Buhagiar 2019 Oi 190124

Original Investigation | Physical Medicine and Rehabilitation
Assessment of Outcomes of Inpatient or Clinic-Based

vs Home-Based Rehabilitation After Total Knee Arthroplasty
A Systematic Review and Meta-analysis
Mark A. Buhagiar, PhD, MHM, BAppSc; Justine M. Naylor, PhD, BAppSc; Ian A. Harris, MBBS, MMed, PhD, FRACS, FAHMS; Wei Xuan, PhD;
Sam Adie, BSc, MBBS, MSpMed, MPH, PhD, FRACS; Adriane Lewin, MSc
Abstract Key Points

Question Is inpatient or clinic-based
IMPORTANCE Recent publication of the largest trials to date investigating rehabilitation after total
rehabilitation associated with superior
knee arthroplasty (TKA) necessitate an updated evidence review.
outcomes after total knee arthroplasty
compared with home programs?
OBJECTIVE To determine whether inpatient or clinic-based rehabilitation is associated with superior
function and pain outcomes after TKA compared with any home-based program. Findings This systematic review and
meta-analysis included 5 unique studies
DATA SOURCES MEDLINE, Embase, CINAHL, and PubMed were searched from inception to involving 752 unique participants
November 5, 2018. Search terms included knee arthroplasty, randomized controlled trial, comparing clinic- and home-based
physiotherapy, and rehabilitation. rehabilitation and 1 study comparing
inpatient rehabilitation with a home-
STUDY SELECTION Published randomized clinical trials of adults who underwent primary unilateral based program. Based on low- to
TKA and commenced rehabilitation within 6 postoperative weeks in which those receiving postacute moderate-quality evidence, no
inpatient or clinic-based rehabilitation were compared with those receiving a home- associations between settings, no
based program. clinically important differences for
mobility or patient-reported pain and
DATA EXTRACTION AND SYNTHESIS Two reviewers extracted data independently and assessed function at 10 and 52 postoperative
data quality and validity according to the PRISMA guidelines. Data were pooled using a random- weeks, and no significant differences in
effects model. Data were analyzed from June 1, 2015, through June 4, 2018. quality of life or range of motion
were found.
MAIN OUTCOMES AND MEASURES Primary outcomes were mobility (6-minute walk test [6MWT])
Meaning For adults who underwent
and patient-reported pain and function (Oxford knee score [OKS] or Western Ontario and McMaster
total knee arthroplasty, clinic or
Universities Osteoarthritis Index) reported at 10 to 12 postoperative weeks. The GRADE assessment
inpatient vs home-based rehabilitation
(Grading of Recommendations, Assessment, Development, and Evaluation) was applied to the
appeared to offer no clinically important
primary outcomes.
advantages.
RESULTS Five unique studies involving 752 unique participants (451 [60%] female; mean [SD] age,
68.3 [8.5] years) compared clinic- and home-based rehabilitation, and 1 study involving 165 + Supplemental content
participants (112 [68%] female; mean [SD] age, 66.9 [8.0] years) compared inpatient and home- Author affiliations and article information are
based rehabilitation. Low-quality evidence showed no clinically important difference between clinic- listed at the end of this article.
and home-based programs for mobility at 10 weeks (6MWT favoring home program; mean
difference [MD], −11.89 m [95% CI, −35.94 to 12.16 m]) and 52 weeks (6MWT favoring home
program; MD, −25.37 m [95% CI, −47.41 to −3.32 m]). Moderate-quality evidence showed no clinically
important difference between clinic- and home-based programs for patient-reported pain and
function at 10 weeks (OKS MD, −0.15 [95% CI, −0.35 to 0.05]) and 52 weeks (OKS MD, 0.10 [95% CI,
−0.14 to 0.34]).
(continued)
Open Access. This is an open access article distributed under the terms of the CC-BY License.
JAMA Network Open. 2019;2(4):e192810. doi:10.1001/jamanetworkopen.2019.2810 (Reprinted) April 26, 2019 1/14
Downloaded From: https://jamanetwork.com/ on 11/23/2022

JAMA Network Open | Physical Medicine and Rehabilitation Inpatient or Clinic-Based vs Home-Based Rehabilitation After Total Knee Arthroplasty
Abstract (continued)
CONCLUSIONS AND RELEVANCE Based on low- to moderate-quality evidence, no superiority of

clinic-based or inpatient programs compared with home-based programs was found in the early
subacute period after TKA. This evidence suggests that home-based rehabilitation is an appropriate
first line of therapy after uncomplicated TKA for patients with adequate social supports.
JAMA Network Open. 2019;2(4):e192810. doi:10.1001/jamanetworkopen.2019.2810
Introduction
Total knee arthroplasty (TKA) was the most frequently performed inpatient operating room
procedure in the United States in 2012.1 From 2003 to 2012, the incidence of TKA increased from
145.4 to 223.0 per 100 000 population (a 4.9% mean annual increase), with the total number
performed in the United States projected to increase from 711 000 in 2011 to 3.48 million by 2030.2
Similarly in Australia, the incidence increased from 108.3 per 100 000 population in 2003 to 222.3
per 100 000 population in 2017, with more than 54 000 TKAs performed in 2017.3
The increased volume of surgery constitutes a significant burden on the acute health care
budget, but because the surgery is typically followed by a protracted rehabilitation period, the latter
can add significantly to the cost of care. Several studies4-6 describe a significant cost differential
among rehabilitation pathways involving inpatient rehabilitation after TKA, ranging from a 5-fold to a
26-fold cost differential between a rehabilitation pathway that included inpatient therapy and one
that did not despite no differences in outcomes between groups. Concern about the total episode-
of-care costs for arthroplasty, including the rehabilitation period, has led to the introduction of
bundled payments in the United States, consisting of a single bundled payment to health care
organizations for all services related to the TKA to 90 days after surgery.7,8 This payment approach
has subsequently driven health care providers to reconsider the use of the more expensive inpatient
rehabilitation pathways.7
Outside inpatient rehabilitation, the setting, cost, and modes of provision vary greatly when
rehabilitation is delivered in the community.6,9 Available options include one-to-one or group-based
interventions (land or water) and various iterations of home-based care, including domiciliary
programs (physiotherapy visits in the home), telerehabilitation, or more simple monitored (via
occasional clinic visits or telephone contact) or unmonitored home programs.9-13 Previous systematic
reviews of randomized clinical trials14,15 have concluded that no single setting—clinic- or home-
based, in water or on land—appears to be associated with better recovery across a range of
outcomes. Despite this finding, to date, no evidence-based clinical practice guideline exists to
promote the use of home-based programs after uncomplicated TKA. Trials published since the
aforementioned reviews,16-18 however, have included new comparisons (inpatient and 3-arm trials)
and constitute the largest TKA rehabilitation trials to date. Thus, a more contemporary review is
warranted, potentially as a precursor to development of a much-needed clinical practice guideline.
The aim of this systematic review and meta-analysis was to investigate the importance of the
rehabilitation setting on outcomes for adults after elective, primary, unilateral TKA. Specifically, we
aimed to determine whether inpatient or clinic-based rehabilitation is associated with superior
function and pain outcomes after TKA compared with any home-based physiotherapy program
(monitored or unmonitored, or domiciliary [physiotherapy home visitation]). Superiority was defined
as a change considered to be clinically important for each outcome assessed.
Methods
This systematic review of randomized clinical trials follows the methods described in the Cochrane
Handbook for Systematic Reviews of Interventions, Version 5.1.019 and is reported in accordance with
the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting

guideline.20 The protocol was updated to include the GRADE (Grading of Recommendations,
Assessment, Development and Evaluation) component for assessing quality of evidence.21
Eligibility Criteria
Types of Studies
Published randomized clinical trials were eligible for inclusion. We excluded studies reported only as
abstracts if adequate data could not be obtained from the authors, studies in which TKA data could
not be separated from other procedures (eg, total hip arthroplasty), and studies for which we were
unable to obtain potentially relevant data from the authors on request. No language restrictions
were applied.
Types of Participants
We included studies of adults (age ⱖ18 years) who had undergone a primary unilateral TKA and
commenced rehabilitation within 3 months of surgery. We excluded studies of unicompartmental
surgery, revision TKA, or TKA secondary to trauma.
Types of Interventions
We included studies investigating rehabilitation after TKA in which patients who had received
postacute inpatient or clinic-based rehabilitation were compared with others who had received a
monitored or an unmonitored home-based or domiciliary program after discharge from the acute-
care facility. We excluded telerehabilitation because other reviews in progress during conduct of our
study and subsequently published22-24 have investigated this option.
Outcomes
The goal of physiotherapy-based rehabilitation after TKA is to improve physical function, including
walking, activities of daily living, and knee mobility. We grouped outcomes into the following
categories that broadly reflect these goals: physical performance test results (6-minute walk test
[6MWT],25 measured as laps walked on a flat surface in 6 minutes; walking speed, stair ascent and
descent tests, and chair rise test), patient-reported pain and function (Oxford knee score26 [OKS;
range, 0-48, with higher scores indicating best outcomes], Knee Injury and Osteoarthritis Outcome
Score27 [range, 0-100, with higher scores indicating worse outcomes], and Western Ontario and
McMaster Universities Osteoarthritis Index28 [5 items for pain, 2 for stiffness, and 17 for functional
limitation, with higher scores indicating worse outcomes]), generic health-related quality-of-life
measures (12- and 36-Item Short Form Health Surveys and EuroQol-5D29), and knee range of motion
(ROM), expressed as active or passive ROM, extension, and/or flexion.
Primary and Secondary Outcomes

Primary outcomes were mobility (6MWT) and patient-reported pain and function (OKS) measured
at 10 to 12 postoperative weeks. Secondary outcomes included mobility and patient-reported pain
and function, knee ROM, postoperative complications, and health-related quality of life measured at
10, 26, and/or 52 weeks.
Identification and Selection of Studies

We searched Embase, PubMED, MEDLINE, and CINAHL from inception to June 19, 2018, using search
terms that included knee arthroplasty, randomized controlled trial, physiotherapy and related terms,
and rehabilitation (eMethods 1 in the Supplement). We later scanned references of all
included studies.
Two reviewers (M.A.B. and J.M.N.) independently screened titles and abstracts of the search
output to identify studies suitable for further scrutiny. They discussed inconsistencies in the
screening process before a decision was made to review the full text. The same reviewers then

screened full-text articles to determine inclusion in the review. Discrepancies in the final list were
discussed, and a consensus was reached for all articles.
Assessment of Study Quality, Risk of Bias, and GRADE Assessment

Two reviewers (M.A.B. and J.M.N.) independently assessed study quality using the Cochrane
Collaboration Risk of Bias tool,30 which includes the following variables: random allocation sequence,
allocation concealment, blinding (of patients, therapists, and outcome assessors), attrition (loss to
follow-up and intention-to-treat analysis), and selective outcome reporting. Disagreements in risk of
bias were resolved by discussion, or, when necessary, a third person arbitrated. Included studies
were also assessed using the Physiotherapy Evidence Database scale,31 used to identify trials that are
more likely to be valid and to contain sufficient information to guide clinical practice.
Two reviewers (S.A. and A.L.) independently used the GRADE component to categorize the
quality and strength of the evidence as high, moderate, low, and very low for the 6MWT and patient-
reported pain and function at 10 to 12 postoperative weeks (the primary outcomes) and at 52
weeks.32 Disagreements were resolved by consensus between the 2 reviewers. To ensure
reproducibility and consistency, the reviewers used a checklist to rate each component of the GRADE
assessment.33 We used GRADEpro software to create summary of findings tables.34 Because 4 of
the investigators (M.A.B., J.M.N., I.A.H., and W.X.) were involved in randomized clinical trials relevant
to this review, the GRADE assessment was undertaken by reviewers not involved in any of the
included studies.
Data Extraction
Four reviewers (M.A.B., J.M.N., S.A., and A.L.) independently extracted data. We collected data
related to participants (diagnosis, age, sex, and body mass index); country; study dates; inclusion and
exclusion criteria; setting, timing, duration, and intensity of the intervention and comparison
(control) conditions; duration of follow-up; losses to follow-up and reasons; and outcomes. Means
(SDs) were extracted for outcomes reported as continuous variables. Proportions were extracted for
categorical outcomes. Appropriate conversions were applied when outcomes were reported as
medians and interquartile ranges or means and 95% CIs.19
For studies with incomplete data, we attempted to contact the corresponding author. We also
asked whether any outcomes not reported in their publications had been collected. When authors of
included studies were unable to provide additional data, all available data were included in the
review. If data had been provided by authors to other reviewers in published reviews, these were
included in the analyses in the case of failure to retrieve data from the primary source and
acknowledged appropriately. Authors of included studies were also contacted when there was
incomplete reporting of data. Where possible, we used data from intention-to-treat analyses in our
calculations to determine between-group differences.
For studies with 3 randomized arms, we adopted a strategy described by Higgins and Green19
of including each comparison separately but with the shared intervention group divided evenly
among the comparisons. For continuous outcomes, only the total number of participants was
divided, with the means (SDs) left unchanged.
Statistical Analysis
Measures of Treatment Effect
Data were analyzed from June 1, 2015, through June 4, 2018. We used the mean differences (MDs)
and 95% CIs for continuous outcomes with the same units (eg, 6MWT). We presented continuous
outcomes with different units as standardized MDs and 95% CIs. Categorical outcomes were
expressed as a number with percentage.

