402696

yai et alNeurorehabilitation and Neural Repair

© The Author(s) 2010

Reprints and permission: http://www.

NNR25610.1177/1545968311402696Mi

sagepub.com/journalsPermissions.nav

Research Articles
Neurorehabilitation and

Results of New Policies for Inpatient Neural Repair

25(6) 540­–547
© The Author(s) 2011
Rehabilitation Coverage in Japan Reprints and permission: http://www.
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1545968311402696
http://nnr.sagepub.com

Ichiro Miyai, MD, PhD1,2, Shigeru Sonoda, MD, PhD1,3,

Shota Nagai1,4,Yuko Takayama1,5, Yukiko Inoue, PhD6,
Atsuo Kakehi, PhD6, Masaki Kurihara, MD1,7,
and Makoto Ishikawa, MD1,8

Abstract
Background. A new interdisciplinary postacute rehabilitation unit, the Kaifukuki (convalescent) rehabilitation ward (KRW),
has been incorporated into the Japanese medical insurance system since 2000. More than 57 000 beds (45 beds per 100 000
population) are currently available nationwide. The maximal coverage for therapy sessions increased from 2 to 3 hours per
day, 7 days a week, in 2006. Objective. To investigate how changes in policy affected rehabilitation outcomes of KRWs in
a retrospective cohort study of 87 917 patients over 10 years. Results. The mean (standard deviation) age of the patients
was 73.0 (13.8) years, and 55.4% were women. Diagnoses included stroke (47.9%); orthopedic diseases, including hip
fracture (35.2%); and traumatic brain and spinal cord injury (5.4 %). Onset–admission interval (OAI) was 31.5 (18.6)
days, length of stay was 75.9 (46.1) days, and 69.1% were discharged home. Daily therapy time was 79.4 (34.5) minutes.
Admission/discharge scores of the Barthel Index and the Functional Independence Measure were 49.3 (31.0)/70.4 (31.9) and
75.3 (31.2)/91.7 (31.8), respectively. Year-by-year comparison revealed that older age, greater initial disability, and shorter OAI
were coupled with a higher dose of rehabilitative interventions and a higher rate of home discharge. Longitudinal data from
a cohort of hospitals implied a small but significant dose-dependent effect of hours of therapy on rehabilitation outcome
after stroke. Conclusions. Although the organization of KRWs is in flux as the system of hospitals grows, results over the
past 8 years suggest that changes in national insurance policies are affecting the quantity and organization of rehabilitation
interventions and improvement in patient outcomes.

Keywords
inpatient rehabilitation facility, rehabilitation outcome, stroke, medical insurance

