Full text of DOI:10.1093/gerona/56.3.M167

Journal of Gerontology: MEDICAL SCIENCES
Copyright 2001 by The Gerontological Society of America
2001, Vol. 56A, No. 3, M167–M175
Functional Status and Health-Related Quality of Life
of Elderly Osteoarthritic Patients Treated
With Celecoxib
Jeffrey Lisse,1 Luis Espinoza,2 Sean Z. Zhao,3 Seema D. Dedhiya,4 and Jane T. Osterhaus4
1University of Texas, Galveston.
2Louisiana State University, New Orleans.
3Pharmacia Corporation, Peapack, New Jersey.
4Pharmacia Corporation, Skokie, Illinois.
Background. This study evaluates the impact of celecoxib on functional status, health-related quality of life
(HRQOL), and safety of elderly patients (Ն70 years) with osteoarthritis (OA) of the knee and/or hip.
Methods. Data were pooled from three prospective, randomized, multicenter, double-blind, parallel group trials,
each having a 12-week treatment period. Multicenter studies were conducted in the United States and Canada. Data for
patients diagnosed with active OA of the knee and/or hip in a flare state who were 70 years of age and older were in-
cluded in the comparison of therapeutic doses of celecoxib or naproxen versus placebo (N ϭ 768). Elderly patients from
each of the three trials who were randomly assigned to groups treated with a placebo, 200 mg/day of celecoxib, 400 mg/
day of celecoxib, or 1000 mg/day of naproxen were included in this analysis. The Western Ontario and McMaster Uni-
versities Osteoarthritis Index was used to measure functional status. The Short Form-36 was used as a general measure
of HRQOL. Safety was assessed according to the incidence and type of adverse reactions as reported by the patients and
the rate of withdrawal due to adverse events.
Results. At the end of the treatment period, patients in the celecoxib groups had significant improvement in both
functional status and HRQOL in comparison with the placebo group. The effects of total daily doses of 200 mg of cele-
coxib, 400 mg of celecoxib, and 1000 mg of naproxen on functioning and HRQOL were not found to be significantly
different from each other. The incidence of serious adverse events and withdrawal from the studies due to adverse events
were similar in the celecoxib groups as they were in the placebo group. Overall, the naproxen group reported a signifi-
cantly higher incidence of gastrointestinal adverse events than did the placebo and the 200-mg-daily celecoxib groups.
Conclusions. This study showed that celecoxib and naproxen significantly improved functional status and HRQOL
in elderly patients compared with those treated with a placebo. Celecoxib-treated patients were also found to experience
safety and tolerability similar to that of the placebo-treated patients.
HE elderly population is of special interest because
pharmacological outcomes are affected by the combi-
paired functional status or inability to perform activities of
daily living leads to interference with social lives and per-
sonal relationships, reduces levels of independence, and in-
creases the need for assistance (6). Because treatments for
OA are largely aimed at the reduction of symptoms, treat-
ments that significantly reduce the signs and symptoms of
arthritis should also lead to an improvement in patients’
functional status as the severity of their symptoms decline.
The assessment of functional status in OA patients, there-
fore, is an important means of evaluating the benefits of
treatment.
Traditional treatment for OA, consisting mainly of the
use of nonsteroidal anti-inflammatory drugs (NSAIDs) and
analgesics, is often suboptimal, requiring a balance between
efficacy and tolerance of associated side effects (8). Al-
though efficacious at relieving pain, NSAIDs are frequently
associated with a number of side effects, some of which are
potentially life threatening. These include gastrointestinal
(GI) events (e.g., upset stomach, ulceration, and hemor-
rhagic ulceration), alteration of platelet function, impair-
ment of renal function, and interaction with other medica-
T
nation of multiple disease processes and the physiologic
effects of aging (1–3). Problems in this population can man-
ifest themselves as exaggerated or less-than-optimal re-
sponses to medication, symptoms of drug-induced illness,
lack of therapeutic response to medication, and adverse events
(4,5). It is therefore especially important to monitor the ef-
fects of treatment on safety, efficacy, and health-related
quality of life (HRQOL) in elderly persons.
One of the prevalent chronic conditions in this population
is osteoarthritis (OA), a degenerative joint disorder charac-
terized by pain, joint swelling, stiffness, and limitations in
physical functioning. About 80% of the population over the
age of 55 years are affected by OA (6). Impairments, as a
direct result of OA and due to the associated pain and im-
mobility, can impose substantial humanistic and economic
burdens on patients and health care providers. Patients’ pain,
joint swelling and stiffness, and forced immobility gradu-
ally lead to difficulty with physical functioning, such as
walking and ascending or descending stairs (7). Also, im-
M167

M168
LISSE ET AL.
tions commonly used in the elderly population (4,5,9–18).
Alternative treatments with long-standing analgesics such
as acetaminophen may not provide adequate pain relief and,
in high doses, can be associated with a host of undesirable
side effects (e.g., dizziness, nausea, abdominal pain, and
shortness of breath).
for periods no greater than 3 consecutive days) were al-
lowed as concomitant medications, except within 48 hours
prior to arthritis assessments during which no concomitant
analgesics were allowed. The use of other NSAIDs, oral or
injectable corticosteroids, or anticoagulants were prohib-
ited. Data were collected on patients’ demographic charac-
teristics, functional status, HRQOL, and safety.
The mechanism of action for currently available NSAIDs
is the prevention of the formation of prostaglandins by inhi-
bition of the enzyme cyclooxygenase (COX) (19). COX ex-
ists in at least two forms: COX-1 (constitutive form) and
COX-2 (inducible form). Preclinical data suggest that inhibi-
tion of COX-2 may be responsible for the anti-inflammatory
and analgesic effects of COX enzyme inhibition, whereas in-
hibition of COX-1 may be responsible for the adverse events
in the upper GI tract and platelets (20–23). Most NSAIDs
currently available are nonselective inhibitors of both COX-1
and COX-2. In contrast, celecoxib is a COX-2 specific inhib-
itor that maintains the beneficial effects of traditional
NSAIDs with a superior safety profile. Studies in OA pa-
tients have demonstrated that at full therapeutic doses, cele-
coxib has an equivalent efficacy to first-generation NSAIDs,
but with fewer adverse events (24), leading to the expect-
ancy that celecoxib will improve functional status and
HRQOL among elderly OA patients and cause fewer side ef-
fects. Although these studies included elderly patients, their
number was limited in each of these studies, restricting the
opportunity to assess treatment effects in this population.
