After applying the inclusion and exclusion criteria, 7078 patients in the AOM cohort and 91,137 patients in the non-AOM cohort were identified.
The average patient age was 72.03 years in the AOM group and 74.25 years in the non-AOM group (p<.0001). High-comorbidity patients were defined as those with a comorbidity score of 2 and above. Individuals in the AOM cohort had a significantly higher proportion of patients with a high CDS (81.41% vs. 37.03%, p<.0001), high CCI (60.94% vs. 37.53%, p<.0001), and a high Elixhauser Index (93.60% vs. 75.97%, p<.0001). The AOM cohort was also significantly more likely to live in low-SES regions (34.90% vs. 32.85%, p<.001) and significantly less likely to live in medium-SES regions than the non-AOM cohort (31.25% vs. 33.19%, p=.001). The most prevalent baseline OA-related comorbidities in the AOM and non-AOM groups, respectively, were hypertension (81.53% vs. 70.37%, p<.0001) and diabetes (79.10% vs. 35.73%, p<.0001). COPD (17.94% vs. 16.55%, p<.05), congestive heart failure (13.99% vs. 12.35% p=.0001), myocardial infarction (2.25% vs. 1.74%, p<.05), peripheral vascular disease (13.01% vs. 11.78%, p<.05), and depression (15.50% vs. 11.60%, p<.0001) was also significantly higher in the AOM cohort (Table 1).
Table 1. Baseline Characteristics of the AOM and Non-AOM Cohorts
Characteristics
|
With AOM (Ozempic or Wegovy) (N = 7078)
|
Without AOM (N = 91,137)
|
P Value
|
Std
|
N/Mean
|
%/Std
|
N/Mean
|
%/Std
|
Age group (years)
|
72.03
|
4.74
|
74.25
|
5.87
|
<.0001*
|
0.3836
|
65-70
|
3,211
|
45.37%
|
29,003
|
31.82%
|
<.0001*
|
0.2893
|
71-80
|
3,428
|
48.43%
|
47,123
|
51.71%
|
<.0001*
|
0.0655
|
≥80
|
439
|
6.20%
|
15,011
|
16.47%
|
<.0001*
|
0.2828
|
Gender
|
|
|
|
|
|
|
Male
|
3,238
|
45.75%
|
42,008
|
46.09%
|
.5739
|
0.0069
|
Female
|
3,840
|
54.25%
|
49,129
|
53.91%
|
.5739
|
0.0069
|
Comorbidity Scores
|
|
|
|
|
|
|
CCI ≥2
|
4,313
|
60.94%
|
34,203
|
37.53%
|
<.0001*
|
0.4831
|
CDS ≥2
|
5,762
|
81.41%
|
33,744
|
37.03%
|
<.0001*
|
0.9310
|
Elixhauser Index ≥2
|
6,625
|
93.60%
|
69,236
|
75.97%
|
<.0001*
|
0.4230
|
SES
|
|
|
|
|
|
|
Low
|
2,423
|
34.90%
|
29,385
|
32.85%
|
.0005*
|
0.0435
|
Medium
|
2,170
|
31.25%
|
29,685
|
33.19%
|
.0010*
|
0.0411
|
High
|
2,350
|
33.85%
|
30,375
|
33.96%
|
.8489
|
0.0024
|
Baseline OA-related comorbidities
|
|
|
|
|
|
|
Hypertension
|
5,771
|
81.53%
|
64,130
|
70.37%
|
<.0001*
|
0.2470
|
Congestive heart failure
|
990
|
13.99%
|
11,255
|
12.35%
|
.0001*
|
0.0496
|
Myocardial infarction
|
159
|
2.25%
|
1,583
|
1.74%
|
.0018*
|
0.0386
|
Cerebrovascular disease
|
294
|
4.15%
|
3,384
|
3.71%
|
.0600
|
0.0232
|
Peripheral vascular disease
|
921
|
13.01%
|
10,736
|
11.78%
|
.0020*
|
0.0381
|
Diabetes
|
5,599
|
79.10%
|
32,566
|
35.73%
|
<.0001*
|
0.9143
|
COPD
|
1,270
|
17.94%
|
15,087
|
16.55%
|
.0025*
|
0.0373
|
Depression
|
1,097
|
15.50%
|
10,576
|
11.60%
|
<.0001*
|
0.1204
|
*Significant at p<.05.
Abbreviations: CCI, Charlson Comorbidity Index; COPD, chronic obstructive pulmonary disease; OA, osteoarthritis; SES, socioeconomic status; Std, standardized difference.
Comparison of the Ozempic and Wegovy cohorts indicated that Ozempic users were older (72.08 vs. 70.93, p<.0001), had a higher proportion of males (46.38% vs. 33.33%, p<.0001), had a higher CDS (82.25% vs. 64.93%, p<.0001), a higher CCI (62.65% vs. 27.54%, p<.0001), and a higher Elixhauser Index (94.07% vs. 84.35%, p<.0001). Patients in the Ozempic cohort were significantly less likely to live in a high SES region (33.50% vs. 40.72%, p=.0065). Some of the most prevalent comorbidities in both cohorts included hypertension (81.97% vs. 73.04%, p<.001) and diabetes (82.03% vs. 22.03%, p<.001). Patients in the Ozempic cohort had a significantly higher prevalence of congestive heart failure (14.35% vs. 6.96%, p<.001) and peripheral vascular disease (13.34% vs. 6.67%, p=.0003) prior to initiation of the medication use (Table 2).