Introduction The Japanese national insurance system introduced the

Based on the findings from randomized controlled trials of Kaifukuki rehabilitation ward (KRW) in 2000. Kaifuku-ki
organized stroke units, there is an international consensus represents the recovery or convalescent phase. To facilitate
that early, interdisciplinary, and more hours of rehabilitation
improves activities of daily living (ADLs), walking status, 1
Annual Survey Committee, Kaifukuki Rehabilitation Ward Association,
and percentage of patients discharged home.1-6 Length of Tokyo, Japan
stay (LOS) in inpatient rehabilitation facilities (IRFs) has 2
Neurorehabilitation Research Institute, Morinomiya Hospital, Osaka,
been considerably shortened over the past 2 decades, espe- Japan
cially in North America.7 One of the possible solutions to
3
Nanakuri Sanatorium, Fujita Health University, Mie, Japan
4
shorter LOS is early supported discharge, where patients are Kinjo University, Ishikawa, Japan
5
Ukai Rehabilitation Hospital, Aichi, Japan
managed in the community by an interdisciplinary team.8-11 6
National Institute of Public Health, Saitama, Japan
In Japan, there had been no definitive system for an inpa- 7
Nagasaki Rehabilitation Hospital, Nagasaki, Japan
tient facility to provide early and intensive rehabilitation. 8
Hatsudai Rehabilitation Hospital, Tokyo, Japan
Consequently, patients with stroke often received low-intensity
Corresponding Author:
inpatient rehabilitation for more than 6 months. With the Ichiro Miyai, Neurorehabilitation Research Institute, Morinomiya
increasing number of elderly patients, rehabilitation resources Hospital, 2-1-88, Morinomiya, Joto-ku, Osaka, 536-0025, Japan
had to be more effectively used. Email: webeo@ga2.so-net.ne.jp
Miyai et al 541
Figure 1. Number of beds in Kaifukuki rehabilitation wards (KRWs; there has been a constant increase in the number of KRW beds).
an interdisciplinary team approach, the KRW team had to
provide patients and their families with a comprehensive
monthly rehabilitation plan, including information about
achieved goals, planned goals, and rehabilitative approaches
to achieve the goals; discharge planning; and social resources
necessary for a home discharge. Timing for discharge was
set when patients reached a plateau in ADLs according to an
interactive evaluation, which has facilitated home discharge
of severely disabled patients with stroke.12 The more orga-
nized rehabilitative intervention in the KRW also resulted in
improved functional outcome.13
The Ministry of Health, Labor and Welfare of Japan has
defined the KRW as the main system of IRF covered by the
medical insurance system. All patients who still need assis-
tance in ADLs after treatment in acute hospitals are trans-
ferred to KRWs. The number of beds has continuously
increased—3651 beds in 2000, 14 865 in 2002, 27 261 in
2004, 40 403 in 2006, 53 802 in 2008, and 57 028 in 2009—
comprising 4.5% of total hospital beds in Japan (Figure 1).
There have been several changes in the insurance policy for
the KRW. Initially, patients within 3 months after the onset of
disabling diseases, including stroke, traumatic brain injury,
and other neurological diseases as well as orthopedic dis-
eases such as hip fracture, were eligible for admission to the
KRW. The maximal onset–admission interval (OAI) was short-
ened to 2 months in 2006 to facilitate earlier transfer to the
KRW. Maximal coverage for inpatient rehabilitation increased
from 2 hours up to 3 hours per day in 2006. These decisions
were made based on accumulating evidence that early and
intensive interventions improve rehabilitation outcome.1-6
Currently, the maximal LOS covered by the insurance is
150 days for stroke, 180 days for stroke and other neurologi-
cal diseases with severe disability and cognitive disorders,
and 90 days for orthopedic diseases and the disuse/myopathy
syndrome. Accordingly, the maximal duration of coverage,
including acute and postacute inpatient care, for patients with
severe stroke is 8 months after onset.14 The KRW, comparable
with IRFs in North America and Europe, includes extended
coverage of inpatient rehabilitation up to 180 days after stroke
and 3 hours of rehabilitation per day, including weekends
(21 hours per week).
Because there is no similar rehabilitation system in the
world in terms of LOS and intensity of rehabilitative interven-
tions, we describe how the KRW has contributed to improving
the functional outcome of patients as it has evolved.
Methods
We retrospectively analyzed 87 917 patients from the data-
base of the annual survey for the KRW from 2001 through
2008. The data included adult patients (≥16 years old) who
were discharged from a KRW in August and September each
542 Neurorehabilitation and Neural Repair 25(6)
year after more than a 6-day stay on the ward. Data sheets
were sent to the member hospitals of the KRW association.
They were required to provide demographic data of patients;
dose of rehabilitation; outcome measures, including either the
Barthel Index (BI)15 or the Functional Independence Measure
(FIM)16; discharge disposition; and LOS. The data collection
rate improved year by year, from 46.8% (2001) to 64.1%