Data Synthesis
The 2 main comparisons (clinic- vs home-based and inpatient vs home-based) were considered
separately. Where possible, we pooled data using random-effects meta-analysis.35 Because the
standardized MD can be artifactually affected by correlation between baseline and follow-up
measurements when including the SD of change along with the SD of absolute values, we used the
baseline SD for change score values when combining change scores and absolute values. For
dichotomous outcomes, we used a pooled odds ratio.
We used the I2 statistic to assess statistical heterogeneity among included studies. We planned
to explore publication bias using funnel plots if we had a minimum of 10 included studies, but the
number found did not reach this. We used RevMan software (version 5.3)36 to compile data and
perform statistical analyses.
Results
Results of the Search
The search strategy yielded 2286 references. After duplicates were removed using the duplicate
removal program within EndNote commercial reference management software (Clarivate Analytics)
and titles and abstracts were screened, we retrieved 15 studies for evaluation, of which 9 studies
were excluded (eMethods 2 in the Supplement). Six eligible randomized clinical trials were included
in the review (Table 1),16,17,37-40 5 of which were included in the meta-analysis (Figure 1).17,37-40
Included Studies
Five unique studies17,37-40 with a total of 752 unique participants (451 [60%] female; mean [SD] age,
68.3 [8.5] years) included in the meta-analysis compared outpatient rehabilitation (individual and/or
group) with home-based rehabilitation (monitored or unmonitored). The sixth study16 with 165
patients (112 [68%] female; mean [SD] age, 66.9 [8.0] years) compared inpatient rehabilitation with
home-based rehabilitation monitored by a health care professional. This study was included in a
qualitative synthesis only. In all studies, rehabilitation commenced within 3 months of surgery, and
participants were followed up for 26 to 52 weeks. Among studies that reported diagnostic data, the
most common diagnosis was osteoarthritis. The mean (SD) age of study patients ranged from 66.2
(8.2) to 70.9 (SD not provided) years among studies reporting age. All studies excluded patients with
complications in the acute postoperative period. Four studies reported patient adherence to the
program (88%16; 77%17; 96%37; and 61%38) (Table 1). One study17 included in this meta-analysis had
3 randomized arms. We included each comparison separately (in this case, outpatient group–based
vs home-based rehabilitation and outpatient one-to-one therapy vs home-based rehabilitation) but
with the shared intervention group (ie, the home-based treatment arm) divided evenly among the
comparisons as previously described in the data extraction section.19
Risk of Bias in Included Studies

As shown in Figure 2, 5 studies16,17,38-40 used adequate methods for generating the randomization
sequence, with the sixth study37 not providing this information. Four studies16,17,38,39 described the
use of adequate methods to conceal allocation. Blinding of participants and therapists was not
possible in any of the studies owing to the nature of the intervention, but all included studies blinded
assessors of objective outcomes to group allocation. Two studies16,17 were free of selective outcome
reporting. Risk of bias was present in 2 studies37,38 owing to uneven losses to follow up.
Physiotherapy Evidence Database scores assessing study quality ranged from 5 to 8 (maximum of 10)
(eTable 1 in the Supplement). We did not use funnel plots to explore publication bias owing to the
small number of included studies.

Intervention Outcomes and Comparisons

The results for all outcomes and comparisons of clinic-based rehabilitation with monitored or
unmonitored home-based programs are summarized in eTable 2 in the Supplement. Because only 1
study assessed inpatient rehabilitation,16 meta-analysis was not possible, and a brief narrative
summary is provided.
Primary Outcomes
Mobility
Two studies17,37 reported the 6MWT at 1 or more follow-up points (eTable 2 and eFigure 1 in the
Supplement). Participants who received clinic-based rehabilitation (n = 231) had walked an MD in
6MWT of −11.89 m (95% CI, −35.94 to 12.16 m; P = .33) compared with those who received a home-
based program (n = 142) at 10 to 12 weeks and an MD in 6MWT of −3.05 m (95% CI, −29.75 to 23.66
m; P = .82) compared with those receiving a home-based program in both studies with 243
participants at 26 weeks. At 52 weeks, participants who had undergone clinic-based rehabilitation
had walked an MD in 6MWT of −25.37 m (95% CI, −47.41 to −3.32 m; P = .02) compared with
participants who had undergone clinic-based rehabilitation. Based on GRADE assessment,
Table 1. Summary of Included Studies

