Abstract
Background: The relief of pain and the restoration of functional
activities are the main outcomes of primary total knee arthroplasty for the
treatment of osteoarthritis. This paper examines the preoperative predictors
of pain and functional outcome at one and two years following total knee
arthroplasty.
Methods: Patients were recruited for a prospective observational
study of primary total knee arthroplasty for the treatment of osteoarthritis
from centers in the United States, the United Kingdom, and Australia. Research
assistants recruited the patients and collected the clinical history and
physical examination data preoperatively and at three, twelve, and twenty-four
months postoperatively. The Western Ontario and McMaster University
Osteoarthritis Index (WOMAC), Short Form-36 (SF-36), and demographic data were
obtained by self-administered patient questionnaires.
Results: We recruited 860 patients and obtained one-year WOMAC data
on 759 patients (88%) and two-year data on 701 (82%). The mean age was seventy
years, and 59% of the patients were female. Using hierarchical regression
models, we found that the most significant preoperative predictors of worse
scores on the pain and function domains of the WOMAC scale and on the physical
functioning domain of the SF-36 at one and two years postoperatively were low
preoperative scores, a higher number of comorbid conditions, and a low SF-36
mental health score. After adjusting for these predictors, we found that the
functional status of the patients from the United Kingdom was significantly
worse than that of the patients from the other countries and the difference
was clinically important at both the one-year and two-year follow-up
examination (p < 0.0005). The mean WOMAC pain scores for the three
countries were not significantly different at one year, and, although they
were significantly different at two years (p = 0.025), the difference was not
clinically important.
Conclusions: Patients who have marked functional limitation, severe
pain, low mental health score, and other comorbid conditions before total knee
arthroplasty are more likely to have a worse outcome at one year and two years
postoperatively. After adjusting for these predictors, it was found that
patients from the United Kingdom had significantly worse functional outcomes
but similar pain relief compared with those from the United States and
Australia.
Level of Evidence: Prognostic study, Level I-1
(prospective study). See Instructions to Authors for a complete description of
levels of evidence.
Knee osteoarthritis is a common cause of severe pain and functional
limitation, affecting approximately 6% of the adult population; this
percentage increases in people who are more than fifty-five years
old1,2.
Total knee arthroplasty is now the accepted optimal management for patients
with severe arthritis of the
knee3-6.
Previous research has shown that there may be gender and racial inequality in
the provision of total knee
arthroplasty7.
Hawker et al.8 found
that underuse of joint arthroplasty was more than three times greater in women
than in men. Katz et
al.9 demonstrated
that women had consistently worse preoperative self-reported pain and function
but had comparable or greater functional improvement following elective hip
and knee arthroplasty.
Differences in health-care systems and the type of access that patients
have to total knee replacement may contribute to the variation in the status
of patients at the time of surgery. In the United Kingdom, patients wait up to
twelve months for surgery (generally according to the chronological order in
which they enter the waiting list) once the decision to operate has been made.
In a study comparing patients from a center in Montreal, Quebec, Canada, and a
center in Boston, Massachusetts, important differences between the two centers
were identified with respect to the preoperative status of the
patients10.
Additionally, preoperative status was the strongest determinant of functional
outcomes at six months and two years following hip and knee
surgery10,11.
The ability to identify patients who have worse pain and functional status
could act as an adjunct to clinical and radiographic data for surgeons and
health-care providers who refer patients for surgery. In addition, it could
help surgeons to counsel patients with regard to what they may expect in terms
of pain and function following surgery, and it could help the patient to set
realistic expectations. These preoperative assessment tools, in conjunction
with clinical and radiographic data, may assist in prioritizing patients with
more severe symptoms for surgery in centers that have limited resources and a
long waiting list.
To date, no prospective study has compared preoperative status and outcomes
of total knee arthroplasty across multiple centers in different countries.
This study compares the preoperative pain and functional status of patients
and their corresponding outcome at one year and two years following primary
total knee replacement. Patients were recruited from twelve centers globally:
six centers in the United Kingdom, four centers in the United States, and two
centers in Australia.