(2009) of KRWs nationwide.
In addition to changes of maximal OAI and rehabilitation
coverage in 2006, the policy introduced an incentive for
KRWs that admitted severely disabled patients, who repre-
sented more than 15% of total patients and whose rate of
home discharge was more than 60% in 2008. Influences of
these changes of policy on outcome measures were analyzed.
Because hospitals with a KRW included in the survey dif-
fered across the years, we identified the 110 hospitals that
consecutively participated in the survey from 2006 to 2009.
We also analyzed the serial changes of rehabilitation out-
come in this hospital group.
Longitudinal changes of demographic data and outcome
measures were analyzed using 1-way analysis of variance
(ANOVA) or χ2 tests as applicable. Assessment of the effect
of augmented therapy dose on functional outcome used a
multiple stepwise regression model. Statistical analyses
were performed using SPSS version 18.0J (SPSS Inc,
Chicago, Illinois). Statistical significance was set at P < .05.
Results
General Description
The mean (standard deviation [SD]) age of all patients was 73.0
(13.8) years, and 55.0% were women. Age (F8,87 908 = 102.9;
P < .001) and the proportion of women (P < .001) were sig-
nificantly higher in recent years (Table 1). Diagnostic cat-
egories were stroke (47.9%); orthopedic diseases, including
hip fracture and disuse myopathy (35.2%); and neurological
diseases, including traumatic brain and spinal cord injury
(5.3 %). The proportion of stroke decreased and that of
orthopedic diseases increased year by year (P < .001).
Mean OAI (SD) and LOS were 31.5 (18.6) and 75.9 (46.1)
days; 69.1% of patients were discharged home. The policy
change for maximal OAI in 2006 resulted in significantly
shorter OAIs (ie, shorter LOSs in acute hospitals; F8,87 908 =
675.5, P < .001) and LOSs (F8,87 908 = 68.3; P < .001). The
dose of rehabilitation is expressed as the total time spent in
physical, occupational, and speech therapy sessions per day
in Table 1. The increase of maximal coverage for therapy in
2006 produced a significantly higher dose of therapy thereaf-
ter (F7,77 432 = 1106.289; P < .001). Consequently, augmented
therapy of more than 15 hours per week was provided to
15.5% of patients in 2009. In terms of weekend intervention,
96.3% and 41.1% of KRWs provided therapy sessions on
Saturdays and Sundays in 2009 (Table 1).
Mean (SD) scores of admission/discharge BI and FIM
were 49.3 (31.0)/70.4 (31.9) and 75.3 (31.2)/91.7 (31.8),
respectively (Table 1). Patients with severe disability were
prominent in recent years both on admission (F5,46 039 = 14.9;
P < .001 for FIM) and discharge (F5,46 039 = 16.4; P < .001 for
FIM). The policy change in 2008 resulted in higher rate of
home discharge in spite of more patients with severe disabil-
ity (Table 1).
Analysis of Stroke Data
Age of patients with stroke was significantly higher (F8,41 959 =
23.1; P < .001) in recent years, and the proportion of males
was consistently high (Table 2). The policy change in 2006
resulted in significantly shorter OAIs (F8,41 959 = 379.8; P <
.001). LOS shortened in 2006 but significantly increased in
2008 and 2009 (P < .001). The increase of maximal cover-
age for therapy in 2006 led to greater doses of therapy
thereafter (F7,36 709 = 889.7; P < .001). Therapy for more
than 15 hours per week was provided to 24.6% of patients
in 2009. The number of patients with severe disability
increased in recent years both on admission (F5,21 894 = 7.7;
P < .001 for total FIM) and discharge (F5,21 894 = 8.6; P <
.001 for total FIM). Similar changes were also seen in
admission/discharge scores of the motor component of the
FIM (F5,21 894 = 8.1; P < .001/F5,21 894 = 8.4; P < .001). The
policy change in 2008 resulted in a higher rate of home dis-
charges in spite of there being more patients with severe
disability (Table 2).
Analyses for Serial Data From
the Identical Hospital Group
To investigate whether augmented therapy time affected
functional outcome, we analyzed serial data of patients with
stroke (n = 10 078) from the very same hospitals that con-
secutively participated in the survey from 2006 to 2009
(Table 3). A year-by-year comparison revealed that OAI
significantly decreased (F3,10 074 = 4.1; P < .01), whereas LOS
significantly increased (F3,10 074 = 17.8; P < .001). In terms
of dose of intervention, therapy time (F3,9414 = 265.3; P < .001)
and rate of patients who were provided more than 15 hours
of therapy significantly increased (P < .01). For the FIM, the
total (F3,9745 = 5.1; P < .05), motor (F3,9745 = 3.4; P < .05), and
cognition (F3,9745 = 3.7; P < .05) scores on admission and
the total (F3,9745 = 5.2; P < .005) and motor (F3,9745 = 6.0;
P < .001) scores on discharge were significantly different
among the years. However, post hoc Bonferroni testing failed
to show consistent yearly changes in these measures. Of note,
the gains of total FIM scores were significantly greater in
2009 than in 2006 (P < .001) and 2008 (P < .005), although
FIM efficiencies were unchanged. The data also revealed
that significantly more patients were discharged home in
2009 (P < .01).
 Table 1. Clinical Characteristics of Patients Admitted to KRWs