Table 2. Baseline Characteristics of the Ozempic and Wegovy Cohorts
Characteristics
|
Ozempic (N = 6,733)
|
Wegovy (N = 345)
|
P Value
|
Std
|
N/Mean
|
%/Std
|
N/Mean
|
%/Std
|
Age group (years)
|
72.08
|
4.76
|
70.93
|
4.20
|
<.0001*
|
0.2442
|
65-70
|
3,016
|
44.79%
|
195
|
56.52%
|
<.0001*
|
0.2358
|
71-80
|
3,287
|
48.82%
|
141
|
40.87%
|
.0040*
|
0.1591
|
≥80
|
430
|
6.39%
|
9
|
2.61%
|
.0045*
|
0.1567
|
Gender
|
|
|
|
|
|
|
Male
|
3,123
|
46.38%
|
115
|
33.33%
|
<.0001*
|
0.2623
|
Female
|
3,610
|
53.62%
|
230
|
66.67%
|
<.0001*
|
0.2623
|
Comorbidity Scores
|
|
|
|
|
|
|
CCI ≥2
|
4,218
|
62.65%
|
95
|
27.54%
|
<.0001*
|
0.7283
|
CDS ≥2
|
5,538
|
82.25%
|
224
|
64.93%
|
<.0001*
|
0.4473
|
Elixhauser Index ≥2
|
6,334
|
94.07%
|
291
|
84.35%
|
<.0001*
|
0.3988
|
SES
|
|
|
|
|
|
|
Low
|
2,197
|
33.24%
|
96
|
28.74%
|
.0880
|
0.0957
|
Medium
|
2,198
|
33.26%
|
102
|
30.54%
|
.3030
|
0.0578
|
High
|
2,214
|
33.50%
|
136
|
40.72%
|
.0065*
|
0.1526
|
Baseline OA-related comorbidities
|
|
|
|
|
|
|
Hypertension
|
5,519
|
81.97%
|
252
|
73.04%
|
<.0001*
|
0.2303
|
Congestive heart failure
|
966
|
14.35%
|
24
|
6.96%
|
.0001*
|
0.2133
|
Myocardial infarction
|
156
|
2.32%
|
3
|
0.87%
|
.0768
|
0.0977
|
Cerebrovascular disease
|
283
|
4.20%
|
11
|
3.19%
|
.3569
|
0.0509
|
Peripheral vascular disease
|
898
|
13.34%
|
23
|
6.67%
|
.0003*
|
0.1984
|
Diabetes
|
5,523
|
82.03%
|
76
|
22.03%
|
<.0001*
|
1.5562
|
COPD
|
1,217
|
18.08%
|
53
|
15.36%
|
.2003
|
0.0707
|
Depression
|
1,031
|
15.31%
|
66
|
19.13%
|
.0560
|
0.1055
|
*Significant at p<.05.
Abbreviations: CCI, Charlson Comorbidity Index; COPD, chronic obstructive pulmonary disease; OA, osteoarthritis; SES, socioeconomic status; Std, standardized difference.
Table 3 provides Cox regression results for the risk of OA in the AOM and non-AOM cohorts. Among individuals with obesity, AOM (Wegovy and Ozempic) was associated with a 10% reduction in the risk of OA by 10% compared with those who were not on AOM (HR=0.90, p<.0001). Female patients were more likely to have OA than male patients. (HR=1.39, p<.0001). Patients between the ages of 65 and 70 had a 9% reduced risk of OA compared with those over 80 years of age (HR=0.91, p<.0001). CDS≥2 (HR=1.21, p<.001), peripheral vascular disease (HR=1.09, p<.0001), COPD (HR=1.09, p<.0001), and depression (HR=1.18, p<.0001) were associated with an increased risk of OA. Our results indicated that having diabetes was associated with a reduced risk of OA (HR=0.94, p<.0001). This finding is inconsistent with previous literature [27].