Source No. of Primary Focus of Follow-up
(Country) Participants Diagnosis Intervention Setting Intervention Condition Control Condition Outcome Assessment Point, wk
Buhagiar 165 TKA and Simple and advanced Inpatient Inpatient rehabilitation Home: 2-3 OP PT 6MWT, 10MWT, OKS, 10, 26,
et al,16 osteoarthritis functional, aerobic, rehabilitation and home: 10 d of twice- sessions for 10 wk, knee ROM ≥100°, and 52
2017 and strengthening and home-based daily inpatient PT; 2-3 starting 2-3 wk after KOOS, EuroQol-5D, PO
(Australia) exercises groups OP physiotherapy surgery complications
sessions for 10 wk,
starting 2-3 wk after
surgery
Ko 249 TKA and Simple and advanced 1:1:1 Clinic and group clinic: 2 Home: 2 OP PT visits OKS, WOMAC 2, 10, 26,
et al,17 osteoarthritis functional, aerobic, Randomized OP PT sessions per week with follow-up function, knee ROM, and 52
2013 and strengthening clinic-, group for 6 wk, starting 2-3 wk telephone call for 6 wk, 6MWT, timed stairs
(Australia) exercises clinic–, and after surgery starting 2-3 wk after ascent and descent,
home-based surgery SF-12 physical and
groups mental scores, PO
complications
Kramer 160 TKA and Simple and advanced Clinic- and Clinic: two 1-h OP PT Home: monitored via 2 WOMAC, SF-36, Knee 12, 26,
et al,37 osteoarthritis strengthening and home-based sessions per week for 10 telephone calls between Society Scale, timed and 52
2003 ROM exercises groups wk, starting 1 wk after 2-12 wk after surgery stair ascent and
(Canada) surgery. Home: exercise descent, 6MWT
program upgraded by
treating therapist
Madsen 80 TKA and Strengthening, Group clinic- Group clinic: 2 PT Home: 2 OP PT visits in OKS, SF-36 physical 12 and 26
et al,38 osteoarthritis endurance, and home-based sessions per week for 6 total, with additional function, EuroQol-5D,
2013 functional, and ROM groups wk, starting 4-8 wk after OP visits allowed (not knee ROM, peak leg
(Denmark) exercises surgery. Strength and exceeding 12) for extensor power,
endurance training and participants with balance test, 10MWT,
patient education and physical limitations sit-to-stand tests, VAS
discussion. Home: pain during leg
exercises twice weekly extensor power test
with strength training,
endurance training on
exercise bike, walking,
and balance training
Mockford 143 TKA and Functional, Clinic- and Clinic: home exercise Given home exercise OKS, Bartlett Patellar 12 and 52
et al,39 osteoarthritis strengthening, and home-based regime to follow on regime to follow on Score, SF-12, PO
2008 or ROM exercises groups discharge; PT sessions discharge, with no OP complications
(Northern rheumatoid for 6 wk, starting within PT
Ireland) arthritis 3 wk of hospital
discharge
Rajan 120 TKA and No information on OP clinic vs Clinic: PT sessions Home: No information ROM 12, 26,
et al,40 monoarticular primary focus of unmonitored (mean, 4-6) after on program. Patients and 52
2004 arthrosis intervention home-based discharge from hospital. from both groups given
(England) group No information on a home exercise regime
program content. on discharge
Abbreviations: KOOS, Knee Injury and Osteoarthritis Outcome Score; OKS, Oxford knee total knee arthroplasty; VAS, visual analog scale; WOMAC, Western Ontario and
score; OP, outpatient; PO, postoperative; PT, physiotherapy; ROM, range of motion; McMaster Universities Osteoarthritis Index; 6MWT, 6-minute walk test; 10MWT,
SF-12, 12-Item Short Form Health Survey; SF-36, 36-Item Short Form Health Survey; TKA, 10-minute walk test.

low-quality evidence suggests that there may be no clinically important difference between clinic-
and home-based programs for mobility at 10 and 52 weeks (Table 2). Minimal heterogeneity was
found across studies reporting mobility outcomes.
Patient-Reported Pain and Function

Three studies17,38,39 reported a pain and function outcome at 1 or more follow-up points using the
OKS (absolute values or change from baseline). Based on the GRADE component, moderate-quality
evidence suggests little or no difference between clinic- and home-based programs for patient-
reported pain and function in 457 patients at 10 weeks (MD, −0.15; 95% CI, −0.35 to 0.05) and in 388
patients at 52 weeks (MD, 0.10; 95% CI, −0.14 to 0.34) (Table 2 and eFigure 2 in the Supplement).
Secondary Outcomes
Patient-Reported Quality of Life
Two studies17,38 reported quality-of-life outcomes at 1 or more follow-up points using the 12- or
36-Item Short Form Health Survey. No superiority of outcomes was found for patients receiving
Figure 1. Flow of Studies Through the Review
2286 Records identified through database search and other sources
404 Duplicates removed
1882 Abstracts screened
1867 Abstracts excluded
15 Full-text articles assessed for eligibility
9 Full-text articles excludeda

2 Research design not RCT
5 Intervention not comparing inpatient
or clinic-based vs monitored or
unmonitored home setting
1 Participants underwent procedures other
than TKA
1 Results of combined patient groups
(TKA and THA) unable to be adequately RCT indicates randomized clinical trial; THA, total hip
separated or distinguished
arthroplalsty; and TKA, total knee arthroplasty.
a
Studies may have been excluded for failing to meet
6 Articles included in review
more than 1 inclusion criterion.
Figure 2. Cochrane Risk of Bias Table
Risk of bias
Low High Unclear
Buhagiar et al, 201716

Ko et al, 201317
Kramer et al, 200337
Madsen et al, 201338
Mockford et al, 200839
Rajan et al, 200440
Random Allocation Blinding of Blinding of Incomplete Selective Other Bias

Sequence Concealment Participants Outcome Outcome Reporting
Generation (Selection and Personnel Assessment Data (Reporting
(Selection Bias) (Performance (Detection (Attrition Bias) Study quality was assessed using the Cochrane
Bias) Bias) Bias) Bias) Collaboration Risk of Bias Tool.30

clinic-based rehabilitation compared with those who received a monitored or unmonitored home-
based program in 314 participants at 10 to 12 weeks, 313 participants at 26 weeks, and 242
participants at 52 weeks (eFigure 3 in the Supplement). Less than 50% heterogeneity was found
across studies reporting quality-of-life outcomes.
Active ROM
Active ROM flexion data suitable for meta-analysis were available from 3 studies37,39,40 with 386
participants and for active ROM extension from 1 study39 with 143 participants. No benefit was seen
for these outcomes at any point (eFigure 4 in the Supplement). Greater heterogeneity was found
across studies reporting ROM outcomes.
Passive ROM
Two studies17,38 with 314 participants reported passive ROM data suitable for meta-analysis
(eFigure 5 in the Supplement). No superiority of outcomes in passive ROM was found between
randomized groups at 10 to 12, 26, or 52 weeks after surgery.
Inpatient Rehabilitation vs Home-Based Rehabilitation