The initial aims of this study were to determine (1) whether patients in
different countries varied in terms of preoperative pain and functional
status, and (2) whether it was possible to identify consistent characteristics
across countries and centers that correlated with worse preoperative pain and
functional status. We hypothesized that (1) there would be a significant
difference in pain and functional status at the time of the preoperative
assessment among the patients from the United Kingdom, the United States, and
Australia after adjusting for independent covariates, and (2) women would have
significantly worse preoperative pain and functional status than would men
after adjusting for independent covariates.
The secondary aim of this study was to confirm, in a multicenter,
multinational study, the influence of preoperative pain and functional status
on outcomes at one year and two years postoperatively. We hypothesized that
preoperative pain and functional status would be the strongest predictors of
pain and functional outcomes at one year and two years after surgery.
Design
Data for these analyses were obtained as part of the Kinemax Outcomes
Study, which was a prospective observational study of primary total knee
arthroplasty for the treatment of osteoarthritis in twelve centers: four in
the United States, six in the United Kingdom, and two in Australia. The
appropriate institutional review board or ethical committee approved the study
at each of the participating centers. Independent research assistants at the
participating sites recruited patients from September 1997 to December 1998 in
the United Kingdom, the United States, and Australia.
Patients
All patients undergoing primary total knee arthroplasty with use of the
Kinemax prosthesis (Stryker Howmedica, Mahwah, New Jersey) for the treatment
of osteoarthritis were included. Patients were excluded if they had a history
of knee joint infection or prior implant surgery to the index knee or were
unable to complete the questionnaires because of cognitive or language
difficulties. Patients who had a bilateral total knee arthroplasty within one
year were excluded from these analyses to ensure that the one-year results
reflected the outcome of the index operation and not a subsequent
operation.
Data Collection Procedures
Independent research assistants working at the various sites recruited
eligible patients and obtained informed consent. Research assistants collected
data using a standardized protocol. They carried out a clinical history and
physical examination of the patients and administered the questionnaires.
Preoperative data were collected within six weeks prior to the total knee
arthroplasty, and follow-up data were collected at three, twelve, and
twenty-four months following surgery. The lead author (E.A.L.) trained all
research assistants to standardize data collection, and the data were entered
into a single database at the coordinating center.
Data Elements
The preoperative questionnaire included demographic details, socioeconomic
data (education, income, working status, and living arrangements), height,
weight, and history of comorbid conditions. The self-administered comorbidity
questionnaire has been validated with a medical record-based comorbidity
instrument as well as with subsequent health status and
utilization12. At
each evaluation, two health-status scales were administered: the Western
Ontario and McMaster University Osteoarthritis Index (WOMAC), a
disease-specific measure of pain, stiffness, and
function13,14,
and the Short-Form-36 (SF-36), a generic health-status
measure15-17.
We transformed WOMAC scores to a 0 to 100-point scale for each domain
(with 100 being the best score). The standardized method of calculating the
SF-36 domains was used so that each of the eight subscales had a score of 0 to
100 (with 100 being the best
score)18.
Data collected by the research assistants for the clinical history included
previous orthopaedic surgery on the lower limb and the current level of pain
and functional ability (walking distance, stair-climbing ability, and use of a
walking aid). The physical examination included an assessment of the range of
motion, stability, alignment, and muscle power of the knee.
Statistical Methods
Statistical analyses were performed with use of SAS statistical software
(version 8.0; SAS Institute, Cary, North Carolina). Comparisons of the
patients from the United Kingdom, United States, and Australia with regard to
preoperative demographic, socioeconomic, and clinical characteristic data were
made with use of analysis of variance and chi-square analyses where
appropriate.
