on a treadmill and riding stationarybicycle). Speech therapy, psychological evalua-tion, social worker intervention (with bothpatients and their caregivers) and medical carewere supplied according to the patients’ needs.Each patient attended three therapy sessions perweek, which included at least three of the aboveactivities.

The multidisciplinary team met weekly todecide on whether to continue and reviewprogress in treatment in light of targets. Lengthof stay was calculated as the sum of day atten-dance in a day hospital.

Statistical analysis was performed using SPSSfor Windows 10.01 (standard version, SPSS Inc.,1989–1999). Paired t-test was used to present: (a)the signi�cance (Bonferroni probability p <0.003) of change in FIM score and (b) the signif-icance of difference between two D-FIM (dis-charge FIM score minus admission score) groups(patients with D-FIM > 0 and those with D-FIM £0), in relation to age and admission FIM score.Chi-squared test was used to present the signi�-cance of difference between the two D-FIMgroups in relation to gender and main diagnosis.Logistic regression was used to detect the rela-tionship between the predictor variables and D-FIM. ANOVA with repeated measures was usedto present: (a) the difference between the threeadmission diagnosis groups (stroke, decondition-ing and orthopaedics) in relation to D-FIM; (b)the difference between the three admission FIMscore groups (£90, between 90 and 107 and >107)in relation to D-FIM; and (c) the signi�cance ofchange in the timed ‘get up and go’ score.Wilcoxon signed rank sum test was used to pre-sent the signi�cance of score change for the Not-tingham Extended ADL Index. The correlationbetween the motor part of FIM and the timed‘get up and go’ test was measured by Pearson cor-relation.

Results

Of the 353 patients admitted to the day hospital,236 patients (67%) were included in the study.Two patients were admitted for a one-session

at CORNELL UNIV on May 29, 2014cre.sagepub.comDownloaded from

Programme evaluation for geriatric day hospital 753

lower FIM score on admission than the patientsin the second group (p < 0.001). Logistic regres-sion analysis has also shown that FIM score onadmission is the only variable that can signi�-cantly predict improvement in basic ADL (oddsratio (OR) = 0.93; 95% con�dence interval (CI):0.89–0.97; p < 0.001).

All patients were assigned to three groupsaccording to their admission FIM score. GroupA (49 patients) scored less than 90, group B (145patients) scored between 90 and 107, and groupC (42 patients), scored over 107. The threegroups differed signi�cantly (p < 0.001) in theirD-FIM (6.1 (5.3), 3.4 (3.1) and 1.3 (1.8) respec-tively).

The Nottingham Extended ADL Index wasused in 84 stroke patients (out of 163). Of the 79patients who were not evaluated by this test, 20attended less than 5 days. Data were not col-lected in 59 patients due to language barrier,aphasia and cognitive decline (25 patients), unex-pected termination of the rehabilitation process(10), refused to participate (5), and investigatorerror (19). Most of the enrolled patients n = 77(92%) showed an improvement in their Notting-ham Extended ADL score, �ve patients (6%)had lower score and two (2%) showed no change.Mean (SD) score on admission was 21 (11.9) and31 (14.6) at discharge. In all but three items (self-feeding, using a telephone and driving a car),mean scores at discharge were signi�cantlyhigher than admission scores (p < 0.001). Anotable change in scores (>0.5 point) was foundin nine items: mobility (walking, climbing stairsand crossing the road), simple instrumental activ-ities (carrying a hot drink from one room toanother, washing dishes, preparing a snack andshopping) and in social activities.

Of the 353 admitted patients, 230 (65%) weremeasured by the timed ‘get up and go’ test onadmission and at discharge from the day hospi-tal. Fifty patients (14%) left rehabilitation, 21(6%) were returning patients, 13 (4%) were read-mitted, 14 (4%) were unable to complete the test,and for 25 (7%) data are missing. One hundredand seven were stroke patients, 59 deconditionedand 64 orthopaedic.

The mean (SD) time score on admission was32.6 (24.2) seconds and at discharge 21.8 (18.3),indicating an improvement of 33% in perfor-

mance. The mean (SD) time score on admissionfor stroke patients was 30.0 (25.7), deconditionedpatients 33.04 (25.7) and orthopaedic patients36.5 (19.3) seconds. The mean (SD) time score atdischarge was 20.4 (18.7), 23.6 (19.5) and 22.5(16.5) seconds respectively. The decrease in timescores was statistically signi�cant for all threegroups (p < 0.001). Changes in mean time scoreswere similar in all patients’ groups (p = 0.354).The mean time score at discharge was close tothe desirable value of 20 seconds for all patients’groups (75% had a mean score of less than 25seconds).