This study, using pooled data from three OA studies,
evaluated the impact of celecoxib on OA in elderly patients
(Ն70 years). The study determined whether celecoxib im-
proved functional status and HRQOL and was safe and well
tolerated by elderly patients.
Description of the Western Ontario and McMaster
Universities Osteoarthritis Index.—The functional status
of OA patients was assessed using the Western Ontario and
McMaster Universities Osteoarthritis Index (WOMAC). The
WOMAC is a reliable, validated instrument that has been
used extensively to determine the functional status in pa-
tients with OA of the knee and hip (25–34). It has three sub-
scales with a total of 24 items (pain with 5 items, stiffness
with 2 items, and physical functioning with 17 items) and
can be self-administered in less than 5 minutes. The Likert
version of the WOMAC reports responses on a 5-point scale
where 0 represents “none,” 1 is “mild,” 2 is “moderate,” 3 is
“severe,” and 4 is “extreme.” Therefore, the 5-item pain sub-
scale score ranges from 0 to 20, the 2-item stiffness subscale
score ranges from 0 to 8, the 17-item physical functioning
subscale score ranges from 0 to 68, and the composite
WOMAC score has a range of 0 to 96. A lower score on any
subscale represents better functional status and, when change
in scores over time is calculated, a negative score represents
an improvement in functional status (35).
Description of the Short Form-36.—HRQOL was as-
sessed using the acute version of the Short Form-36 Health
Survey (SF-36). The SF-36 is a reliable, validated, generic
instrument that has been used extensively to measure
HRQOL in diverse patient groups, including arthritis pa-
tients (28,36–41). It measures HRQOL using eight separate
domains (physical functioning, role physical, bodily pain,
general health, vitality, social functioning, role emotional,
and mental health). Two summary scores, the physical com-
ponent score (PCS) and the mental component score (MCS),
can then be computed by weighting the corresponding do-
mains. A higher score on the eight domains and two sum-
mary scores represents a higher level of HRQOL (42).
Both the WOMAC and the SF-36 were administered to
patients at baseline (prior to receiving the study drug) and at
Weeks 2 and 12.
METHODS
Data Collection and Study Design
This study used data from three prospective, randomized,
double-blind, parallel group trials conducted at various sites
in the United States and Canada. The trials were conducted in
accordance with the principles of good clinical practice and
the Helsinki Declaration of 1975. The institutional review
board at each clinical site approved the protocol, and all pa-
tients were required to provide written informed consent.
The design of the three trials was similar with the excep-
tion that two studies assessed patients with OA of the knee
and/or hip and one study assessed patients with OA of the
hip only. Male and female outpatients, 18 years of age or
older, were eligible to participate in the trials if they met
predetermined criteria; however, only patients 70 years of
age or older were included in this analysis. Following a 2- to
7-day washout period of NSAIDs or any analgesic medica-
tion, symptomatic OA (“flare”) was confirmed at a baseline
visit according to predefined criteria for changes in the re-
sults of the arthritis assessments performed at screening.
Elderly patients who demonstrated an OA flare within 14
days of discontinuing NSAID or analgesic treatment for OA
and who were randomly assigned to placebo, 100-mg BID
(twice daily) celecoxib, 200-mg BID celecoxib, or 500-mg
BID naproxen groups were included in this analysis. Low-
dose aspirin (Յ325 mg/d) and acetaminophen (up to 2 g/d
Description of safety assessment.—Safety was assessed
according to the incidence and type of adverse reactions re-
ported by the patients and the rate of withdrawal due to ad-
verse events. Adverse events were recorded and coded ac-
cording to the World Health Organization Adverse Reaction
Terminology list. These events were graded by the investi-
gator as mild (causing no limitation of usual activities),
moderate (causing some limitation of usual activities), or
severe (causing inability to carry out usual activities). Safety
was assessed at baseline, Week 2, and Week 12.
Statistical Methods
All randomized elderly patients were included in the anal-
yses of pretreatment variables. All statistical analyses on

CELECOXIB IN OSTEOARTHRITIC PATIENTS
M169
treatment period variables were based on the intent-to-treat
population, defined as all patients who were randomized and
took at least one dose of study medication. The Last Obser-
vation Carried Forward approach was used for imputing
missing values (43). All statistical testing was two-sided at a
5% level of significance, and changes in WOMAC and SF-
36 scores from baseline are reported as least square means.
The statistical package SAS for Windows version 6.03 (SAS
Institute, Cary, NC) was used to conduct all analyses. For
statistical analyses with center as a factor, any center with no
patients in at least one of the five treatment groups was
pooled with another center geographically. Chi-square tests
were used to analyze homogeneity of treatment groups for
categorical variables. A two-way analysis of variance with
treatment and center included in the model was used to ex-
amine the homogeneity among treatment groups for continu-
ous variables and also to analyze baseline WOMAC and SF-
36 scores. Treatment comparisons were assessed using
change from baseline scores for each of the WOMAC sub-
scales and for the eight domains and two summary scores of
the SF-36. Statistical significance was tested using a two-
way analysis of covariance with treatment and center as fac-
tors and baseline scores included in the model as a covariate.
If the change in domain or composite score was significant,
then pairwise comparisons were performed.