The independent variables were age, age-squared, gender, income (coded as 0
= lowest income category or 1 = above lowest income category; income
categories were country-specific and based on median household income for each
country), educational status (coded as 0 = less than a high school education
or 1 = a high school education or greater), working status (coded as 0 = not
working or 1 = working part-time or full-time), body mass index (calculated as
weight in kilograms divided by height in meters squared), preoperative SF-36
mental health score, number of comorbid conditions, preoperative knee flexion,
country, and center within country. We used both age and age-squared as
independent variables to analyze the linear and quadratic effects of age. When
data were missing, we substituted the mean value for each country.
We used hierarchical models to determine the significant correlates (p <
0.05) of pain and functional status at the preoperative assessment. The first
set of factors to be entered were age and gender, which were retained in
future steps regardless of the level of significance. The second set of
variables to be entered were the socioeconomic variables
(Table I). The third set of
variables to be entered were age, gender, and the significant variables from
the second set and the preoperative clinical variables
(Table II). The fourth set of
variables were age, gender, the significant variables from the previous step,
and the center. The final set of variables included age, gender, the
accumulated significant variables from the previous steps, and country. With
use of this method, the final models were defined for the three dependent
variables, the pain and function scores on the WOMAC scale and the physical
functioning score on the SF-36.
We used a similar hierarchical model to determine the significant
preoperative predictors of pain and functional status at one year and two
years following surgery, with the exception that the preoperative scores for
each of the outcome measures were added to the third set of variables. All
final models were tested for robustness by running the analysis without the
imputed values to test whether the same variables remained significant.
The final models, which adjusted for age, gender, country, and center
within country, in all instances were reported with the respective F value
(equal to t2) and the parameter estimate with 95% confidence
intervals for all of the variables. The F value is a measure of the strength
of the effect, with higher scores indicating a stronger effect. Parameter
estimates for continuous variables can be interpreted as that amount of change
per unit increase, and parameter estimates for dichotomous variables can be
interpreted as that amount of difference between the two groups. As country is
a categorical variable with two degrees of freedom, the interpretation of the
parameter estimates was complex and we therefore reported the F value and the
difference between adjusted means, which gave a clear indication of the
absolute difference between countries in mean scores, adjusting for
significant covariates. Australia was used as the reference country for
comparisons between countries. Differences in the WOMAC and SF-36 scores of 9
to 12 points on a 100-point scale have been shown to be perceptible to
patients and are clinically
meaningful19,20.
Recruitment
Between September 1997 and December 1998, 1100 (76.6%) of all 1435 eligible
patients were recruited. One hundred and twenty-four eligible patients (8.6%)
were not recruited because they refused consent; 197 patients (13.7%) were
missed preoperatively because the research assistant was absent as a result of
sickness or vacation or because the patients were brought in as last-minute
admissions on the day of surgery, which did not allow sufficient time to
inform, obtain consent, and evaluate them prior to surgery; and fourteen
patients (1.0%) were already enrolled in another study, and the institutional
review board protocol at this site did not allow recruitment of patients into
more than one study. During the recruitment period, only 6% of all patients
who had a primary total knee arthroplasty performed by the participating
surgeons were not managed with a Kinemax prosthesis, frequently because a more
constrained prosthesis was required.
After the exclusion of 240 patients who had a bilateral total knee
arthroplasty within twelve months, the final sample for this analysis was 860
patients. We had complete data on 741 patients (86.2%) at one year and on 678
patients (78.8%) at two years. Data on the physical examination only were
available for an additional eighteen patients (2.1%) at one year and for
twenty-three patients (2.7%) at two years. The reasons for the missing data
are listed in the Appendix.
Patient Features and Outcomes
Table I shows the
demographic and socioeconomic characteristics of the patients. The mean age of
the 860 patients was 69.9 years (range, thirty-eight to eighty-nine years),
and the majority of the patients (509; 59.2%) were female. A total of 430
patients (50.0%) were recruited by the centers in the United Kingdom; 260
patients (30.2%), by the centers in the United States; and 170 patients
(19.8%), by the centers in Australia. The patients in the United Kingdom were
significantly older than the patients in the United States (F = 4.31, p =
0.014), but no significant difference was detected among the three countries
with respect to the proportion of patients who were female (chi square = 2.60,
p = 0.27). The patients in the United States had a dramatically higher
educational level (chi square = 225.1, p < 0.0001) and reported higher
incomes (chi square = 10.78, p = 0.0046); in addition, a significantly higher
proportion of them were still working (chi square = 43.2, p < 0.0001)
compared with the patients in Australia and the United Kingdom
(Table I).