One hundred and ninety-six patients were eval-uated by both FIM and the timed ‘get up and go’test: 102 were stroke patients, 42 deconditionedand 52 orthopaedic. A signi�cant correlation wasfound between the motor part of FIM (whichincluded �ve parameters: transferring from bedto a chair, to a toilet and to a bath, walking andclimbing stairs) and the timed ‘get up and go’ test(r = 0.48; p < 0.001).

The average (SD) length of stay of admittedpatients was 18 (7), 16 (8) and 18 (7) for stroke,deconditioned and orthopaedic patients, respec-tively.

Discussion

In the present study, day hospital patients(regardless of age, gender and main diagnoses)did not demonstrate signi�cant clinical change inbasic ADL, as measured by FIM. This may resultfrom the ‘ceiling effect’ attributed to disabilitymeasures.19,20 The majority of our patients (54%)received one to three months of inpatient reha-bilitation care prior to their referral to the dayhospital. During that period, the patients hadalready achieved signi�cant progression in basicADL and mobility and most of their rehabilita-tion potential had already been exhausted. Our�nding that only patients with lower admissionFIM score achieved clinical meaningful D-FIMsupports the above notion. The fact that most ofour disabled and handicap patients received 24-hour home assistance by a skilled worker couldalso contribute to the minimal change in basicADL. Most of the patients preferred the conve-nience of the workers’ help over personal physi-

at CORNELL UNIV on May 29, 2014cre.sagepub.comDownloaded from

754 A Hershkovitz et al.

ment (33%) in mobility, as re�ected by the timed‘get up and go’ test. This test is one of severalmobility outcome measures used for evaluatingday hospital patients.3,8,21,22 Its correlation withthe motor part of FIM (r = 0.48) attests to itsvalidity and hence its utility as an appropriateoutcome measure to be used in day hospitals.This notion is also supported by the fact that asigni�cant correlation exists between timed ‘getup and go’ test scores and functional capacitymeasured by the Barthel Index.17

The major goal in our rehabilitation pro-gramme was to reach mobility independence, i.e.,to perform the timed ‘get up and go’ test in lessthan 20 seconds (time necessary to cross a streetsafely). Most patients (75%), regardless of theiradmission diagnosis, achieved this average timed‘get up and go’ score of approximately 20 sec-onds. Since most of our patients (92%) wereindependent or partially dependent in basicADL, and therefore expected to leave the house,this outcome became one of the criteria for ter-minating the rehabilitation treatment.

It is necessary to identify concrete goals andincorporate them into suitable programmes inday hospitals. The results of several studies,including ours, support the assumption that dayhospital patients have reached a plateau on therehabilitation curve in basic ADL during hospi-tal rehabilitation and, therefore, repeated train-ing in a day hospital might be ineffective.Therefore, it is essential that day hospitals pro-mote programmes that focus on achieving moreindependence in extended activities of daily liv-ing (e.g., bring the patients to leave the house,use the stairs, participate in household tasks andtake part in social activities). Improving thepatients’ mobility should be a major goal of dayhospital rehabilitation care. We suggest that atimed ‘get up and go’ score of 20 seconds mightbe used as a discharge criterion from day hospi-tals for patients who are not homebound.

AcknowledgementsThe authors wish to thank Yael Oran, BOT,

Shirley Fox, BOT, Ilana Akrish, BOT, HavaGilad, BPT, Yael Gozlav, BPT and Rachel Alter,BPT, for assistance with data collection and IlanaGelernter, MA, for assistance with data analysis.

cal challenges. Our results concur with otherstudies that failed to identify signi�cant clinicalchanges in basic ADL among day hospitalpatients, using another disability measure(Barthel Index).5–7,19

Unlike basic ADL, a great improvement ininstrumental ADL was noted in day hospitalstroke patients, as re�ected by an average of a48% increment in Nottingham Extended ADLscore. The use of extended ADL measure is moreappropriate for day hospital patients (who gen-erally return home and struggle to restore theirformer life schedule and hobbies), re�ectingmore reliably their functional improvement. Thisinstrument was found to be a valid measure ofinstrumental ADL performance.20 Parker et al.did not �nd a signi�cant change between admis-sion and discharge scores.19 Burch et al. found asigni�cant improvement in only two subscales ofthis instrument.3 Our results show a signi�cantchange in all four subscales (mobility, kitchenand domestic tasks and leisure activities). Inthree tasks: self-feeding, using a telephone anddriving a car, no signi�cant change in scores wasfound. This may be related to the fact that mostpatients on admission gave a high score for self-feeding and using the telephone, and most didnot drive before the stroke. The low scoreobtained for gardening derived from the patients’living habitat (mainly in small apartments).

