Skip to main content

Table 5 T-test results for WHODAS and WEMWBS cluster-level scores

From: An intervention to support stroke survivors and their carers in the longer term: results of a cluster randomised controlled feasibility trial (LoTS2Care)

 

Time point

Cluster point estimates

Mean difference (CI: 95%, 67%, 51%)

T-test p-valuea

Significant at 5%, 33%, 49% level

New start

Usual care

n/Nb

Mean (SD)

n/N

Mean (SD)

WHODAS simple (score 0–100; higher score = higher level of disability)

Baseline

4/5

26.2 (4.34)

5/5

23.9 (5.81)

-2.3

0.53

(-10.60, 6.01)

No

(-5.97, 1.38)

No

(-4.85, 0.26)

No

6 months

4/5

21.12 (2.93)

5/5

24.2 (3.69)

3.14

0.21

(-2.23, 8.50)

No

(0.76, 5.51)

Yes

(1.48, 4.79)

Yes

9 months

4/5

24.2 (4.71)

5/5

23.34 (4.38)

-0.87

0.52

(-8.05, 6.30)

No

(-4.05, 2.30)

No

(-3.08, 1.34)

No

WHODAS complex (score 0–100; higher score = higher level of disability)

Baseline

4/5

28.0 (5.34)

5/5

24.7 (7.72)

−3.26

0.49

(-14.05, 7.53)

No

(-8.04, 1.51)

No

(-6.59, 0.06)

No

6 months

4/5

23.9 (4.56)

5/5

26.0 (6.89)

2.07

0.62

(-7.46, 11.59)

No

(-2.15, 6.29)

No

(-0.87, 5.00)

No

9 months

4/5

26.2 (6.22)

5/5

26.0 (5.99)

-0.16

0.97

(-9.82, 9.50)

No

(-4.44, 4.11)

No

 (-3.14, 2.82)

No

WEMWBS (score 14–70; higher score = better state of mental well-being)

Baseline

4/5

46.9 (2.06)

5/5

47.2 (1.55)

0.29

0.82

(-2.54, 3.12)

No

(-0.96, 1.54)

No

(-0.58, 1.16)

No

6 months

4/5

47.2 (1.80)

5/5

44.4 (5.70)

-2.87

0.37

(-9.96, 4.21)

No

(-6.01, 0.26)

No

(-5.06, − 0.69)

Yes

9 months

4/5

45.8 (2.71)

5/5

47.2 (2.80)

1.42

0.47

(-2.96, 5.80)

No

(-0.52, 3.36)

No

(0.07, 2.77)

Yes

  1. aUnpaired t-tests were performed at three different significance levels (5%, 33% and 49%) on cluster-level questionnaire data at each time point, producing corresponding confidence intervals (at 95%, 67% and 51%). These results are not adjusted for baseline scores. For data completeness, see Table I in Supplementary data
  2. bOnly four of the five sites in the intervention arm were analysed at cluster level. This is due to one site containing a very small number of stroke survivors, and thus not providing enough information about the cluster for accurate analysis