Preoperative Clinical Characteristics
Preoperative clinical characteristics are shown in
Table II. The patients in the
United States had a significantly higher body mass index (F = 23.79, p <
0.0001) and a larger proportion (39.6%) were classified as obese (that is, a
body mass index of
>30)21 compared
with the patients in Australia (18.5%) and the United Kingdom (19.1%). The
patients in the United Kingdom had significantly fewer comorbid conditions (F
= 9.03, p < 0.0001) than the patients in Australia and United States. The
mean SF-36 mental health scores were similar to the population-based norms as
reported by the Medical Outcomes
Trust18, and
Australian patients had significantly better mental health scores than did the
other patients (p = 0.05).
The proportion of patients who reported having had knee pain for greater
than five years was significantly higher for the United Kingdom than for the
United States or Australia (chi square = 18.1, p < 0.0001,
Table II). Patients in
Australia reported significantly higher unadjusted WOMAC pain scores than did
those in the United States and United Kingdom (F = 11.83, p < 0.0001). In
addition, patients in the United Kingdom had significantly less preoperative
flexion (F = 38.11, p < 0.0001) than did patients in the United States and
Australia. At the preoperative assessment, the mean unadjusted function scores
were significantly different for both WOMAC (F = 17.81, p < 0.0001) and
SF-36 (F = 9.83, p < 0.0001) instruments among all three countries, with
the worst scores for the United Kingdom and the best scores for Australia
(Table II).
Correlates of Preoperative Pain and Functional Status
A hierarchical regression model was used to determine the significant
correlates of preoperative pain and function. The final models, which
controlled for age and gender in all instances, are presented in the Appendix.
Age is significant in the models where the parameter estimates are recorded as
having a p value of <0.05. When the parameter estimate is positive (for
example, preoperative WOMAC pain = 0.36 [p < 0.0005]), it means that, for
each year of age, the WOMAC pain score increases by 0.36, indicating that
younger patients report significantly more pain (i.e., they have lower WOMAC
pain scores than do older patients) preoperatively. Conversely, when the
parameter estimates are negative and are recorded as being significant (e.g.,
WOMAC function at one year and SF-36 physical functioning at one year and two
years), increasing age would be associated with a significantly lower score,
indicating more pain. For SF-36 physical functioning, we found that age did
not have a linear relationship preoperatively and that both very old patients
and younger patients had worse scores (see Appendix).
After adjusting for significant covariates, we found that the patients in
the United Kingdom had significantly worse WOMAC pain and function scores than
did the patients in the United States and that female patients were
significantly worse overall as well as within each country and each center
within countries. Knee flexion and mental health status were two of the
strongest determinants. The patients in the United Kingdom had significantly
lower SF-36 physical functioning scores after adjusting for these significant
covariates, and there was a significant difference between the centers within
each country. Female patients were not significantly worse than male patients
with respect to SF-36 physical functioning (p = 0.17).
Preoperative Predictors of Postoperative WOMAC Pain and Functional
Status
The final models for the one and two-year outcomes that control for age and
gender in all instances are presented in the Appendix. The preoperative WOMAC
pain and SF-36 mental health scores were the strongest determinants of the
postoperative WOMAC pain score at one year and two years. Although no
significant difference among the countries was found with respect to pain at
one year, the patients in the United Kingdom were significantly worse than the
other patients at two years (Fig.
1). With a large cohort of patients, we can demonstrate a
significant difference, but it is important to note when the differences in
the WOMAC scores are clinically important. Differences in the WOMAC scores of
9 to 12 points on a 100-point scale have been shown to be perceptible to
patients and are clinically
meaningful19,20.