A single study16 compared inpatient with home-based rehabilitation. In an intention-to-treat
analysis, the authors reported no difference in the primary outcome of the 6MWT between the 2
randomized groups. A per protocol analysis of the primary outcome yielded similar results. The
unadjusted and adjusted group effects were nonsignificant for all secondary outcomes at 10, 26, and
52 weeks (OKS, knee ROM ⱖ100°, 10MWT, Knee Injury and Osteoarthritis Outcome Score, and
EuroQol-5D score). Per protocol analyses yielded the same results across all points. Moderate-quality
evidence suggested that inpatient rehabilitation was not associated with superior mobility and
Table 2. GRADE Component for Clinic-Based Compared With Home-Based Rehabilitation for Total Knee
Arthroplasty at 10 and 52 Weeks
No. of Certainty of the Anticipated Absolute Effectsb

Participants Evidence, Mean Value With
(No. of RCTs) GRADE Home-Based MD With Clinic-Based
Outcomes in Follow-up Componenta Rehabilitation Rehabilitation (95% CI)
At 10- to 12-wk
follow-up
Mobility assessed 373 (3) Lowc,d 371.0 m −11.89 m
with 6MWT (−35.94 to 12.16 m)
Pain and function 457 (4) Moderatec,f NE −0.15 points
assessed with (−0.35 to 0.05 points)g
OKSe
At 52-wk follow-up
Mobility assessed 369 (3) Lowc,d 414.6 m −25.37 m
with 6MWT (−47.41 to −3.32 m)
Pain and function 388 (3) Moderatec,f NE 0.10 points
assessed with (−0.14 to 0.34 points)g
OKSe
Abbreviations: GRADE, Grading of Recommendations, Assessment, Development and Evaluation; MD, mean difference;
NE, not estimable; OKS, Oxford knee score; RCT, randomized clinical trial; 6MWT, 6-minute walk test.
a
High indicates very confident that the true effect lies close to that of the estimate of the effect; moderate, moderately
confident that the true effect is likely to be close to the estimate of the effect, but with a possibility that it is substantially
different; low, limited confidence and the true effect may be substantially different from the estimate of the effect; and
very low, very little confidence and the true effect is likely to be substantially different from estimate of effect.
b
The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative
effect of the intervention (and its 95% CI).
c
Differential loss to follow-up in 1 study: 19% (clinic-based group) and 28% (home-based group).
d
Total sample size was less than 400 (optimal information size for continuous outcomes).
e
Range of possible scores was 0 to 48, with higher scores indicating best outcomes.
f
Outcome assessment was not described in 1 study.
g
Standardized MD.

patient-reported pain and functional outcomes compared with the outcomes from a monitored
home program; however, this result must be interpreted cautiously because it is based on a single
study (Table 3).
Discussion
Summary of Main Findings
This systematic review and meta-analysis found that, based on low- to moderate-quality evidence,
clinic-based rehabilitation after TKA was not associated with superior outcomes compared with a
home-based program, whether monitored or unmonitored, when considering mobility, pain,
function, quality of life, active knee flexion and extension, and passive knee ROM. Similarly, inpatient
rehabilitation after TKA does not deliver superior outcomes compared with monitored home-based
rehabilitation when considering mobility, pain, function, quality of life, and knee flexion.
Home-based rehabilitation provided greater mobility (approximately 25 m more in the 6MWT)
at 52 weeks compared with a clinic-based program. However, research indicates that this difference
is not clinically important. Using a triangulation of methods, including patient-perceived anchor-
based thresholds and distribution-based thresholds, Naylor and colleagues41 proposed that the
threshold for minimal or more improvement for the 6MWT after TKA is expected to range from 26 to
55 m. For patients with chronic obstructive pulmonary disease, Rasekaba and colleagues42
determined the minimal clinically important distance for the 6MWT is 54 m, with a similar figure (50
m) determined for a population of older adults and those with stroke by Perera et al.43
A single study provided evidence that inpatient rehabilitation is not associated with better
mobility and patient-reported pain and function outcomes compared with a monitored home-based
program among adults undergoing uncomplicated TKA.16 This study reported that inpatient
rehabilitation was associated with higher levels of patient-reported satisfaction. Understanding the
reason for this finding would be useful for informing alternative models of rehabilitation provision.
Table 3. GRADE Component for Inpatient Compared With Home-Based Rehabilitation for Total Knee
Arthroplasty at 10 and 52 Weeks
No. of Certainty of the Anticipated Absolute Effectsb

Participants Evidence, Mean Value With
(No. of RCTs) GRADE Home-Based MD With Inpatient
Outcomes in Follow-up Componenta Rehabilitation Rehabilitation (95% CI)
At 10- to 12-wk
follow-up
Mobility assessed 158 (1) Moderatec 383.2 m 3.6 m
with 6MWT (−23.2 to 30.4 m)
Pain and function 157 (1) Moderatec 32.1 points 1.21 points
assessed with (−1.45 to 3.88 points)
OKSd
At 52-wk follow-up
Mobility assessed 150 (1) Moderatec 404.8 m −13.5 m
with 6MWT (−40.7 to 13.6 m)
Pain and function 160 (1) Moderatec 37.0 points −0.55 points
assessed with (−3.21 to 2.1 points)
OKSd
Abbreviations: GRADE, Grading of Recommendations, Assessment, Development and Evaluation; MD, mean difference;
OKS, Oxford knee score; RCT, randomized clinical trial; 6MWT, 6-minute walk test.
a
High indicates very confident that the true effect lies close to that of the estimate of the effect; moderate, moderately
confident that the true effect is likely to be close to the estimate of the effect, but with a possibility that it is substantially
different; low, limited confidence and the true effect may be substantially different from the estimate of the effect; and
very low, very little confidence and the true effect is likely to be substantially different from estimate of effect.
b
The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative
effect of the intervention (and its 95% CI).
c
Only 1 study and a small sample size.
d
Range of possible scores was 0 to 48, with higher scores indicating best outcomes.