In addition, the proportion of patients in each treatment
group who were able to improve from reporting “moderate”
to “extreme” pain on the WOMAC at baseline to reporting
“none” or “mild” pain at Week 12 were compared between
the total-daily-dose 200-mg celecoxib, 1000-mg naproxen,
and placebo groups. To test the differences in pain scores
between treatment groups, patients who experienced “mod-
erate,” “severe,” or “extreme” pain were combined. The per-
centages of patients who moved from these categories at
baseline to “none” or “mild” pain at Week 12 were calcu-
lated. The denominator of the proportion was the total num-
ber of patients in the categories of “moderate,” “severe,” or
“extreme” pain at baseline. The numerator was the differ-
ence in the total number of patients in the categories of
“moderate,” “severe,” or “extreme” pain at Week 12 and
baseline. This means that patients who moved into these
three categories from the “none” and “mild” pain categories
by the end of the treatment period were also considered. A
two-sample test for binomial proportions was used to deter-
mine treatment differences for each item (44). A similar
item analysis was conducted for the physical functioning
Table 1. Baseline Demographic Characteristics and Disease Activity
Celecoxib,
200 mg/day
n ϭ 191
Celecoxib,
400 mg/day
n ϭ 183
Naproxen,
1000 mg/day
n ϭ 206
Placebo
n ϭ 188
p-Value
Gender
Female
Age (y)
Mean (SD)
NS
66.0%
68.1%
63.9%
70.4%
NS
74.4 (3.9)
58.5%
30.9%
7.5%
75.0 (4.0)
51.3%
32.5%
14.7%
1.6%
75.0 (4.0)
50.3%
36.6%
9.8%
74.8 (4.1)
57.3%
30.1%
7.8%
70–74
75–79
80–84
85ϩ
3.2%
3.3%
4.9%
Race/Ethnic Origin
NS
NS
NS
White
93.1%
92.7%
88.0%
90.8%
Body Mass Index (Kg/m2)
Mean (SD)
Index Knee/Hip
29.1 (6.1)
28.7 (5.3)
28.7 (5.7)
29.2 (5.6)
Hip
Knee
Both
3.5%
82.5%
14.0%
4.7%
79.7%
15.6%
4.6%
80.0%
15.4%
3.0%
84.9%
12.1%
Disease Duration (y)
Mean (SD)
Functional Capacity Classification
NS
NS
9.4 (10.2)
10.0 (8.9)
10.3 (9.6)
10.4 (10.7)
I
II
III
1.6%
79.2%
19.1%
2.2%
76.9%
20.9%
2.3%
78.3%
19.4%
1.0%
72.2%
26.8%
OA Severity Index
Mean (SD)
Patient’s Global Assessment
NS
NS
14.7 (3.5)
15.2 (3.4)
15.4 (3.4)
15.4 (3.7)
Fair
Poor
Very poor
29.5%
60.7%
9.8%
26.9%
61.0%
12.1%
24.6%
63.4%
12.0%
23.7%
64.7%
11.6%
Patient’s Assessment of Arthritis Pain (0–100 mm VAS)
Mean (SD)
Physician’s Global Assessment
NS
NS
66.1 (16.7)
68.0 (16.4)
67.0 (16.2)
69.1 (16.3)
Fair
Poor
Very poor
30.1%
64.5%
5.5%
35.2%
57.1%
7.7%
30.3%
61.1%
8.6%
27.3%
63.1%
9.6%
Notes: No statistically significant difference between treatment groups was observed. NS ϭ not significant; OA ϭ osteoarthritis; VAS ϭ visual analog scale.

M170
LISSE ET AL.
Table 2. WOMAC and SF-36 Scores at Baseline
Treatment Groups
Celecoxib,
200 mg/day
Celecoxib,
400 mg/day
Naproxen,
1000 mg/day
Placebo
Measures
Mean
SD
Mean
SD
Mean
SD
Mean
SD
p Value
WOMAC†
Pain
9.9
4.4
33.6
46.7
3.2
1.4
11.3
15.8
10.3
4.6
35.0
48.9
3.3
1.4
10.7
15.3
10.3
4.7
35.5
49.5
3.2
1.4
11.1
16.0
10.4
4.6
34.5
48.7
3.4
1.5
11.1
15.7
NS
NS
NS
NS
Stiffness
Physical functioning
Composite score
SF-36‡
Physical functioning
Role physical
Bodily pain
General health
Vitality
Social functioning
Role emotional
Mental health
Physical component score§
Mental component score§
34.0
19.5
34.3
63.3
43.4
65.8
48.8
71.9
29.6
49.9
21.8
29.8
16.2
18.5
19.9
26.2
43.8
18.5
8.0
33.3
16.4
32.3
63.0
40.7
60.5
44.0
70.9
29.3
48.4
20.2
28.1
14.5
19.5
19.5
26.6
43.7
17.6
6.6
33.6
21.9
35.7
61.7
44.1
68.1
49.2
71.5
30.0
50.1
20.7
34.3
16.5
17.8
18.5
26.6
44.2
17.3
7.8
31.8
22.3
33.4
62.1
40.0
68.2
52.4
72.4
29.0
50.5
20.9
33.8
17.0
19.6
19.2
27.6
44.9
19.2
7.9
NS
NS
NS
NS
NS
Ͻ.05
NS
NS
NS
NS
10.7
11.1
10.7
11.4
Note: WOMAC ϭ Western Ontario and McMaster Universities Osteoarthritis Index; SF-36 ϭ Short Form-36; NS ϭ not significant.
†Lower scores represent better functional status.
‡Higher scores represent better health-related quality of life.
§Standardized with mean ϭ 50 and SD ϭ 10.
subscale to calculate the percentage of patients who im-
proved from the more limited physical functioning catego-
ries to the less limited physical functioning categories.
summary scores of the SF-36 compared with patients in the
placebo group (Table 3). There was no difference in change
scores among the 200-mg-daily celecoxib, 400-mg-daily
celecoxib, and naproxen groups.
At Week 12, the two celecoxib groups and the naproxen
group had statistically significant improvement on the
WOMAC pain, stiffness, physical functioning, and total
scores compared with the placebo-treated group (p Ͻ
0.001; Table 4). These improvements were not significantly
different between the celecoxib groups and the naproxen
group. Similarly, the HRQOL scores among patients in the
200-mg- and 400-mg-daily celecoxib groups, as well as the
naproxen group, showed significant improvement in six
(i.e., physical functioning, role physical, bodily pain, vital-
ity, social functioning, and mental health) of the eight SF-36
domains and the PCS, compared with patients in the pla-
cebo group (Table 4). In addition, the two celecoxib groups
showed significant change over the placebo group on the
general health domain. The 200-mg-daily celecoxib group
also showed significant improvement in the MCS compared
with the placebo-treated group.
