| Group |
N |
Missing |
| Effusion pre |
37 |
16 |
| Effusion 1w |
37 |
16 |
| Effusion 2w |
37 |
16 |
| Effusion 3w |
37 |
16 |
| Effusion 3m |
37 |
17 |
| Effusion 6m |
37 |
16 |
|
| Group |
Median |
25% |
75% |
| Effusion pre |
3.00 |
2.00 |
4.000 |
| Effusion 1w |
3.00 |
1.75 |
3.000 |
| Effusion 2w |
2.00 |
1.00 |
3.000 |
| Effusion 3w |
1.00 |
0.00 |
2.000 |
| Effusion 3m |
0.00 |
0.00 |
1.000 |
| Effusion 6m |
0.00 |
0.00 |
0.250 |
|
H=57.5 with 5 degrees of freedom (P=<0.0001).
The differences in the median values among the treatment groups are greater
than would be expected by chance; there is a statistically significant difference
(P=3.91E-011).
All pairwise multiple comparison procedures (Dunn's method): |
| Comparison |
Diff of Ranks |
P |
Q |
P<0.05 |
| Effusion pre vs. 6m |
63.62 |
6 |
5.857 |
Yes |
| Effusion pre vs. 3m |
56.25 |
5 |
5.116 |
Yes |
| Effusion pre vs. 3w |
44.00 |
4 |
4.051 |
Yes |
| Effusion pre vs. 2w |
24.79 |
3 |
2.282 |
No |
| Effusion pre vs. 1w |
7.19 |
2 |
0.662 |
Do not test |
| Effusion 1w vs. 6m |
56.43 |
5 |
5.195 |
Yes |
| Effusion 1w vs. 3m |
49.06 |
4 |
4.462 |
Yes |
| Effusion 1w vs. 3w |
36.81 |
3 |
3.389 |
Yes |
| Effusion 1w vs. 2w |
17.60 |
2 |
1.620 |
Do not test |
| Effusion 2w vs. 6m |
38.83 |
4 |
3.575 |
Yes |
| Effusion 2w vs. 3m |
31.47 |
3 |
2.862 |
No |
| Effusion 2w vs. 3w |
19.21 |
2 |
1.769 |
Do not test |
| Effusion 3w vs. 6m |
19.62 |
3 |
1.806 |
No |
| Effusion 3w vs. 3m |
12.25 |
2 |
1.114 |
Do not test |
| Effusion 3m vs. 6m |
7.37 |
2 |
0.670 |
Do not test |
|
| Table 3: Statistical analysis of effussion at each time point. Effusion: Kruskal-
Wallis one way analysis of variance on ranks. |
