Skip to main content
. 1998 Nov 7;317(7168):1292–1296. doi: 10.1136/bmj.317.7168.1292

Table 3.

Mean (SD) improvement in outcome measures in patients with painful stiff shoulder and differences between groups by treatment at different points in follow up

Patients treated with corticosteroid injection* Patients treated with physiotherapy* Mean (95% CI) difference between groups P value
Improvement in rating of severity
Associated with main complaint:
 3 weeks 32 (26) 17 (21) 15 (6 to 24)     (0.071)
 7 weeks 58 (28) 32 (29)  26 (15 to 37)
 13 weeks 66 (28) 47 (33) 19 (7 to 31)
 26 weeks 63 (31) 54 (33)   9 (−3 to 22)
 52 weeks 70 (24) 59 (30) 11 (1 to 23)
Of pain during the day:
 3 weeks 22 (20) 10 (15) 12 (5 to 18)     < 0.001
 7 weeks 35 (20) 23 (24) 12 (4 to 21)
 13 weeks 36 (26) 27 (31)   9 (−3 to 20)
 26 weeks 32 (25) 32 (28)    0 (−10 to 10)
 52 weeks 38 (23) 35 (26)   3 (−7 to 13)
Of pain at night:
 3 weeks 21 (26) 9 (23) 12 (2 to 21)     0.015
 7 weeks 36 (28) 22 (30) 14 (3 to 25)
 13 weeks 37 (33) 28 (36)   9 (−4 to 23)
 26 weeks 34 (36) 33 (41)    1 (−13 to 17)
 52 weeks 37 (33) 35 (39)    2 (−12 to 16)
As rated by observer:
 3 weeks 13 (17) 0 (18) 13 (6 to 20)   <0.001
 7 weeks 24 (20) 9 (20) 15 (7 to 22)
 26 weeks 29 (24) 27 (27)   2 (−9 to 11)
Improvement in rating of shoulder disability§
 3 weeks 19 (27) 6 (22) 13 (4 to 23)     0.024
 7 weeks 39 (27) 14 (27)  25 (14 to 35)
 13 weeks 38 (31) 28 (32)  10 (−2 to 22)
 26 weeks 45 (30) 33 (34) 12 (0 to 25)
 52 weeks 42 (33) 38 (34)    4 (−10 to 17)
Improvement in degree of restriction of range of motion
External rotation:
 3 weeks  6 (14) –3 (12)  9 (3 to 14)   0.002
 7 weeks 13 (16) –2 (14) 15 (9 to 20)
 26 weeks 16 (18)  7 (21)  9 (1 to 16)
Abduction:
 3 weeks  2 (12) –3 (13) 5 (0 to 9)   0.065
 7 weeks  4 (11) –1 (14)  5 (0 to 10)
 26 weeks  9 (12)  7 (17)  2 (−3 to 8)
*

At 3 and 13 weeks there is one missing value in each group. At 7 weeks there is one missing value in the injection group. At 26 weeks there are one missing value in the injection group and two in the physiotherapy group. At 52 weeks there are four missing values in the injection group and one in the physiotherapy group. 

Pain as rated on visual analog scale in which scores range from 0-100; 100 indicates very severe pain. 

The change in scores for this outcome measure show a non-gaussian distribution. Non-parametric testing results in statistically significant differences at 3, 7, 13, and 52 weeks 

§

As rated on shoulder disability questionnaire in which scores range from 0-100; 100 indicates severe disability.11 12