Another way of evaluating the impact on aspects measured
by the WOMAC is to consider the proportion of patients who
improved from a “moderate,” “severe,” or “extreme” pain or
difficulty category to a “none” or “mild” pain or difficulty
category after treatment (Table 5). At Week 12, 33% (range,
21%–39%) of patients in the 200-mg-daily celecoxib group
and 29% (range, 18%–41%) of patients in the 1000-mg-daily
naproxen group improved from the “moderate,” “severe,” or
“extreme” pain categories to the “none” or “mild” pain cate-
gories for each of the five WOMAC pain items, compared
with 15% (range, 10%–21%) in the placebo group. The least
frequent activity to achieve a “none” or “mild” rating was for
RESULTS
Descriptive Statistics
The study had a total of 768 patients with OA of the knee
and/or hip who were 70 years of age or older. The baseline
demographic characteristics and disease activity of the pa-
tients are summarized in Table 1. There were no statistically
significant differences between the four treatment groups
with respect to demographic characteristics or disease activ-
ity at baseline.
Functional Status and HRQOL Evaluations
Baseline WOMAC and SF-36 scores showed no statisti-
cally significant differences between the treatment groups
except for the social functioning domain of the SF-36 (p Ͻ
.05; Table 2). The social functioning score in the 200-mg-
daily celecoxib group was the lowest (60.5), while scores
in the placebo, 400-mg-daily celecoxib, and 1000-mg-daily
naproxen groups were 65.8, 68.1, and 68.2, respectively.
At Week 2, the 200-mg- and 400-mg-daily celecoxib
groups and the 1000-mg-daily naproxen group showed sta-
tistically significant improvement in the WOMAC pain,
stiffness, physical functioning, and composite scores com-
pared with the placebo-treated group (p Ͻ .001; Table 3).
No significant differences between the celecoxib groups
and the naproxen group were observed. Similarly, the gen-
eral HRQOL among patients in the 200-mg- and 400-mg-
daily celecoxib groups and the naproxen group showed
significant improvement in all eight domains and the two

CELECOXIB IN OSTEOARTHRITIC PATIENTS
M171
Table 3. Change From Baseline in WOMAC and SF-36 Scores at Week 2
Treatment Groups
Celecoxib,
200 mg/day
n ϭ 191
Celecoxib,
400 mg/day
n ϭ 183
Naproxen,
1000 mg/day
n ϭ 206
Placebo
n ϭ 188
Measures
p Value
WOMAC†
Pain
Ϫ0.7
Ϫ0.4
Ϫ1.6
Ϫ2.4
Ϫ2.6
Ϫ1.1
Ϫ8.4
Ϫ2.4
Ϫ1.0
Ϫ7.5
Ϫ2.3
Ϫ1.0
Ϫ6.7
.0001
.0001
.0001
.0001
Stiffness
Physical functioning
Composite score
SF-36‡
Ϫ11.9
Ϫ10.8
Ϫ10.1
Physical functioning
Role physical
Bodily pain
General health
Vitality
Social functioning
Role emotional
Mental health
Physical component score§
Mental component score§
2.2
5.9
4.7
Ϫ0.1
Ϫ0.5
0.3
5.7
0.3
1.4
0.2
7.5
17.8
14.0
2.6
6.9
14.1
14.3
1.8
6.5
6.9
13.6
3.8
3.8
8.3
19.6
15.8
1.7
8.0
8.9
15.2
4.0
4.8
.0005
.0001
.0001
.0416
.0001
.0001
.0154
.0048
.0001
.0051
8.4
10.9
14.8
3.5
4.4
3.1
2.4
2.5
Notes: All values are statistically significantly different from placebo. WOMAC ϭ Western Ontario and McMaster Universities Osteoarthritis Index; SF-36 ϭ
Short Form-36.
†Negative change scores indicate improvement in functional status.
‡Positive change scores indicate improvements in health-related quality of life.
§Standardized with mean ϭ 50 and SD ϭ 10.
pain experienced while going up or down stairs. The 200-mg-
daily celecoxib group was significantly different from the
placebo-treated group on all items and had a larger percent-
age of change on four of the five items in comparison with
the naproxen group, although these differences were not sta-
tistically significant. The naproxen group showed significant
differences from the placebo group on four of the five items.
A similar item level analysis was conducted for the phys-
ical functioning items (Table 5). The reduction in the per-
centage of patients in the “moderate,” “severe,” or “ex-
Table 4. Change From Baseline in WOMAC and SF-36 Scores at Week 12
Treatment Groups
Celecoxib,
Celecoxib,
Naproxen,
Placebo
200 mg/day
400 mg/day
1000 mg/day
p Value
WOMAC†
Pain
Ϫ1.0
Ϫ0.3
Ϫ2.4
Ϫ3.4
Ϫ2.7*
Ϫ1.1*
Ϫ8.2*
Ϫ11.5*
Ϫ2.4*
Ϫ1.0*
Ϫ6.5*
Ϫ9.5*
Ϫ2.3*
Ϫ1.0*
Ϫ7.2*
Ϫ10.4*
Ͻ.001
Ͻ.001
Ͻ.001
Ͻ.001
Stiffness
Physical functioning
Composite score
SF-36‡
Physical functioning
Role physical
Bodily pain
General health
Vitality
Social functioning
Role emotional
Mental health
Physical component score§
Mental component score§
1.8
8.1
3.7
Ϫ0.9
Ϫ0.6
Ϫ2.4
3.1
1.3
1.8
Ϫ1.1
8.9*
17.9*
13.1*
2.7*
7.3*
5.6*
9.3
3.2*
4.8*
1.5*
7.6*
14.8*
12.0*
2.3*
5.0*
5.4*
7.3
2.2*
4.4*
0.8
9.5*
17.0*
14.2*
0.6
6.7*
5.1*
7.2
5.0*
5.0*
0.3
Ͻ.001
Ͻ.001
Ͻ.001
Ͻ.05
Ͻ.001
Ͻ.001
NS
Ͻ.01
Ͻ.01
Ͻ.01
Note: WOMAC ϭ Western Ontario and McMaster Universities Osteoarthritis Index; SF-36 ϭ Short Form-36; NS ϭ not significant.
*Statistically significant difference from placebo; p Ͻ .05.
†Negative change scores indicate improvement in functional status.
‡Positive change scores indicate improvements in health-related quality of life.
§Standardized with mean ϭ 50 and SD ϭ 10.

M172
LISSE ET AL.