The preoperative WOMAC function score was noticeably the strongest
determinant of the WOMAC function score at one year and two years after the
operation. Despite the marked differences between male and female patients at
the preoperative assessment, no significant difference in the WOMAC function
scores was found between men and women at one year and two years. After
adjusting for the other significant predictors, we found that the patients in
the United Kingdom were significantly worse than those in the United States
and Australia in terms of the WOMAC function score at one year; this
difference was also evident at two years
(Fig. 1).
To demonstrate the strong influence of the preoperative WOMAC function
score on postoperative outcomes at one year and two years, we analyzed the
total patient group according to the preoperative WOMAC function score
quartiles (<34, 34 to 45, 46 to 57, and >57). The mean WOMAC function
scores for each group at the preoperative, one-year, and two-year assessments
are shown in Figure 2. At one
year and two years, the patients with a preoperative WOMAC function score in
the lowest quartile (<34) had mean scores of 62.1 and 59.8, respectively,
which indicates that these patients had, on the average, considerable
functional disability after total knee arthroplasty. The patients with scores
in the lowest quartile group had the greatest improvement in their scores
compared with the other groups. We found that patients with a preoperative
WOMAC function score in the lowest quartile were over four times (odds ratio,
4.12; 95% confidence interval, 2.86, 6.25) more likely to have a score of
=60 at two years following surgery than were patients who had a
preoperative WOMAC function score of >35.
Preoperative Predictors of Postoperative SF-36 Physical
Functioning
The final models for the one and two-year outcomes that control for age and
gender in all instances are presented in the Appendix. The preoperative SF-36
physical functioning score was the strongest predictor of the SF-36 physical
functioning score at one year. At two years, the preoperative SF-36 physical
functioning score was still a strong predictor of outcome, but a greater
number of comorbid conditions and older age were equally powerful predictors
in this model. After adjusting for other significant predictors, we found that
patients in the United Kingdom remained significantly worse in terms of this
score at the postoperative assessment at both one year and two years (p <
0.0005).
Validation of Final Models
All final models were rerun without the imputed values for missing data to
check for the robustness of our results. The final models are identical to the
ones reported in the Appendix with the following exceptions: (1) the
preoperative WOMAC pain model did not include income but did include work
status (p < 0.05), (2) the preoperative WOMAC function model did not
include income, and (3) the one-year WOMAC function model did not include body
mass index.
Our results demonstrate that the preoperative pain and functional status of
patients varies among countries and that the patients in the United Kingdom
had consistently worse status in all measures of pain and function, even after
adjusting for independent covariates. In each country, female patients were
significantly worse than their male counterparts preoperatively. At the time
of follow-up, female patients were not significantly different from the male
patients with respect to the WOMAC pain and function scores. Our results also
confirmed that preoperative status was the strongest determinant of outcome at
both one year and two years.
The finding that the patients in the United Kingdom were significantly
worse preoperatively in terms of pain and functional status was not surprising
as, at the time of recruitment, we found that patients in the United Kingdom
were waiting up to twelve months for an orthopaedic consultation and a further
twelve to eighteen months for surgery after they had been listed for surgery.
These waiting times were considerably longer than those in the United States
and Australia22.
Differences in the preoperative status may also arise from differences among
the countries with respect to the patient's willingness to seek treatment and
to consider surgical options. Cultural differences in terms of lifestyles and
the criteria by which patients determine their level of pain and functional
status may also play a role in patient-reported health status. Further
cross-cultural population-based studies should examine these factors.
Our results demonstrate that female patients were significantly worse than
the male patients at the preoperative assessment in terms of pain and
functional status, which substantiates the findings in prior
work8,9,23.
It has been shown that men and women differ in their preferences for surgery
and that women are more likely than men are to wait until their function has
deteriorated before accepting the offer of
surgery24.