No studies included in this review considered whether outcomes of post-TKA rehabilitation

delivered in the domiciliary setting differed from those in other rehabilitation settings. One study
comparing inpatient with domiciliary rehabilitation44 combined data from recipients of total hip and
knee arthroplasty and concluded that the combined cohort had no difference in pain, functional
outcomes, or patient satisfaction between the 2 treatment groups and that inpatient rehabilitation
was not cost-effective. We were not able to obtain individual joint data from the authors, so were not
able to include these data in our meta-analysis.
We were also unable to include the largest randomized clinical trial conducted to date
concerning rehabilitation after TKA (n = 390).18 This study compared usual care with a home-based
exercise program; however, usual care consisted of any combination of clinic- or inpatient-based
programs, and many in the home-based program also accessed clinic-based care. Thus, we were
unable to assign their participants to exclusively home-based or facility-based care. Those authors
concluded that a home-based exercise program was not inferior to usual care for a range of patient-
reported and objectively measured outcomes, including the Western Ontario and McMaster
Universities Osteoarthritis Index, walking speed, and knee ROM.
Comparison With Other Reviews

A systematic review and meta-analysis14 published in 2015 examined the effectiveness of
physiotherapy exercise after TKA and found no differences for outpatient compared with home-
based physiotherapy exercise for physical function or pain outcomes. A short-term benefit that
favored home-based physiotherapy exercise for ROM flexion was not clinically important. These
findings are consistent with those of our review.
Quality of the Evidence

The risk of bias in the 5 studies included in the review was variable. The primary source of potential
bias was from uneven losses to follow-up in 2 studies.37,38 Another potential source of bias, because
of the nature of the intervention, was that participants could not be blinded to their treatment.
Future Considerations
We identified a number of ongoing randomized trials comparing rehabilitation settings after TKA for
future consideration. One, identified via a search in ClinicalTrials.gov,45 plans to evaluate
unsupervised home exercise with and without a web-based recovery platform compared with
traditional outpatient physiotherapy after TKA. Another, with a published protocol46 and feasibility
study,47 will compare clinic-based group physiotherapy with usual (home-based) care. However, the
method proposed suggests possible crossover between settings in the latter arm because some
patients were referred to physiotherapy services on an individual basis at the discretion of the
hospital physiotherapist, orthopedic team, or general physician, with no indication of how many such
referrals were made. We also updated our search to November 2018, with no new eligible trials
identified.
Strengths and Limitations

This review has specific strengths. We included only studies in which treatment assignments were
randomized, enhancing the strength of the conclusions that could be drawn from the findings. Our
review was also comprehensive because we included non–English-language articles in the search
strategy, although none were found to be suitable for inclusion.
This review also has several limitations. First, the failure to identify all relevant studies is a
common source of bias in systematic reviews. We conducted thorough searches of research
databases as well as clinical trial registries, including studies in all languages, using reference list
searches of included studies and forward citation tracking, and corresponded with authors of
included studies. We identified 1 study that was reported only in a conference proceeding48 and
compared clinic-based rehabilitation with an unmonitored home-based program. However, that

study had significant risk of bias, including failure to mask outcome assessors and incomplete
outcome data, and data obtained from the author were insufficient for inclusion in the analysis.
Despite these efforts, failing to identify unpublished studies may have introduced bias.
Second, the number of studies in this area of research is small, limiting the precision of the
findings and influencing the ability to assess for publication bias graphically or statistically. Third, the
method used to divide data from the 3-arm study17 only partially overcomes unit-of-analysis error
because the resulting comparisons remain correlated,19 meaning that the narrowness of the 95% CIs
may have been overestimated. Also, we did not consider complications occurring in the subacute
care period as part of this review because of uncertainty around the standardization of reporting in
this area. However, our conclusions apply to patients who did not experience major complications
during the acute-care period that would have prohibited their involvement in the programs
prescribed.
Conclusions
Several clinical trials have investigated the influence of setting on the effectiveness of rehabilitation
delivered in the early subacute phase after TKA. This review found consistent evidence suggesting no
clinically important differences between clinic-based or inpatient rehabilitation after TKA compared
with a home-based program across a range of outcomes. Care that aligns with this evidence would
incorporate home-based rehabilitation as the first line of therapy, reserving the more intensely
supervised approaches for the most impaired patients or those without adequate social supports. In
our view, the development of an evidence-based clinical practice guideline appears to be the next
step in synthesizing this literature and aligning practice with the most up-to-date evidence.
ARTICLE INFORMATION
Accepted for Publication: March 3, 2019.
Published: April 26, 2019. doi:10.1001/jamanetworkopen.2019.2810
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Buhagiar MA
et al. JAMA Network Open.
Corresponding Author: Mark A. Buhagiar, PhD, MHM, BAppSc, Catholic Diocese of Parramatta, Level 2, 12 Victoria
Rd, Parramatta, New South Wales 2150, Australia (mmmbuhagiar@hotmail.com).
Author Affiliations: Catholic Diocese of Parramatta, Parramatta, Australia (Buhagiar); South West Sydney Clinical
School, University of New South Wales, Sydney, Australia (Buhagiar, Naylor, Harris, Xuan, Adie, Lewin); Whitlam
Orthopaedic Research Centre, Liverpool, Australia (Naylor, Harris); South Western Sydney Local Health District,
Sydney, Australia (Naylor, Harris, Lewin); Ingham Institute of Applied Medical Research, Liverpool,
Australia (Xuan).
Author Contributions: Dr Buhagiar had full access to all the data in the study and takes responsibility for the
integrity of the data and the accuracy of the data analysis.
Concept and design: Buhagiar, Naylor, Harris, Xuan.
Acquisition, analysis, or interpretation of data: Buhagiar, Naylor, Xuan, Adie, Lewin.
Drafting of the manuscript: Buhagiar, Naylor, Lewin.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Buhagiar, Xuan, Adie.
Administrative, technical, or material support: Buhagiar, Naylor, Adie, Lewin.
Supervision: Naylor, Harris, Adie, Lewin.
Conflict of Interest Disclosures: Drs Naylor and Harris reported receiving grants from the HCF Research
Foundation outside the submitted work. Dr Adie reported receiving unrestricted grants from the National Health
and Medical Research Council outside the submitted work. No other disclosures were reported.