Table 5. Patients’ Shift From WOMAC Categories of “Moderate,” “Severe,” or “Extreme” to “None” or “Mild” at Week 12 Compared
With Baseline
Treatment Groups
Placebo
Celecoxib, 200 mg/day
Reduction,
Naproxen, 1000 mg/day
Reduction,
Reduction,
Percent
WOMAC Items
No. of Cases†
No. of Cases
Percent
No. of Cases
Percent
Pain Subscale
How much pain do you have?
Walking on a flat surface
Going up or down stairs
At night while in bed
Sitting or lying
140
163
128
118
141
20.7
9.8
13.3
19.5
12.8
148
177
126
115
148
34.5*
20.9*
38.9*
37.4*
33.1*
155
186
140
125
160
27.1
18.3*
41.4*
31.2*
26.3*
Standing upright
Physical Functioning Subscale
What degree of difficulty do you have?
Descending stairs
141
154
162
143
158
124
149
152
131
144
107
99
125
97
113
162
118
6.4
6.5
157
164
165
141
155
136
154
153
128
150
111
111
140
98
21.0*
20.7*
25.5*
30.5*
20.6*
37.5*
27.3*
20.9*
30.5*
34.7*
31.5*
43.2*
26.4*
34.7*
32.8*
18.2*
26.8*
171
181
176
154
175
133
169
163
134
153
116
120
144
107
126
184
133
24.6*
22.7*
21.0*
24.0
Ascending stairs
Rising from sitting
Standing
11.7
17.5
10.1
14.5
10.1
7.9
11.5
14.6
12.1
13.1
4.0
Bending to floor
21.1*
28.6*
29.0*
19.6*
20.9*
30.7*
25.0*
39.2*
34.0*
36.4*
38.1*
13.5*
28.6*
Walking on flat surface
Getting in/out of car
Going shopping
Putting on socks/stockings
Rising from bed
Taking off socks/stockings
Lying in bed
Getting in/out bath
Sitting
11.3
5.3
2.5
Getting on/off toilet
Heavy domestic duties
Light domestic duties
122
176
123
1.7
Note: WOMAC ϭ Western Ontario and McMaster Universities Osteoarthritis Index.
*Statistically significant difference from placebo; p Ͻ .05.
†Number of moderate, severe, or extreme cases at baseline.
treme” difficulty categories for the 17 physical functioning
items in the placebo, 200-mg-daily celecoxib, and naproxen
treatment groups was 9% (range, 2%–18%), 28% (range,
18%–43%), and 27% (range, 14%–39%), respectively. The
200-mg-daily celecoxib group was significantly different from
the placebo group on all physical functioning items. On 9 of
the 17 physical functioning items, higher percentages of pa-
tients moved to the less severe category in the 200-mg-daily
celecoxib treatment group than in the naproxen-treated group,
but these differences were not statistically significant. In the
naproxen group, with the exception of one item (i.e., stand-
ing), all other items were found to be statistically signifi-
cantly different from the placebo-treated group.
200-mg-daily celecoxib, 119 (59.5%) 400-mg-daily cele-
coxib, and 121 (58.7%) 1000-mg-daily naproxen group pa-
tients. Only the 200-mg-daily celecoxib group showed a sta-
tistically significant difference from the placebo group with
regard to the incidence of nonserious adverse events. With re-
spect to the overall incidence of GI-related adverse events,
there were no significant differences between the two cele-
coxib groups and the placebo group. In contrast, statistically
significant differences were found between the naproxen
group and the placebo group, and the naproxen group and the
200-mg-daily celecoxib group. No differences were observed
between the treatment groups among other more commonly
reported non-GI-related adverse events. Differences in the
rates of withdrawal due to adverse events in the two cele-
coxib groups and the naproxen group were not statistically
significant from the placebo-treated group.
Safety Evaluations
Both doses of celecoxib were well tolerated. In total, 92
(48.9%), 125 (65.4%), 128 (64.0%), and 125 (60.7%) pa-
tients in the placebo, 200-mg-daily celecoxib, 400-mg-daily
celecoxib, and 1000-mg-daily naproxen groups, respectively,
reported at least one adverse event (Table 6). Serious adverse
events occurred in eight (4.3%) placebo, seven (3.7%) 200-
mg-daily celecoxib, nine (4.9%) 400-mg-daily celecoxib, and
four (1.9%) 1000-mg-daily naproxen group patients, but dif-
ferences were not statistically significant. Nonserious adverse
events were reported in 84 (44.7%) placebo, 118 (61.8%)
DISCUSSION
It has become increasingly important to study the effects
of medications in the elderly population because demo-
graphic changes in the United States over the past century
have led to an increased attention to special needs of this
population within the health care system. Over the next de-
cade, the number of people over the age of 65 is expected to
increase substantially, with the elderly population estimated

CELECOXIB IN OSTEOARTHRITIC PATIENTS
M173
Table 6. Comparison of Adverse Events and Withdrawal Rates
Placebo
Celecoxib, 200 mg/day
Celecoxib, 400 mg/day
Naproxen, 1000 mg/day
Characteristics
n ϭ 188
n ϭ 191
n ϭ 183
n ϭ 206
Adverse Events (AE)
Incidence of AEs
Patients with at least one serious AE
8 (4.3%)
7 (3.7%)
9 (4.9%)
4 (1.9%)
Patients with at least one nonserious AE
Incidence of GI-related AEs¶
Abdominal pain
84 (44.7%)
118 (61.8%)*
119 (59.5%)
121 (58.7%)
7 (3.7%)
3 (1.6%)
11 (5.9%)
13 (6.9%)
3 (1.6%)
3 (1.6%)
1 (0.5%)
0 (0.0%)
32 (17.0%)
10 (5.2%)
3 (1.6%)
17 (8.9%)
16 (8.4%)
2 (1.0%)
10 (5.2%)
0 (0.0%)
3 (1.6%)
11 (6.0%)
7 (3.8%)
16 (8.7%)
11 (6.0%)
3 (1.6%)
7 (3.8%)
2 (1.1%)
2 (1.1%)
37 (20.2%)
14 (6.8%)
9 (4.4%)
13 (6.3%)
25 (12.1%)
8 (3.9%)
15 (7.3%)*
0 (0.0%)
5 (2.4%)*
62 (30.1%)* ‡
Constipation
Diarrhea
Dyspepsia
Flatulence
Nausea
Stomatitis
Vomiting
Overall incidence
Incidence of non-GI-related AEs#
Headache
49 (25.7%)
11 (5.9%)
9 (4.8%)
2 (1.1%)
3 (1.6%)
4 (2.1%)
24 (12.6%)
5 (2.6%)
6 (3.1%)
5 (2.6%)
4 (2.1%)
18 (9.8%)
20 (10.9%)
10 (5.5%)