Male and female patients were not significantly different at the one-year
and two-year follow-up evaluation in terms of WOMAC function score despite the
marked differences at the preoperative assessment. This observation is
consistent with the findings of Fortin et
al.10, who studied
outcomes at six months following primary total knee arthroplasty. Men had
significantly lower WOMAC pain scores at one year and two years, but this
difference would not be perceptible to patients and is not considered a
clinically meaningful
difference19,20.
Socioeconomic variables were significant correlates of preoperative pain
and functional status as has been reported
previously25,26.
The presence of comorbid medical conditions in our patient cohort did not have
a significant impact on the preoperative WOMAC scores, but it did have a
significant influence on patient outcomes, especially functional outcomes, at
one year and two years. Patients with other medical problems may be unable to
participate fully in the demanding postoperative rehabilitation, and surgeons
may need to counsel patients that their outcome may be compromised by the
presence of other medical problems.
Preoperative pain and functional status were overwhelmingly the strongest
determinants of outcomes at one year, and this remained true at two years.
This observation adds to the findings of Fortin et
al.10,11
and emphasizes the important influence of preoperative functional status on
outcome at one year and two years. In our analysis, patients with a
preoperative WOMAC function score in the lowest quartile made the greatest
improvements at one year, with only a slight decline at two years
(Fig. 2). However, they were
more than four times more likely to have a score of =60 points two years
after surgery than were patients in the other groups, indicating moderate
functional limitation. Further research needs to be undertaken to ascertain
whether poor access to total knee arthroplasty can cause functional status to
deteriorate to a level that may compromise the functional results. In the
determination of the optimal time for surgery, these findings need to be
carefully balanced with the possibility that earlier surgery for younger, more
active patients may potentially lead to higher rates of revision.
At one year and two years postoperatively, the patients in the United
Kingdom had significantly worse functional status than did those in the United
States and Australia, even after adjusting for the significant independent
covariates including preoperative status. This difference was not only
significant but also clinically important as the adjusted mean scores for the
patients in the United Kingdom were 8 to 11 points less than those for the
patients in the United States at one year and two
years19,20.
The reasons that the patients in the United Kingdom reported more functional
limitation require further investigation. We know from our survey of care
pathways at all of the different centers that the patients in the United
Kingdom were not referred to inpatient rehabilitation facilities and that
relatively few, compared with the patients in the United States, had either
home or outpatient rehabilitation once they were discharged from the
hospital22. No
difference between the countries was found in terms of WOMAC pain scores at
one year, and the difference at two years was significant but not clinically
important19,20.
A limitation of our analysis of mental health status is that we were unable
to prove cause and effect with this data, and we were unable to conclude
whether poor mental health causes poor function or whether poor functional
status leads to poor mental health. However, we were able to show that, even
after we adjusted for preoperative pain and functional status, a low mental
health score at the preoperative assessment was significantly associated with
a lower outcome score at one year and two years.
The relationship between the pain and disability caused by osteoarthritis
and mental health status is complex, and we do not in any way imply that
surgeons should focus on a patient's mental health status or indeed limit
access to total knee arthroplasty for patients with poor mental health. It is
important to be aware that a low score for mental health does not conclusively
indicate clinical depression but is a general measure of a patient's mental
health status and may be affected by an acute event or a chronic
condition.
All clinicians involved with the care of patients managed with a total knee
arthroplasty should be aware that mental health status can have an impact on
the pain and functional status of the patient, and the assessment of mental
health status may in some instances enable us to modify the way in which we
manage these patients to optimize the outcome following total knee
arthroplasty. At a minimum, it will allow clinicians to counsel patients that
their outcome may be compromised. We hope that these results will stimulate
other researchers to explore these relationships in more detail.
The present study identifies a set of factors that portend a worse outcome
following primary total knee arthroplasty for the treatment of osteoarthritis.
Surgeons should discuss these factors with patients and indicate frankly to
patients who have these risk factors that they are at risk of a poorer
outcome. This list is by no means exhaustive, and further research needs to
build on these findings to identify other factors that may influence the
outcome. The WOMAC questionnaire asks patients to rate their functional
limitation or level of pain due to the "knee that was replaced,"
and differences in these scores are largely due to a change in knee status.