Additional Contributions: Vicki Ko, PhD, BAppSc(PT), Macquarie University, Sydney, Australia, provided
supplementary information and additional data for the meta-analyses. She did not receive any compensation in
association with her contribution.
REFERENCES
1. Fingar KR, Stocks C, Weiss AJ, et al. Most frequent operating room procedures performed in US hospitals, 2003-
2012. https://www.hcup-us.ahrq.gov/reports/statbriefs/sb186-Operating-Room-Procedures-United-States-2012.pdf.
Published December 2014. Accessed February 27, 2019.
2. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the
United States from 2005 to 2030. J Bone Joint Surg Am. 2007;89(4):780-785.
3. Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR). Demographics of Hip,
Knee and Shoulder Arthroplasty Supplementary Report 2018. South Australia, Australia: Australian Orthopaedic
Association; 2018.
4. Lavernia CJ, D’Apuzzo MR, Hernandez VH, Lee DJ, Rossi MD. Postdischarge costs in arthroplasty surgery.
J Arthroplasty. 2006;21(6)(suppl 2):144-150.
5. Tribe KL, Lapsley HM, Cross MJ, Courtenay BG, Brooks PM, March LM. Selection of patients for inpatient
rehabilitation or direct home discharge following total joint replacement surgery: a comparison of health status
and out-of-pocket expenditure of patients undergoing hip and knee arthroplasty for osteoarthritis. Chronic Illn.
2005;1(4):289-302.
6. Naylor JM, Hart A, Mittal R, Harris I, Xuan W. The value of inpatient rehabilitation after uncomplicated knee
arthroplasty: a propensity score analysis. Med J Aust. 2017;207(6):250-255.
7. Finkelstein A, Ji Y, Mahoney N, Skinner J. Mandatory Medicare bundled payment program for lower extremity
joint replacement and discharge to institutional postacute care: interim analysis of the first year of a 5-year
randomized trial. JAMA. 2018;320(9):892-900.
8. Piccinin MA, Sayeed Z, Kozlowski R, Bobba V, Knesek D, Frush T. Bundle payment for musculoskeletal care:
current evidence (part 1). Orthop Clin North Am. 2018;49(2):135-146.
9. Naylor J, Harmer A, Fransen M, Crosbie J, Innes L. Status of physiotherapy rehabilitation after total knee
replacement in Australia. Physiother Res Int. 2006;11(1):35-47.
10. Peter WF, Tilbury C, Verdegaal SHM, et al. The provision of preoperative and postoperative physical therapy in
elderly people with hip and knee osteoarthritis undergoing primary joint replacement surgery. Curr Orthop Pract.
2016;27(2):173-183.
11. Okoro T, Ramavath A, Howarth J, et al. What does standard rehabilitation practice after total hip replacement
in the UK entail? results of a mixed methods study. BMC Musculoskelet Disord. 2013;14:91.
12. Artz N, Dixon S, Wylde V, Beswick A, Blom A, Gooberman-Hill R. Physiotherapy provision following discharge
after total hip and total knee replacement: a survey of current practice at high-volume NHS hospitals in England
and Wales. Musculoskeletal Care. 2013;11(1):31-38.
13. Moffet H, Tousignant M, Nadeau S, et al. In-home telerehabilitation compared with face-to-face rehabilitation
after total knee arthroplasty: a noninferiority randomized controlled trial. J Bone Joint Surg Am. 2015;97(14):
1129-1141.
14. Artz N, Elvers KT, Lowe CM, Sackley C, Jepson P, Beswick AD. Effectiveness of physiotherapy exercise
following total knee replacement: systematic review and meta-analysis. BMC Musculoskelet Disord. 2015;16:15.
15. Minns Lowe CJ, Barker KL, Dewey M, Sackley CM. Effectiveness of physiotherapy exercise after knee
arthroplasty for osteoarthritis: systematic review and meta-analysis of randomised controlled trials. BMJ. 2007;
335(7624):812.
16. Buhagiar MA, Naylor JM, Harris IA, et al. Effect of inpatient rehabilitation vs a monitored home-based program
on mobility in patients with total knee arthroplasty: the HIHO randomized clinical trial. JAMA. 2017;317(10):
1037-1046.
17. Ko V, Naylor J, Harris I, Crosbie J, Yeo A, Mittal R. One-to-one therapy is not superior to group or home-based
therapy after total knee arthroplasty: a randomized, superiority trial. J Bone Joint Surg Am. 2013;95(21):
1942-1949.
18. Han AS, Nairn L, Harmer AR, et al. Early rehabilitation after total knee replacement surgery: a multicenter,
noninferiority, randomized clinical trial comparing a home exercise program with usual outpatient care. Arthritis
Care Res (Hoboken). 2015;67(2):196-202.
19. Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions, Version 5.1.0. http://
handbook-5-1.cochrane.org/. Updated March 2011. Accessed March 27, 2019.

20. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-
analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009;6(7):
e1000100. Accessed February 27, 2019.
21. Brozek JL, Akl EA, Alonso-Coello P, et al; GRADE Working Group. Grading quality of evidence and strength of
recommendations in clinical practice guidelines, part 1 of 3: an overview of the GRADE approach and grading
quality of evidence about interventions. Allergy. 2009;64(5):669-677.
22. Panda S, Bali S, Kirubakaran R, Hagenberg A. Telerehabilitation and total knee arthroplasty: a systematic
review and meta-analysis of randomized controlled trials. Int J Ther Rehabil. 2015;22(suppl 8):S6.
23. Cottrell MA, Galea OA, O’Leary SP, Hill AJ, Russell TG. Real-time telerehabilitation for the treatment of
musculoskeletal conditions is effective and comparable to standard practice: a systematic review and meta-
analysis. Clin Rehabil. 2017;31(5):625-638.
24. Shukla H, Nair SR, Thakker D. Role of telerehabilitation in patients following total knee arthroplasty: evidence
from a systematic literature review and meta-analysis. J Telemed Telecare. 2017;23(2):339-346.
25. Troosters T, Gosselink R, Decramer M. Six-minute walk test: a valuable test, when properly standardized. Phys
Ther. 2002;82(8):826-827.
26. Dawson J, Fitzpatrick R, Murray D, Carr A. Questionnaire on the perceptions of patients about total knee
replacement. J Bone Joint Surg Br. 1998;80(1):63-69.
27. Roos EM, Lohmander LS. The Knee injury and Osteoarthritis Outcome Score (KOOS): from joint injury to
osteoarthritis. Health Qual Life Outcomes. 2003;1(1):64.
28. American College of Rheumatology. Western Ontario and McMaster Universities Osteoarthritis Index
(WOMAC). http://www.rheumatology.org/practice/clinical/clinicianresearchers/outcomes-instrumentation/
WOMAC.asp. Accessed March 27, 2019.
29. Xie F, Pullenayegum EM, Li SC, Hopkins R, Thumboo J, Lo NN. Use of a disease-specific instrument in
economic evaluations: mapping WOMAC onto the EQ-5D utility index. Value Health. 2010;13(8):873-878.
30. Higgins JPT, Altman DG, Gøtzsche PC, et al; Cochrane Bias Methods Group; Cochrane Statistical Methods
Group. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928.
31. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M. Reliability of the PEDro scale for rating quality of
randomized controlled trials. Phys Ther. 2003;83(8):713-721.
32. Schünemann H, Brożek J, Guyatt G, Oxman A, eds. Handbook for Grading the Quality of Evidence and the
Strength of Recommendations Using the GRADE Approach. https://gdt.gradepro.org/app/handbook/handbook.html.
Updated October 2013. Accessed March 27, 2019.
33. Meader N, King K, Llewellyn A, et al. A checklist designed to aid consistency and reproducibility of GRADE
assessments: development and pilot validation. Syst Rev. 2014;3:82.
34. GRADEpro GDT [software]. Evidence Prime, Inc. https://gradepro.org/. Accessed March 27, 2019.
35. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177-188.
36. RevMan [computer program]. Version 5.3. Copenhagen, Denmark: Nordic Cochrane Centre, Cochrane
Collaboration; 2014.
37. Kramer JF, Speechley M, Bourne R, Rorabeck C, Vaz M. Comparison of clinic- and home-based rehabilitation
programs after total knee arthroplasty. Clin Orthop Relat Res. 2003;(410):225-234.
38. Madsen M, Larsen K, Madsen IK, Søe H, Hansen TB. Late group-based rehabilitation has no advantages
compared with supervised home-exercises after total knee arthroplasty. Dan Med J. 2013;60(4):A4607.
39. Mockford BJ, Thompson NW, Humphreys P, Beverland DE. Does a standard outpatient physiotherapy regime
improve the range of knee motion after primary total knee arthroplasty? J Arthroplasty. 2008;23(8):1110-1114.
40. Rajan RA, Pack Y, Jackson H, Gillies C, Asirvatham R. No need for outpatient physiotherapy following total
knee arthroplasty: a randomized trial of 120 patients. Acta Orthop Scand. 2004;75(1):71-73.
41. Naylor JM, Mills K, Buhagiar M, Fortunato R, Wright R. Minimal important improvement thresholds for the
six-minute walk test in a knee arthroplasty cohort: triangulation of anchor- and distribution-based methods. BMC
Musculoskelet Disord. 2016;17(1):390.
42. Rasekaba T, Lee AL, Naughton MT, Williams TJ, Holland AE. The six-minute walk test: a useful metric for the
cardiopulmonary patient. Intern Med J. 2009;39(8):495-501.
43. 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.

44. Mahomed NN, Davis AM, Hawker G, et al. Inpatient compared with home-based rehabilitation following
primary unilateral total hip or knee replacement: a randomized controlled trial. J Bone Joint Surg Am. 2008;90(8):
1673-1680.
45. ClinicalTrials.gov. Unsupervised home exercise with and without a web-based recovery platform as compared
to traditional outpatient physiotherapy after total knee arthroplasty: a prospective, randomized controlled trial.
NCT02911389. https://clinicaltrials.gov/ct2/show/NCT02911389. Accessed February 27, 2019.
46. Wylde V, Artz N, Marques E, et al. Effectiveness and cost-effectiveness of outpatient physiotherapy after knee
replacement for osteoarthritis: study protocol for a randomised controlled trial. Trials. 2016;17(1):289.
47. Artz N, Dixon S, Wylde V, et al. Comparison of group-based outpatient physiotherapy with usual care after
total knee replacement: a feasibility study for a randomized controlled trial. Clin Rehabil. 2017;31(4):487-499.
48. Hensman-Crook A. The effectiveness of physiotherapy intervention with home exercise programme versus
patient directed home exercise programme following total knee replacement. Intern Med J. 2011;41(suppl 1):39.
SUPPLEMENT.
eMethods 1. Search Strings for Systematic Review and Meta-analysis
eMethods 2. Studies Excluded After Full-Text Review
eTable 1. PEDro Scores of Included Studies
eTable 2. Meta-analysis Data
eFigure 1. Mobility
eFigure 2. Patient-Reported Pain and Function
eFigure 3. Patient-Reported Quality of Life
eFigure 4. Active Range of Motion
eFigure 5. Passive Range of Motion Change From Baseline

Buhagiar 2019 Oi 190124

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Buhagiar 2019 Oi 190124

Uploaded by

Copyright:

Available Formats

Original Investigation | Physical Medicine and Rehabilitation

Assessment of Outcomes of Inpatient or Clinic-Based

Abstract Key Points

Downloaded From: https://jamanetwork.com/ on 11/23/2022

CONCLUSIONS AND RELEVANCE Based on low- to moderate-quality evidence, no superiority of

JAMA Network Open. 2019;2(4):e192810. doi:10.1001/jamanetworkopen.2019.2810

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Primary and Secondary Outcomes

Identification and Selection of Studies

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Assessment of Study Quality, Risk of Bias, and GRADE Assessment

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Risk of Bias in Included Studies

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Intervention Outcomes and Comparisons

Table 1. Summary of Included Studies

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Patient-Reported Pain and Function

Figure 1. Flow of Studies Through the Review

2286 Records identified through database search and other sources

404 Duplicates removed

1882 Abstracts screened

1867 Abstracts excluded

15 Full-text articles assessed for eligibility

9 Full-text articles excludeda

Figure 2. Cochrane Risk of Bias Table

Buhagiar et al, 201716

Random Allocation Blinding of Blinding of Incomplete Selective Other Bias

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Inpatient Rehabilitation vs Home-Based Rehabilitation

No. of Certainty of the Anticipated Absolute Effectsb

Downloaded From: https://jamanetwork.com/ on 11/23/2022

No. of Certainty of the Anticipated Absolute Effectsb

Downloaded From: https://jamanetwork.com/ on 11/23/2022

No studies included in this review considered whether outcomes of post-TKA rehabilitation

Comparison With Other Reviews

Quality of the Evidence

Strengths and Limitations

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Downloaded From: https://jamanetwork.com/ on 11/23/2022

Downloaded From: https://jamanetwork.com/ on 11/23/2022

You might also like