6 (3.3%)
20 (9.7%)
15 (7.3%)
5 (2.4%)
5 (2.4%)
0 (0.0%)
Upper respiratory tract infection
Peripheral edema
Dizziness
Skin rash
6 (3.3%)
Withdrawal From Study
Patients withdraw from study**
Lost to follow-up
Preexisting violation
Protocol noncompliance
Treatment failure
Withdrawal due to AEs
Incidence of GI-related AEs caused withdrawal¶
Abdominal pain
1 (0.5%)
4 (2.1%)
5 (2.7%)
89 (47.3%)
18 (9.6%)
1 (0.5%)
0 (0.0%)
7 (3.7%)
47 (24.6%)‡
29 (15.2%)
0 (0.0%)
0 (0.0%)
5 (2.7%)
42 (23.0%)ʈ
27 (14.8%)
1 (0.5%)
2 (1.0%)
5 (2.4%)
57 (27.7%)ʈ
25 (12.1%)
1 (0.5%)
0 (0.0%)
1 (0.5%)
0 (0.0%)
0 (0.0%)
0 (0.0%)
2 (1.1%)
3 (1.6%)
3 (1.6%)
3 (1.6%)
0 (0.0%)
0 (0.0%)
0 (0.0%)
9 (4.7%)*
2 (1.1%)
2 (1.1%)
1 (0.5%)
2 (1.1%)
0 (0.0%)
0 (0.0%)
6 (3.3%)
8 (3.9%)*
2 (1.0%)
3 (1.5%)
Diarrhea
Dyspepsia
Flatulence
Nausea
Vomiting
Overall incidence
0 (0.0%)
5 (2.4%)*‡§
1 (0.5%)
16 (7.8%)*‡
Note: GI ϭ gastrointestinal.
*Statistically significant difference from placebo; p Ͻ .05.
‡Statistically significant difference from 200 mg daily of celecoxib; p Ͻ .05.
§Statistically significant difference from 400 mg daily of celecoxib; p Ͻ .05.
ʈStatistically significant difference from 1000 mg daily of naproxen; p Ͻ .05.
¶Statistical tests were based on log-rank tests for time to withdrawal.
#These are the most common non-GI-related AEs reported by the patients.
**If a patient had more than one AE causing withdrawal within the GI system, the patient was only counted once in the overall incidence.
to constitute 13.2% of the U.S. population in 2010 (45). This
population experiences different reactions to medications,
has more comorbidities, and needs more medications. For
these reasons, it was important to study the effect of treat-
ment on the elderly patients suffering from a chronic condi-
tion such as OA. The results of this study using pooled data
of elderly patients from three large multicenter trials show
that celecoxib not only improves functional status and
HRQOL, but is also a safe and well-tolerated treatment op-
tion for OA of the knee and/or hip.
The importance of pain relief and improvement in physi-
cal functioning is highlighted with the improvement in the
ability of patients to care for themselves. Both celecoxib
and naproxen helped patients reduce their pain while per-
forming physical activities such as walking on a flat surface,
going up and down stairs, and standing upright. The medi-
cations also helped them to perform daily activities, such as
rising from a chair, bending, and getting in and out of cars
with more ease. In addition to the improvement observed in
overall pain, stiffness, and physical functioning subscale
scores, it is important to note that improvement was ob-
served across all individual items contributing to the sub-
scale scores, signifying an improvement on all aspects of
functional status. The ramifications of such improvements
may include patients being able to participate in more phys-
ical activities, such as scheduled exercise programs de-
signed to help retard the progress of the disease condition. A
recent study has shown that exercise and physical therapy
can dramatically decrease the pain associated with arthritis
and even prevent the need for surgery (46).
Improvements in HRQOL are important because popula-
tion studies have shown that elderly patients with OA report
significantly lower functioning when compared with healthy
controls and patients with other chronic conditions (47).

M174
LISSE ET AL.
Therapeutic interventions with celecoxib and naproxen re-
sulted in significant improvements in HRQOL when com-
pared with placebo, as measured by the SF-36. These im-
provements in HRQOL mean that elderly patients may be
able to continue with their physical activities, maintain their
desired level of independence, and therefore experience an
overall more enriched life.
elderly patients with OA compared with placebo treatment
and that celecoxib is similar to naproxen in its improve-
ments in functional status and HRQOL. Celecoxib, unlike
naproxen, was also found to be safe to use among this popu-
lation with no significant difference in adverse events be-
tween the two celecoxib and placebo groups.
Improvements in functional status and HRQOL are not
only important to patients’ levels of independence and dis-
ability, but also have economic implications; decrements in
functional status and HRQOL have been shown to be asso-
ciated with increased use of medical resources (48). In addi-
tion to concern about patient outcomes, payers may be inter-
ested in functional status and HRQOL assessments because
they can help predict patients’ resource utilization patterns.
If resource use can be diminished, as would be expected
with higher levels of HRQOL experienced by patients in the
active treatment groups, payers may experience important
budget impacts. Improved HRQOL would also increase
chances of independent living, which would have additional
payer impact. Overall, treatments that provide improved
functional status and relief from symptoms may reduce the
economic burden placed on payers, while improving the
level of patients’ physical activity. Patients and physicians
are also interested in functional status and HRQOL assess-
ments because they help in choosing between treatment op-
tions; functional status and HRQOL are impacted by both
efficacy and safety of treatment options. Improved treat-
ments and evaluations for the management of OA are im-
portant, especially in this subpopulation.