However, it must be noted that, although the WOMAC has been shown to be
responsive and sensitive to detecting changes in pain and functional status in
the study knee following total knee arthroplasty, medical and musculoskeletal
comorbid conditions may also influence these scores.
A limitation of the present study is that we recruited patients from select
surgeons and centers rather than using a population-based cohort. The study
strengths included the use of trained independent research assistants to
recruit patients and follow them at each assessment and the fact that we
obtained detailed preoperative data and high rates of follow-up at one year
and two years for a large cohort of patients.
Tables showing the reasons for missing data and the correlates of worse
preoperative status and postoperative predictors of worse outcome are
available with the electronic versions of this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM).
DieppeP,
Basler HD, Chard J, Croft P, Dixon J, Hurley M, Lohmander S, Raspe H. Knee
replacement surgery for osteoarthritis: effectiveness, practice variations,
indications and possible determinants of utilization. Rheumatology
(Oxford).1999;38:
73-83.3873
1999
[CrossRef]
FelsonDT,
Naimark A, Anderson J, Kazis L, Castelli W, Meenan RF. The prevalence of
knee osteoarthritis in the elderly. The Framingham Osteoarthritis Study.
Arthritis Rheum.1987;30:
914-8.30914
1987
[PubMed][CrossRef]
CallahanCM,
Drake BG, Heck DA, Dittus RS. Patient outcomes following tricompartmental
total knee replacement. A meta-analysis. JAMA.1994; 271:
1349-57.2711349
1994
[PubMed][CrossRef]
CallahanCM,
Drake BG, Heck DA, Dittus RS. Patient outcomes following unicompartmental
or bicompartmental knee arthroplasty. A meta-analysis.
JArthroplasty.1995;10:
141-50.10141
1995
[CrossRef]
HawkerG,
Wright J, Coyte P, Paul J, Dittus R, Croxford R, Katz B, Bombardier C, Heck D,
Freund D. Health-related quality of life after knee replacement.
J Bone Joint Surg Am.1998;80:
163-73.80163
1998
[PubMed]
HeckDA,
Robinson RL, Partridge CM, Lubitz RM, Freund DA. Patient outcomes after
knee replacement. Clin Orthop.1998;356:
93-110.35693
1998
[PubMed][CrossRef]
KatzBP, Freund
DA, Heck DA, Dittus RS, Paul JE, Wright J, Coyte P, Holleman E, Hawker G.
Demographic variation in the rate of knee replacement: a multi-year analysis.
Health Serv Res.1996;31:
125-40.31125
1996
[PubMed]
HawkerGA,
Wright JG, Coyte PC, Williams JI, Harvey B, Glazier R, Badley EM.
Differences between men and women in the rate of use of hip and knee
arthroplasty. N Engl J Med.2000;342:
1016-22.3421016
2000
[PubMed][CrossRef]
KatzJN, Wright
EA, Guadagnoli E, Liang MH, Karlson EW, Cleary PD. Differences between men
and women undergoing major orthopedic surgery for degenerative arthritis.
Arthritis Rheum.1994;37:
687-94.37687
1994
[PubMed][CrossRef]
FortinPR,
Clarke AE, Joseph L, Liang MH, Tanzer M, Ferland D, Phillips C, Partridge AJ,
Belisle P, Fossel AH, Mahomed N, Sledge CB, Katz JN. Outcomes of total hip
and knee replacement: preoperative functional status predicts outcomes at six
months after surgery. Arthritis Rheum.1999;42:
1722-8.421722
1999
[PubMed][CrossRef]
FortinPR,
Penrod JR, Clarke AE, St-Pierre Y, Joseph L, Belisle P, Liang MH, Ferland D,
Phillips CB, Mahomed N, Tanzer M, Sledge C, Fossel AH, Katz JN. Timing of
total joint replacement affects clinical outcomes among patients with
osteoarthritis of the hip or knee. Arthritis Rheum.2002;46:
3327-30.463327
2002
[PubMed][CrossRef]
SanghaO,
Stucki G, Liang MH, Fossel AH, Katz JN. The Self-Administered Comorbidity
Questionnaire: a new method to assess comorbidity for clinical and health
services research. Arthritis Rheum.2003;49:
156-63.49156
2003
[PubMed][CrossRef]
BellamyN,
Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of
WOMAC: a health status instrument for measuring clinically important patient
relevant outcomes to antirheumatic drug therapy in patients with
osteoarthritis of the hip or knee. J Rheumatol.1988;15:
1833-40.151833
1988
[PubMed]
BellamyN.WOMAC osteoarthritis index: a user's guide. London,
ON; 1995.