2001 2002 2003 2004 2005 2006 2007 2008 2009 Total
Number of hospitals included in the    60    86   145   218   272     331    418    509 580  
survey
Number of patients 1665 2536 4807 7643 9044 11 463 15 245 18 278 17 236 87 917
Mean age (SD), y 69.8 (14.1) 69.9 (14.0) 71.1 (14.2) 71.5 (14.0) 71.5 (14.2) 72.6 (13.9) 73.4 (13.4) 73.9 (13.5) 74.5 (13,4) 73.0 (13.8)
Female, % 49.5 53.4 52.9 54.1 53.6 55.2 55.4 57.0 57.1 55.4
Type of disabling disease  
Stroke, % 61.3 59.4 51.1 50.7 52.5 48.4 48.8 44.4 43.2 47.9
Neurological diseases, %   2.9   3.7   4.4   4.3 4.1   5.0   5.4   6.7   5.7   5.3
Orthopedic diseases, % 18.9 27.1 31.0 31.0 31.3 35.8 34.3 39.0 39.5 35.2
Disuse syndrome, %   6.8   6.9   9.6 10.6 9.4 10.9 11.6 10.0 11.6    1053
Others, % 10.2   2.8   3.8   3.4 2.6   0.0   0.0   0.0     0   1.1
Mean onset to admission (SD), d 40.6 (24.4) 41.1 (24.4) 39.0 (22.8) 37.7 (22.5) 37.6 (22.3) 29.0 (15.6) 28.8 (15.8) 28.7 (15.6) 28.4 (15.6) 31.5 (18.6)
Mean length of stay (SD), d 86.8 (50.0) 85.6 (51.5) 79.4 (50.2) 80.9 (50.9) 79.5 (49.4) 71.3 (42.2) 74.4 (45.0) 74.2 (45.1) 74.7 (43.3) 75.9 (46.1)
Discharge disposition  
Home, % 64.9 69.8 67.7 67.6 67.0 67.1 66.9 70.9 73.0 69.1
Acute unit, %   5.3   4.8   5.4   5.5 5.3   5.2   5.5   5.5   5.1   5.3
Other ward, %   5.3   4.0   5.0   4.3 4.7   5.3   4.6   3.6   3.4   4.2
Chronic hospital, % 13.2 10.5   9.4 10.4 10.0   8.3   8.0   6.9   6.5   8.2
Intermediate care, %   8.6   7.7   7.3   7.3 8.3   8.8 10.0   7.3   7.0   8.0
Skilled nursing facility, %   1.9   2.4   3.8   4.0 3.8   4.4   4.1   4.9   4.5   4.2
Died, %   0.8   0.9   1.4   0.9 0.9   1.0   0.9   0.9   0.5   0.9
Mean admission BI (SD) 48.7 (29.0) 51.6 (30.5) 49.5 (32.0) 49.5 (30.7) 48.9 (30.9) 48.7 (30.8) 49.7 (32.1) 48.5 (30.1) 49.7 (30.4) 49.3 (31.0)
Mean discharge BI (SD) 70.5 (30.3) 73.3 (30.8) 73.0 (31.1) 71.6 (31.0) 70.8 (32.1) 70.2 (32.0) 70.2 (32.9) 68.9 (31.6) 69.3 (31.3) 70.4 (31.9)
Mean BI efficiency (SD) 0.33 (0.60) 0.35 (0.48) 0.42 (0.73) 0.37 (0.58) 0.36 (0.55) 0.41 (0.64) 0.36 (0.58) 0.37 (0.60) 0.34 (0.42) 0.37 (0.57)
Mean admission FIM (SD) NA 80.7 (31.9) 78.9 (31.6) NA NA 76.0 (31.9) 74.4 (31.6) 74.7 (31.0) 75.2 (30.7) 75.3 (31.2)
Mean discharge FIM (SD) NA 97.3 (29.7) 95.4 (31.0) NA NA 92.4 (32.4) 91.1 (32.2) 90.6 (31.9) 91.7 (31.2) 91.7 (31.8)
Mean FIM efficiency (SD) NA 0.24 (0.43) 0.27 (0.48) NA NA 0.29 (0.46) 0.28 (0.45) 0.26 (0.42) 0.27 (0.36) 0.27 (0.42)
Mean therapy time per day (SD), min NA 65.1 (29.2) 63.0 (25.8) 59.7 (24.8) 65.0 (25.1) 79.8 (32.1) 83.5 (32.5) 89.8 (34.0) 85.7 (39.5) 79.4 (34.5)
% patients provided >15 hours therapy NA   1.6   0.8   1.1 0.1   8.0   8.4 12.4 15.5   8.3
per week
% services on Saturday NA 76.0 98.8 98.1 95.1 94.6 82.2 96.9 96.3 91.5
% services on Sunday NA 10.7 22.4 15.9 20.6 23.3 23.4 27.7 41.1 26.0
Abbreviations: KRW, Kaifukuki rehabilitation ward; SD, standard deviation; BI, Barthel Index; FIM, Functional Independence Measure; NA, not available.

