Skip to main content
. 2022 Apr 5;5(4):e224852. doi: 10.1001/jamanetworkopen.2022.4852

Table 3. Results of Generalized Estimating Equations With Repeated Measurements for Between-Group Differences Regarding Responders and Perceived Recovery Based on a Per-Protocol Analysis.

Characteristic No./total No. (%)a OR (95% CI)b P valuec
IM injection IA injection
OMERACT-OARSI responder criteria
2 wk 34/68 (50) 41/65 (63) 0.8 (0.4-1.8) .61
4 wk 33/69 (48) 45/66 (68) 0.6 (0.3-1.2) .12
8 wk 36/65 (55) 37/66 (56) 1.3 (0.6-2.7) .45
12 wk 27/67 (40) 36/65 (55) 0.7 (0.4-1.6) .43
24 wk 19/67 (28) 32/65 (49) 0.5 (0.2-1.1) .10
Perceived recovery
2 wk 41/68 (60) 50/65 (77) 0.6 (0.3-1.3) .18
4 wk 43/69 (62) 50/66 (76) 0.6 (0.3-1.3) .22
8 wk 40/65 (62) 41/66 (62) 1.0 (0.5-2.2) .90
12 wk 31/67 (46) 37/65 (57) 0.7 (0.3-1.5) .35
24 wk 27/67 (40) 29/65 (45) 0.9 (0.5-1.9) .82

Abbreviations: IA, intra-articular; IM, intramuscular; OR, odds ratio.

a

Observed and unadjusted values. There were no missing values in the model covariates, and all patients (138) were included in modeling.

b

Intra-articular injection as reference; adjusted for baseline Knee Injury and Osteoarthritis Outcome Score pain score, sex, presence of depression, and duration of knee osteoarthritis symptoms. Calculated based on estimated marginal percentages.

c

P values for superiority tests.