All patients (stroke, deconditioning andorthopaedics) manifested remarkable improve-

Clinical messages

� Day hospital patients did not achieve a sig-ni�cant clinical change in basic activities ofdaily living as measured by Functional Inde-pendence Measure.

� Treatment goals of geriatric rehabilitationday hospitals should be directed towardimproving mobility and instrumental activi-ties of daily living and resuming leisureactivities.

� A timed ‘get up and go’ score of 20 secondsis suggested as a discharge criterion from ageriatric rehabilitation day hospital forpatients who are not homebound.

at CORNELL UNIV on May 29, 2014cre.sagepub.comDownloaded from

Programme evaluation for geriatric day hospital 755

van Gijn JC, Eisma WH. Effects of day hospitalrehabilitation in stroke patients: a review ofrandomized clinical trials. Scand J Rehabil Med1998; 30: 87–94.

12 Keith RA, Granger CV, Hamilton BB, Sherwin FS.The functional independence measure: a new toolfor rehabilitation. Adv Clin Rehabil 1987; 1: 6–18.

13 Hamilton BB, Granger CV. Disability outcomesfollowing inpatient rehabilitation for stroke. PhysTher 1994; 74: 494–503.

14 Hamilton BB, Granger CV, Sherwin FS, ZieleznyM, Tashman JS. A uniform national data systemfor medical rehabilitation. In: Fuhrer MJ ed.Rehabilitation outcomes: analysis andmeasurement. Baltimore: Paul H Brooks Publishing,1987: 137–47.

15 Nouri FM, Lincoln NB. An extended activities ofdaily living scale for stroke patients. Clin Rehabil1987; 1: 301–305.

16 Wade DT. Measurement in neurologicalrehabilitation. New York: Oxford University Press,1995.

17 Podsiadlo D, Richardson S. The timed ‘up & go’: atest of basic functional mobility for frail elderlypersons. J Am Geriatr Soc 1991; 39: 142–48.

18 Collin C, Wade DT, Davies S, Horne V. TheBarthel ADL index: a reliability study. Int DisabilStud 1988; 10: 61–63.

19 Parker SG, Du X, Bardsley MJ et al. Measuringoutcomes in care of the elderly. J R Coll Physicians(Lond) 1994; 28: 428–33.

20 Gompertz P, Pound P, Ebrahim S. Validity of theextended activities of daily living scale. Clin Rehabil1994; 8: 275–80.

21 Wong SF, Yap KB, Chan KM. Day hospitalrehabilitation for the elderly: a retrospective study.Ann Acad Med Singapore 1998; 27: 468–73.

22 Spilg EG, Martin BJ, Mitchell SL, Aitchison TC. Acomparison of mobility assessments in a geriatricday hospital. Clin Rehabil 2001; 15: 296–300.

References

1 A Report of the Research Unit of the RoyalCollege of Physicians and British Geriatric Society.Geriatric day hospital: their role and guidelines forgood practice. London: Royal College of Physicians,1994.

2 Royal College of Physicians and British GeriatricSociety. Standardised assessment scales for elderlypeople. London: Royal College of Physicians, 1992.

3 Burch S, Longbottom J, McKay M, Borland C,Prevost T. The Huntington Day Hospital Trial:secondary outcome measures. Clin Rehabil 2000; 14:447–53.

4 Hui E, Lum CM, Woo J, Or KH, Kay RL.Outcomes of elderly stroke patients. Day hospitalversus conventional medical management. Stroke1995; 26: 1616–19.

5 Harwood RH, Ebrahim S. Measuring the outcomesof day hospital attendance: a comparison of theBarthel Index and London Handicap Scale. ClinRehabil 2000; 14: 527–31.

6 Burch S, Longbottom J, McKay M, Borland C,Prevost T. A randomized controlled trial of dayhospital and day centre therapy. Clin Rehabil 1999;13: 105–12.

7 Eagle DJ, Guyatt GH, Patterson C, Turpie I,Sackett B, Singer J. Effectiveness of a geriatric dayhospital. Can Med Assoc 1991; 144: 699–704.

8 Werner RA, Kessler S. Effectiveness of an intensiveoutpatient rehabilitation program for post acutestroke patients. Am J Phys Med Rehabil 1996; 75:114–20.

9 Young J, Forster A. Day hospital and homephysiotherapy for stroke patients: a comparativecost-effectiveness study. J R Coll Physicians (Lond)1993; 27: 252–58.

10 Forster A, Young J, Langhorne P. Systematic reviewof day hospital care for elderly people. BMJ 1999;318: 837–41.

11 Dekker R, Drost EA, Groothoff JW, Arendzen JH,

at CORNELL UNIV on May 29, 2014cre.sagepub.comDownloaded from