543
544
Table 2. Clinical Characteristics of Patients With Stroke Admitted to KRWs

2001 2002 2003 2004 2005 2006 2007 2008 2009 Total
Number of patients 1020 1506 2450 3855 4747 5546 7376 8027 7441 41 968
Mean age (SD), y 68.5 (11.9) 68.8 (11.5) 70.1 (11.8) 70.6 (11.5) 70.3 (12.0) 71.1 (12.1) 71.2 (12.3) 71.7 (12.3) 71.8 (12.5) 71.2 (12.1)
Female, % 40.5 43.4 44.7 44.9 44.3 44.1 43.4 43.9 43.9 43.9
Mean OAI (SD), d 44.1 (23.5) 46.0 (24.1) 43.3 (23.1) 41.9 (22.4) 41.6 (22.1) 33.1 (14.8) 32.6 (14.9) 32.4 (15.1) 32.4 (14.9) 35.8 (18.5)
Mean LOS (SD), d 95.2 (49.1) 97.0 (49.9) 91.6 (51.9) 94.4 (51.8) 91.4 (50.4) 86.1 (46.7) 90.2 (49.5) 93.1 (50.9) 94.3 (48.8) 91.9 (49.8)
Discharge disposition  
Home, % 63.4 69.1 63.9 64.6 64.7 64.3 64.9 66.9 69.3 66.0
Acute unit, %   5.3   4.8   6.9   6.0   5.5   5.7   5.8   6.4   5.9   5.9
Other ward, %   5.2   3.8   5.4   4.7   5.1   5.1   4.3   4.1   3.5   4.4
Chronic hospital, % 15.3 11.9 11.6 12.5 12.2 10.6   9.8   8.5   8.4 10.2
Intermediate care, %   8.7   7.7   7.5   8.1   8.7   9.8 10.4   8.6   8.8   9.0
Skilled nursing facility, %   1.6   2.1   3.2   3.2   3.0   3.6   3.9   4.3   3.7   3.6
Died, %   0.5   0.7   1.4   0.9   0.8   1.0   0.9   1.2   0.6   0.9
Mean admission BI (SD) 45.6 (28.7) 47.4 (30.6) 43.5 (31.7) 44.3 (30.7) 44.7 (31.3) 43.0 (31.0) 45.5 (32.8) 43.5 (30.9) 44.2 (31.4) 44.6 (31.5)
Mean discharge BI (SD) 68.8 (30.3) 71.1 (30.0) 68.1 (32.1) 67.8 (32.2) 66.9 (33.3) 65.2 (33.6) 67.4 (34.2) 63.9 (33.4) 65.2 (33.4) 66.8 (33.3)
Mean BI efficiency (SD) 0.31 (0.64) 0.31 (0.41) 0.37 (0.66) 0.32 (0.54) 0.30 (0.43) 0.32 (0.56) 0.30 (0.51) 0.29 (0.54) 0.27 (0.37) 0.30 (0.51)
Mean admission FIM (SD) NA 74.6 (31.5) 71.9 (30.9) NA NA 70.3 (32.1) 69.0 (31.7) 68.0 (31.1) 68.9 (31.6) 69.3 (31.6)
Mean discharge FIM (SD) NA 93.6 (30.5) 90.6 (31.8) NA NA 88.6 (33.2) 87.8 (32.9) 86.2 (33.0) 87.8 (33.0) 87.9 (32.9)
Mean FIM efficiency (SD) NA 0.23 (0.33) 0.24 (0.40) NA NA 0.27 (0.45) 0.27 (0.45) 0.24 (0.38) 0.24 (0.30) 0.25 (0.39)
Mean admission motor FIM (SD) NA 51.3 (24.1) 49.1 (23.3) NA NA 47.8 (24.3) 46.8 (24.2) 46.1 (23.7) 46.8 (24.2) 47.0 (24.1)
Mean discharge motor FIM (SD) NA 66.9 (23.4) 65.0 (24.1) NA NA 63.3 (25.2) 62.7 (25.2) 61.4 (25.4) 62.7 (25.5) 62.7 (25.2)
Mean admission cognition FIM (SD) NA 23.4 (9.8) 22.8 (9.8) NA NA 22.5 (9.6) 22.2 (9.6) 21.9 (9.4) 22.2 (9.5) 22.2 (9.6)
Mean discharge cognition FIM (SD) NA 26.8 (8.6) 25.5 (9.3) NA NA 25.4 (9.3) 25.2 (9.1) 24.7 (9.1) 25.2 (9.0) 25.2 (9.1)
Mean therapy time per day (SD), min NA 69.7 (28.9) 66.7 (25.6) 64.8 (24.1) 70.1 (25.0) 86.1 (31.4) 90.5 (32.7) 96.1 (34.0) 101.6 (37.2) 86.7 (34.5)
Percentage of patients provided >15 NA   4.6   2.4   0.8   0.9 10.6 11.6 18.1 24.6 12.4
hours of therapy per week
Abbreviations: KRW, Kaifukuki rehabilitation ward; SD, standard deviation; OAI, onset–admission interval; LOS, length of stay; BI, Barthel Index; FIM, Functional Independence Measure; NA, not available.
Miyai et al 545