Table 3. Cox Regression for Risk of OA in AOM and Non-AOM Cohorts
|
HR
|
CI
|
P Value
|
Lower
|
Upper
|
Treatment
|
|
|
|
|
Yes
|
0.90
|
0.85
|
0.95
|
<.0001*
|
No
|
1.00
|
1.00
|
1.00
|
|
Age group (years)
|
|
|
|
|
65-70
|
0.91
|
0.87
|
0.95
|
<.0001*
|
71-80
|
0.98
|
0.94
|
1.02
|
.3369
|
≥80
|
1.00
|
1.00
|
1.00
|
|
Gender
|
|
|
|
|
Female
|
1.39
|
1.35
|
1.43
|
<.0001*
|
Male
|
1.00
|
1.00
|
1.00
|
|
CDS ≥2
|
1.21
|
1.18
|
1.25
|
<.0001*
|
SES
|
|
|
|
|
Low
|
1.01
|
0.98
|
1.05
|
.5304
|
Medium
|
0.98
|
0.94
|
1.01
|
.1907
|
High
|
1.00
|
1.00
|
1.00
|
|
Comorbidities
|
|
|
|
|
Hypertension
|
1.00
|
0.97
|
1.03
|
.9175
|
Congestive heart failure
|
0.96
|
0.92
|
1.01
|
.0858
|
Myocardial infarction
|
0.91
|
0.81
|
1.02
|
.0980
|
Cerebrovascular disease
|
0.93
|
0.86
|
1.00
|
.0645
|
Peripheral vascular disease
|
1.09
|
1.04
|
1.14
|
<.0001*
|
Diabetes
|
0.94
|
0.91
|
0.97
|
<.0001*
|
COPD
|
1.09
|
1.05
|
1.13
|
<.0001*
|
Depression
|
1.18
|
1.13
|
1.23
|
<.0001*
|
*Significant at p<.05.
Abbreviations: CCI, Charlson Comorbidity Index; CI, confidence interval; COPD, chronic obstructive pulmonary disease; HR, hazard ratio; OA, osteoarthritis; SES, socioeconomic status.
Aalen’s additive regression results are presented in Figure 2. Consistent with Cox regression, AOM treatment (Wegovy and Ozempic) significantly reduces the likelihood of OA. The treatment plot shows an initial reduction in the likelihood of OA during the first 550 to 600 days after initiation of treatment. After that period, the effect of treatment on the likelihood of OA is insignificant until after 900 days, when there is a protective treatment effect. The likelihood of OA increased for peripheral vascular disease and depression initially but then the effect disappeared. The likelihood of OA increases for female patients and patients with COPD and high CDS in the short and long run. The likelihood of OA decreases for patients with diabetes and younger patients also in the short and long run. The congestive heart failure coefficient was insignificant using the Cox regression. However, Aalen’s additive regression shows that the likelihood of OA decreases for congestive heart failure patients at a statistically significant level, which indicates that congestive heart failure could be a protective factor against OA.
Figure 2. Aelan’s Additive Regression Results for Risk of OA
The x-axis denotes the number of days; the y-axis denotes the magnitude of the effect.
Cox regression analysis showed no statistically significant differences in the risk of OA between the Ozempic and Wegovy cohorts (HR: 0.90, p=.4341; results not shown).
Table 4 shows Cox regression results for the risk of OA among the Ozempic and Wegovy groups. Patients taking Ozempic had a 10% decreased risk of OA compared with those taking Wegovy; however, this difference was not statistically significant (HR=0.90, p>.05). Additionally, females had a 38% elevated risk of OA compared with males (HR=1.38, p<.001). Peripheral vascular disease (HR=1.20, p<.05), COPD (HR=1.30, p<.0001), depression (HR=1.22, p<.05), and higher comorbidity scores (HR=1.21, p<.05) were associated with a significantly increased risk of OA in these cohorts.
Table 4. Cox Regression for OA Risk among Ozempic and Wegovy Cohorts
|
HR
|
CI
|
P Value
|
Lower
|
Upper
|
Ozempic
|
Yes
|
0.90
|
0.69
|
1.17
|
0.4341
|
No
|
1.00
|
1.00
|
1.00
|
|
Age group (years)
|
65-70
|
0.93
|
0.74
|
1.17
|
0.5503
|
71-80
|
0.93
|
0.75
|
1.17
|
0.5439
|
≥80
|
1.00
|
1.00
|
1.00
|
|
Gender
|
Female
|
1.38
|
1.23
|
1.54
|
<.0001*
|
Male
|
1.00
|
1.00
|
1.00
|
|
CDS ≥2
|
1.21
|
1.04
|
1.41
|
0.0137*
|
SES
|
Low
|
1.02
|
0.89
|
1.16
|
0.7886
|
Medium
|
1.03
|
0.90
|
1.18
|
0.6379
|
High
|
1.00
|
1.00
|
1.00
|
|
Comorbidities
|
Hypertension
|
1.23
|
1.05
|
1.43
|
0.0112*
|
Congestive heart failure
|
1.17
|
1.00
|
1.37
|
0.0460*
|
Myocardial infarction
|
0.77
|
0.52
|
1.14
|
0.1890
|
Cerebrovascular disease
|
1.15
|
0.90
|
1.48
|
0.2661
|
Peripheral vascular disease
|
1.20
|
1.02
|
1.40
|
0.0244*
|
Diabetes
|
0.97
|
0.83
|
1.13
|
0.6850
|
COPD
|
1.30
|
1.14
|
1.49
|
0.0001*
|
Depression
|
1.22
|
1.06
|
1.40
|
0.0063*
|
*Significant at p<.05.
Abbreviations: CDS, Chronic Disease Score; CI, confidence interval; COPD, chronic obstructive pulmonary disease; HR, hazard ratio; SES, socioeconomic status.