Acknowledgments
This study was supported by Pharmacia Corporation and Pfizer, Inc. We
thank Somali Misra Burgess, PhD, and Steven P. Burch, PhD, of Strategic
Outcomes Services, Inc., for their assistance in preparing this manuscript. Re-
sults from this study were presented as a poster session entitled “Celecoxib
improves health-related quality of life (HRQOL) of elderly patients with os-
teoarthritis (OA)” by Sean Z. Zhao, Seema D. Dedhiya, Ken Verburg, and
Jane T. Osterhaus at the 63rd Annual Scientific Meeting of the American Col-
lege of Rheumatology, Boston, Massachusetts, November 13–17, 1999.
Address correspondence to Seema Dedhiya, Pharmacia Corporation,
5200 Old Orchard Road, Skokie, IL 60077. E-mail: dedhiya_seema@
hotmail.com
References
1. Brawn LA, Casteden CM. Adverse drug reactions: an overview of spe-
cial considerations in the management of the elderly patient. Drug
Safety. 1990;5:421–435.
2. Montamat S, Cusack B, Vestal RE, et al. Management of drug therapy
in the elderly. N Engl J Med. 1989;321:303–309.
3. Stewart RB, Cooper JW. Polypharmacy in the aged: practical solu-
tions. Drugs Aging. 1994;4:449–461.
4. Gabriel SE, Jaakkimainen L, Bombardier C. Risk for serious gas-
trointestinal complications related to use of nonsteroidal anti-
inflammatory drugs. Ann Intern Med. 1991;115:787–796.
5. Garcia Rodriguez LA, Jick H. Risk of upper gastrointestinal bleeding
and perforation associated with individual non-steroidal anti-inflam-
matory drugs. Lancet. 1994;343:769–772.
Results of this study also show that only the 200-mg-
daily celecoxib group was significantly different from the
placebo group with respect to the incidence of nonserious
adverse events. There was no significant difference between
the active treatment and placebo groups with respect to seri-
ous adverse events. The incidence of upper GI-related ad-
verse events is an important consideration for elderly per-
sons with OA because they are at an increased risk of these
adverse events (16). The incidence of GI-related adverse
events, as well as withdrawal from the study due to GI-
related adverse events, was significantly higher for patients
in the naproxen group compared with the placebo group.
However, the celecoxib groups did not experience signifi-
cant differences from the placebo-treated group, with re-
spect to the GI-related adverse event incidence or the indi-
vidual reasons for GI-related withdrawal from the study.
Study findings were consistent with safety and efficacy re-
sults that showed all doses of celecoxib to be safe and gener-
ally well tolerated (24,49,50). The magnitude of improve-
ment in the functional status and HRQOL of patients treated
with a daily dose of 200 mg of celecoxib was comparable to
that of a daily dose of 1000 mg of naproxen, but superior to
that of the placebo treatment results. In addition, the inci-
dence of GI-related adverse events was significantly less than
that in the naproxen group, but comparable to the placebo
group, making celecoxib a viable treatment option for OA.
6. March LM, Bachmeier CJM. Economics of osteoarthritis: a global per-
spective. Baillière’s Clin Rheumatol. 1997;11:817–834.
7. Hudgins TH, Brander VA, Chang RW. Rehabilitation advances in the
treatment of arthritis and musculoskeletal disease. Curr Opin Rheuma-
tol. 1997;9:112–117.
8. Bradley JD, Brandt KD, Katz BP, Kalasinski LA, Ryan SI. Compari-
son of an antiinflammatory dose of ibuprofen, an analgesic dose of
ibuprofen, and acetominophen in the treatment of patients with os-
teoarthritis of the knee. N Engl J Med. 1991;352:87–91.
9. Burch JW, Stanford N, Majerus PW. Inhibition of platelet prostaglan-
din synthetase by oral aspirin. J Clin Invest. 1978;61:314–319.
10. Carmichael J, Shankel SW. Effects of nonsteroidal anti-inflammatory
drugs on prostaglandins and renal function. Am J Med. 1985;78:992–
1000.
11. Gottlieb MS. Conservative management of spinal osteoarthritis with
glucosamine sulfate and chiropractic treatment. J Manipulative Phys-
iol Ther. 1997;20:400–414.
12. Harris C. Osteoarthritis: how to diagnose and treat the painful joint.
Geriatrics. 1993;48:39–46.
13. Houston MC. Nonsteroidal anti-inflammatory drugs and antihyperten-
sives. Am J Med. 1991;90:42S–47S.
14. Langman MJS, Weil J, Wainwright P, et al. Risks of bleeding peptic
ulcer associated with individual non-steroidal anti-inflammatory drugs.
Lancet. 1994;343:1075–1078.
15. Schlondorff D. Renal complications of nonsteroidal anti-inflammatory
drugs. Kidney Int. 1993;44:643–653.
16. Todesco S. Special considerations in the use of NSAIDs in the elderly.
Eur J Rheumatol Inflamm. 1994;14:7–13.
17. Vane JR. Inhibition of prostaglandin synthesis as a mechanism of ac-
tion for aspirin-like drugs. Nat N Biol. 1971;231:232–235.
18. Whelton A, Stout RL, Spilman PS, Klassen DK. Renal effects of ibu-
profen, piroxicam, and sulindac in patients with asymptomatic renal
failure. Ann Intern Med. 1990;112:568–576.
19. Mizuno K, Yamamoto S, Lands WEM. Effects of non-steroidal anti-
inflammatory drugs on fatty acid cyclooxygenase and prostaglandin
hydroperoxidase activities. Prostaglandins. 1982;23:743–757.
CONCLUSIONS
This study showed that celecoxib and naproxen signifi-
cantly improve the functional status and overall HRQOL of

CELECOXIB IN OSTEOARTHRITIC PATIENTS
M175
20. DeWitt DL, Meade EA, Smith WL. PGH synthase isoenzyme selectiv-
ity: the potential for safer nonsteroidal antiinflammatory drugs. Am J
Med. 1993;95:40S–44S.
36. Katz JN, Larson MG, Phillips CB, Fossel AH, Liang MH. Compara-
tive measurement sensitivity of short and longer health status instruments.
Med Care. 1992;30:917–925.
21. Masferrer JL, Zweifel BS, Manning PT, et al. Selective inhibition of
inducible cyclooxygenase 2 in vivo is antiinflammatory and nonul-
cerogenic. Proc Natl Acad Sci USA. 1994;91:3228–3232.