Table 3. Rehabilitation Outcome of Patients With Stroke From the Hospitals That Consecutively Participated in the Survey From
2006 to 2009

2006 2007 2008 2009 Total

Number of patients 2344 2546 2377 2811 10 078
Mean age (SD), y 70.9 (12.0) 70.1 (12.6) 70.8 (12.2) 70.3 70.5 (12.5)
Female, % 44.7 42.9 42.3 41.8 42.9
Mean OAI (SD), d 33.4 (14.8) 32.6 (14.9) 32.1 (14.7) 32.1 (14.4) 32.5 (14.7)
Mean LOS (SD), d 85.9 (44.3) 87.8 (47.2) 91.6 (49.4) 94.7 (47.3) 90.2 (47.3)
Discharge home, % 64.7 64.8 65.5 70.6 66.6
Transfer to acute unit or death, %   6.0   5.8   8.3   6.7   6.6
Mean admission FIM (SD) 69.7 (32.0) 67.6 (31.3) 66.8 (30.8) 68.5 (30.7) 68.2 (31.2)
Mean discharge FIM (SD) 88.3 (33.2) 87.5 (32.8) 85.9 (33.1) 89.5 (32.1) 87.9 (32.7)
Mean FIM gain (SD) 18.7 (18.1) 19.9 (19.6) 19.1 (18.9) 20.9 (17.5) 19.7 (18.5)
Mean FIM efficiency (SD) 0.27 (0.43) 0.28 (0.46) 0.25 (0.40) 0.26 (0.29) 0.26 (0.40)
Mean admission motor FIM (SD) 47.3 (24.4) 45.8 (23.8) 45.2 (23.5) 46.7 (23.5) 46.3 (23.8)
Mean discharge motor FIM (SD) 63.1 (25.3) 62.4 (25.0) 61.2 (25.5) 64.1 (24.6) 62.7 (25.1)
Mean admission cognition FIM (SD) 22.4 (9.8) 21.8 (9.4) 21.5 (9.3) 21.8 (9.3) 21.9 (9.4)
Mean discharge cognition FIM (SD) 25.2 (9.3) 25.1 (9.1) 24.7 (9.0) 25.3 (8.8) 25.1 (9.0)
Mean therapy time per day (SD), min 87.6 (30.4) 94.8 (31.9) 98.4 (29.8) 112.9 (37.4) 99.4 (34.2)
Percentage of patients provided >15 10.6 13.8 14.6 36.6 19.6
hours of therapy per week
Abbreviations: SD, standard deviation; OAI, onset–admission interval; LOS, length of stay; FIM, Functional Independence Measure; NA, not available.

Table 4. Stepwise Multiple Regression Analysis for Discharge FIM Scorea

Standardized 95% Confident 95% Confident

Coefficient Coefficient P Value Interval Lower Limit Interval Upper Limit
Constant 18.44 <.001 16.600 20.276
Admission FIM 0.916 0.871 <.001 0.904 0.9283
LOS 0.094 0.136 <.001 0.086 0.102
OAI -0.144 -0.064 <.001 -0.170 -0.119
Therapy time per day 0.034 0.036 <.001 0.024 0.045
Abbreviations: FIM, Functional Independence Measure; LOS, length of stay; OAI, onset–admission interval.
a 2
r = 0.714, ANOVA < .001.