1995
WareJE Jr,
Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I.
Conceptual framework and item selection. Med Care.1992;30:
473-83.30473
1992
[PubMed][CrossRef]
McHorneyCA,
Ware JE Jr, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36):
II. Psychometric and clinical tests of validity in measuring physical and
mental health constructs. Med Care.1993;
31: 247-63.31247
1993
[PubMed][CrossRef]
McHorneyCA,
Ware JE Jr, Lu JF, Sherbourne CD. The MOS 36-item Short-Form Health Survey
(SF-36): III. Tests of data quality, scaling assumptions, and reliability
across diverse patient groups. Med Care.1994;32:
40-66.3240
1994
[PubMed][CrossRef]
WareJE Jr,
Snow KK, Kosinski M, Gandek B.SF-36 health survey: manual and
interpretation guide. 2nd ed. Boston: The Health Institute, New
England Medical Center; 1997. p
10-26.10
1997
EhrichEW,
Davies GM, Watson DJ, Bolognese JA, Seidenberg BC, Bellamy N. Minimal
perceptible clinical improvement with the Western Ontario and McMaster
Universities osteoarthritis index questionnaire and global assessments in
patients with osteoarthritis. J Rheumatol.2000;27:
2635-41.272635
2000
[PubMed]
AngstF,
Aeschlimann A, Stucki G. Smallest detectable and minimal clinically
important differences of rehabilitation intervention with their implications
for required sample sizes using WOMAC and SF-36 quality of life measurement
instruments in patients with osteoarthritis of the lower extremities.
Arthritis Rheum.2001;45:
384-91.45384
2001
[PubMed][CrossRef]
National Heart, Lung, and Blood Institute, National
Institute of Diabetes and Digestive and Kidney Diseases.Clinical guidelines on the identification, evaluation and treatment
of overweight and obesity in adults: the evidence report.
Bethesda, MD: National Heart, Lung, and Blood Institute, National Institute of
Diabetes and Digestive and Kidney Diseases; 1998. p
xvii.xvii
1998
LingardEA,
Berven S, Katz JN; Kinemax Outcomes Group. Management and care of patients
undergoing total knee arthroplasty: variations across different health care
settings. Arthritis Care Res.2000;13:
129-36.13129
2000
[PubMed][CrossRef]
KantzME,
Harris WJ, Levitsky K, Ware JE Jr, Davies AR. Methods for assessing
condition-specific and generic functional status outcomes after total knee
replacement. Med Care.1992;30(5 Suppl):
MS240-52.30MS240
1992
[PubMed][CrossRef]
KarlsonEW,
Daltroy LH, Liang MH, Eaton HE, Katz JN. Gender differences in patient
preferences may underlie differential utilization of elective surgery.
Am J Med.1997;102:
524-30.102524
1997
[PubMed][CrossRef]
WasielewskiRC,
Weed H, Prezioso C, Nicholson C, Puri RD. Patient comorbidity:
relationship to outcomes of total knee arthroplasty. Clin
Orthop.1998;356:
85-92.35685
1998
[CrossRef]
BrinkerMR,
Lund PJ, Barrack RL. Demographic biases of scoring instruments for the
results of total knee arthroplasty. J Bone Joint Surg
Am.1997;79:
858-65.79858
1997