22. Mitchell JA, Akarasereenont P, Thiemermann C, Flower RJ, Vane JR.
Selectivity of nonsteroidal antiinflammatory drugs as inhibitors of con-
stitutive and inducible cyclooxygenase. Proc Natl Acad Sci USA. 1993;
90:11693–11697.
37. Kvien TK, Smedstad LM, Uhlig T. The responsiveness of generic and
disease specific health status measures in 759 patients with rheumatoid
arthritis (RA) [Poster 1393]. Arthritis Rheum. 1996;39(suppl):S260.
38. McHorney CA, Ware JE, Rachel 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.
23. Siebert K, Zhang Y, Leahy K, et al. Pharmacological and biochemical
demonstration of the role of cyclooxygenase 2 in inflammation and
pain. Proc Natl Acad Sci USA. 1994;91:12013–12017.
24. Celebrex (Celecoxib Capsules) [package insert]. Skokie, IL: GD
Searle & Co; 1998.
39. Ruta DA, Hurst NP, Kind P, Hunter M, Stubbings A. Measuring
health status in British patients with rheumatoid arthritis: reliability,
validity and responsiveness of the Short Form 36-item health survey
(SF-36). Br J Rheumatol. 1998;37:425–436.
40. Talamo J, Frater A, Gallivan S, Young A. Use of the Short Form 36
(SF36) for health status measurement in rheumatoid arthritis. Br J Rheu-
matol. 1997;36:463–469.
41. Ware JE, Kosinski M, Hatoum HT, Kong SX. The SF-36 health sur-
vey (SF-36) as a generic outcome measure in clinical trials of patients
with osteo- and rheumatoid arthritis: relative validity of scales in rela-
tion to clinical measures of arthritis severity [Poster 1383]. Arthritis
Rheum. 1996;39(suppl):S258.
42. Ware JE, Kosinski M, Keller SD. SF-36 Physical and Mental Health
Summary Scales: A User’s Manual. Boston, MA: Health Assessment
Lab; 1994.
25. Barr S, Bellamy N, Buchanan WW, et al. A comparative study of sig-
nal versus aggregate methods of outcome measurement based on the
WOMAC Osteoarthritis Index. J Rheumatol. 1994;21:2106–2112.
26. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW.
Validation study of WOMAC: a health status instrument for measur-
ing clinically important patient relevant outcomes to antirheumatic
drug therapy in patients with osteoarthritis of the hip or knee. J Rheu-
matol. 1988;15:1833–1840.
27. Bellamy N, Kean WF, Buchanan WW, Gerecz-Simon E, Campbell J.
Double blind randomized controlled trial of sodium meclofenamate
(Meclomen) and diclofenac sodium (Voltaren): post validation reap-
plication of the WOMAC Osteoarthritis Index. J Rheumatol. 1992;19:
153–159.
43. Siddiqui O, Ali MW. A comparison of the random-effects pattern mix-
ture model with last-observation-carried-forward (LOCF) analysis in
longitudinal clinical trials with dropouts. J Biopharm Stat. 1998;8:
545–563.
44. Rosner B. Fundamentals of Biostatistics. 4th ed. Belmont, CA:
Wadsworth Publishing; 1995.
28. Kreibich DN, Vaz M, Bourne RB, et al. What is the best way of as-
sessing outcome after total knee replacement? Clin Orthop. 1996;331:
221–225.
29. Laupacis A, Bourne R, Rorabeck C, et al. The effect of elective total
hip replacement on health-related quality of life. J Bone Joint Surg.
1993;75A:1619–1626.
30. Martin DP, Engelberg R, Agel J, Swiontkowski MF. Comparison of
the Musculoskeletal Function Assessment questionnaire with the Short
Form-36, the Western Ontario and McMaster Universities Osteoarthri-
tis Index, and the Sickness Impact Profile health-status measures. J
Bone Joint Surg. 1997;79A:1323–1335.
31. Mazzuca SA, Brandt KD, Katz BP, et al. Comparison of general inter-
nists, family physicians, and rheumatologists managing patients with
symptoms of osteoarthritis of the knee. Arthritis Care Res. 1997;10:
289–299.
45. U.S. Bureau of the Census. Population Projections of the United States
by Age, Sex, Race, and Hispanic Origin: 1995 to 2050. Washington
DC: U.S. Bureau of the Census; 1998. Current population reports, se-
ries P25-1130.
46. Deyle GD, Henderson NE, Matekel RL, Ryder MG, Garber MB, Alli-
son SC. Effectiveness of manual physical therapy and exercise in
osteoarthritis of the knee. Ann Intern Med. 2000;132:173–181.
47. Ware J. SF-36 Health Survey: Manual and Interpretation Guide. Bos-
ton, MA: The Health Institute, New England Medical Center; 1993.
48. Cronan TA, Shaw WS, Gallagher RA, Weisman M. Predicting health
care use among older osteoarthritis patients in an HMO. Arthritis Care
Res. 1995;8:66–72.
32. McGrory BJ, Freiberg AW, Shinar AA, Harris WH. Correlation of
measured range of hip motion following total hip arthroplasty and re-
sponses to a questionnaire. J Arthroplasty. 1996;11:565–571.
33. Slemenda C, Brandt KD, Heilman DK, et al. Quadriceps weakness and
osteoarthritis of the knee. Ann Intern Med. 1997;127:97–104.
34. Stucki G, Sangha O, Stucki S, et al. Comparison of the WOMAC
(Western Ontario and McMaster Universities) osteoarthritis index and
a self-report format of the self-administered Lequesne-Algofunctional
index in patients with knee and hip osteoarthritis. Osteoarthritis Carti-
lage. 1998;6:79–86.
49. Simon LS, Weaver AL, Graham DY, et al. Anti-inflammatory and up-
per gastrointestinal effects of celecoxib in rheumatoid arthritis. JAMA.
1999;282:1921–1928.
50. Bensen WG, Zhao SZ, Burke TA, et al. Upper gastrointestinal tolera-
bility of celecoxib, a COX-2 specific inhibitor, compared to naproxen
and placebo. J Rheumatol. 2000;27:1876–1883.
Received August 14, 2000
35. Bellamy N. WOMAC Osteoarthritis Index: A User’s Guide. London,
Ontario: N. Bellamy; 1995.
Accepted August 25, 2000
Decision Editor: John E. Morley, MB, BCh