Thus, to test whether augmented therapy influenced func-
tional outcome, we performed a stepwise multiple regression
analysis. The model indicated that the discharge FIM score
was significantly affected by daily therapy dose as well as
the admission FIM score, OAI, and LOS (Table 4). However,
compared with the effect of admission FIM score (87.1%),
the effect of therapy dose was minor (3.6%).
Discussion
This retrospective analysis of rehabilitation outcome of the
KRW system revealed that OAI, LOS, therapy dose, and
discharge disposition have been significantly modified by
changes in the national insurance policy in Japan. Outcome
measures, including BI, FIM scores, LOS, and discharge
disposition have significantly changed year by year. The
OAI significantly decreased in 2006 because a new policy
encouraged earlier admissions. However, in patients with
stroke, LOS lengthened in recent years, probably because
policies encouraged the admission of more severely dis-
abled patients who needed a longer time to reach a plateau
before discharge. Increased therapy coverage as well as an
incentive for discharge home induced an improved rate
of home discharge, in spite of the increased number of
severely disabled patients. There was a time lag between
the policy change for increased coverage for rehabilitation
time and the resulting increase in actual rehabilitation time
because KRW hospitals needed to employ new thera-
pists to provide more therapy sessions. As of 2009, only
15.5% of patients were provided more than 15 hours of
therapy per week, but the rate is expected to increase in the
coming years.
Thus, national rehabilitation system policies are a major
determinant of inpatient rehabilitation outcomes. The effects,
546 Neurorehabilitation and Neural Repair 25(6)
however, are still being studied. For example, the overall
effect of providing a higher dose of therapy on outcome
measures is probably obscured by the considerable number
of new KRWs added to the survey every year. Most newly
recruited hospitals provide less organized rehabilitation than
those that are more experienced and organized. The quality
of rehabilitation care is probably still heterogeneous across
KRWs. We tested for a dose-dependent effect on functional
outcome in those hospitals that were consecutively involved
in the survey from 2006 to 2009. Stepwise multiple regres-
sion analysis revealed a small but significantly positive effect
of therapy dose on discharge FIM score. Although the effect
was small and initial functional status was the main predic-
tor of outcome,17 this observation is in accordance with sev-
eral studies reporting the effect of therapy dose on functional
outcome.18-20
In terms of type of disabling diseases, the proportion of
stroke decreased and that of orthopedic diseases increased
year by year (P < .001). This indicates that the total number
of beds for patients with stroke has sufficiently increased in
recent years to manage demand. The KRW association has
set the target number of KRW beds as 50 beds per 100 000
population based on an incidence of stroke in Japan of 200
per 100 000, with half of the patients needing transfer to
a KRW after acute therapy (100 per 100 000). Taking into
account other diseases, the target number of beds is 50 per
100 000. At present, prefectures range from 22 to 130 beds
per 100 000 population.
Limitations of the Study
More detailed data, including type of stroke, location of lesion,
and complications during the hospital stay, are not available.
We did not make these data mandatory because of the bur-
den on sites. A more precise description of patient features
may be necessary to evaluate who will most benefit from
this system and to improve the quality of the KRW orga-
nized rehabilitation system. FIM efficiency was not as high
as that of IRFs in the United States7 mainly because of later
admissions and longer stay.21 This property of KRWs may
cause less sensitivity to detect dose-dependent effects of hours
of therapy and types of therapies.
In 2010, a new incentive has been given to KRWs that
provide more than 14 hours of therapy weekly and also
weekend therapy. Wards that provide less than 280 minutes
of therapy per week were no longer eligible to be a KRW in
2010. These changes of policy may induce bottom-up changes
in quality and intensity of services.
Conclusions
The KRW is a Japanese solution to the increasing number of
elderly people who need inpatient rehabilitation after acutely
disabling diseases. Longitudinal data suggest that national
policy changes directly influence inpatient rehabilitation
services and outcomes, with better results for patients.
Consequences will be continuously monitored for future
reports.
Authors’ Note
Part of the material in the manuscript was presented at the 6th
World Congress of Neurorehabilitation; March 21, 2010; Vienna,
Austria.
Acknowledgments
We thank the other members of Annual Survey Committee of
Kaifukuki Rehabilitation Ward Association (Shinsuke Goto,
Toshio Ohta, Isao Ito, Shin-ichi Yamamoto, Yasuki Takizawa, and
Katsuyuki Kako) for their contribution to the survey. We also
thank all member hospitals of KRW association for submitting
precise data for the annual survey.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect
to the authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research and/or
authorship of this article.
References
  1. Indredavik B, Bakke F, Solberg R, Rokseth R, Haaheim LL,
Holme I. Benefit of a stroke unit: a randomized controlled trial.
Stroke. 1991;22:1026-1031.
  2. Jorgensen HS, Nakayama H, Raaschou HO, Larsen K,
Hubbe P, Olsen TS. The effect of a stroke unit: reductions
in mortality, discharge rate to nursing home, length of hospi-
tal stay, and cost. A community-based study. Stroke. 1995;26:
1178-1182.
  3. Indredavik B, Bakke F, Slordahl SA, Rokseth R, Haheim
LL. Treatment in a combined acute and rehabilitation
stroke unit: which aspects are most important? Stroke.
1999;30:917-923.
  4. Kalra L, Evans A, Perez I, Knapp M, Donaldson N, Swift CG.
Alternative strategies for stroke care: a prospective randomised
controlled trial. Lancet. 2000;356:894-899.
  5. Langhorne P, Duncan P. Does the organization of postacute
stroke care really matter? Stroke. 2001;32:268-274.
  6. Stroke Unit Trialists’ Collaboration. Organised inpatient (stroke
unit) care for stroke. Cochrane Database Syst Rev. 2007;(4):
CD000197.
  7. 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;292:1687-1695.
  8. Indredavik B, Fjaertoft H, Ekeberg G, Loge AD, Morch B. Benefit
of an extended stroke unit service with early supported discharge:
a randomized, controlled trial. Stroke. 2000;31: 2989-2994.
Miyai et al 547
  9. Thorsen AM, Holmqvist LW, de Pedro-Cuesta J, von Koch L.
A randomized controlled trial of early supported discharge
and continued rehabilitation at home after stroke: five-year
follow-up of patient outcome. Stroke. 2005;36:297-303.
10. Langhorne P, Taylor G, Murray G, et al. Early supported dis-
charge services for stroke patients: a meta-analysis of individual
patients’ data. Lancet. 2005;365:501-506.
11. Stuart M, Benvenuti F, Macko R, et al. Community-based adap-
tive physical activity program for chronic stroke: feasibility,
safety, and efficacy of the Empoli model. Neurorehabil Neural
Repair. 2009;23:726-734.
12. Yagura H, Miyai I, Suzuki T, Yanagihara T. Patients with
severe stroke benefit most by interdisciplinary rehabilitation
team approach. Cerebrovasc Dis. 2005;20:258-263.
13. Sonoda S, Saitoh E, Nagai S, Kawakita M, Kanada Y. Full-
time integrated treatment program, a new system for stroke
rehabilitation in Japan: comparison with conventional rehabili-
tation. Am J Phys Med Rehabil. 2004;83:88-93.
14. Yagura H, Miyai I, Seike Y, Suzuki T, Yanagihara T. Benefit
of inpatient multidisciplinary rehabilitation up to 1 year after
stroke. Arch Phys Med Rehabil. 2003;84:1687-1691.
15. Mahoney FI, Barthel DW. Functional evaluation: the Barthel
Index. Md State Med J. 1965;14:61-65.
16. Keith RA, Granger CV, Hamilton BB, Sherwin FS. The
Functional Independence Measure: a new tool for reha-
bilitation. In:Eisenberg MG, Grzesiak RC, eds. Advances
in Clinical Rehabilitation. Vol 2. New York, NY: Springer;
1987:6-18.
17. Shelton FD, Volpe BT, Reding M. Motor impairment as a predic-
tor of functional recovery and guide to rehabilitation treatment
after stroke. Neurorehabil Neural Repair. 2001;15: 229-237.
18. Kwakkel G, Wagenaar RC, Twisk JW, Lankhorst GJ, Koetsier JC.
Intensity of leg and arm training after primary middle-cerebral-
artery stroke: a randomised trial. Lancet. 1999;354:191-196.
19. Kwakkel G, van Peppen R, Wagenaar RC, et al. Effects of
augmented exercise therapy time after stroke: a meta-analysis.
Stroke. 2004;35:2529-2539.
20. Langhorne P, Coupar F, Pollock A. Motor recovery after
stroke: a systematic review. Lancet Neurol. 2009;8:741-754.
21. Miyai I, Suzuki T, Kii K, Kang J, Kajiura I. Functional outcome
of multidisciplinary rehabilitation in chronic stroke. J Neurol
Rehabil. 1998;12:95-99.