Skip to main content
. 2021 Dec 13;23(12):e26794. doi: 10.2196/26794

Table 2.

Composite scores at the 3- and 6-month follow-up.

Time and composite score Control (n=112) Intention-to-treat (n=110)a Per-protocol (n=76)a


Value ORb (95% CI) P value Value OR (95% CI) P value
3 months

Composite score 1c, n (%)


Improved 10 (8.9) 15 (14) 1.663 (0.712-3.884) .24 14 (19.2) 2.420 (1.011-5.792) .047d


Deteriorated or unchanged 102 (91.1) 92 (86) 1.663 (0.712-3.884) .24 59 (80.8) 2.420 (1.011-5.792) .047d

Composite score 2e, n (%) f



Improved 10 (8.9) 15 (14)

14 (19.2)
.04d


Unchanged 89 (79.5) 83 (77.6)

54 (74)
.04d


Deteriorated 13 (11.6) 9 (8.4)

5 (6.8)
.04d
6 months

Composite score 1c, n (%)


Improved 13 (11.6) 16 (15) 1.339 (0.611-2.936) .47 13 (17.8) 1.650 (0.717-3.795) .24


Deteriorated or unchanged 99 (88.4) 91 (85) 1.339 (0.611-2.936) .47 60 (82.2) 1.650 (0.717-3.795) .24

Composite score 2e, n (%)



Improved 13 (11.6) 16 (15)

13 (17.8)
.50


Unchanged 83 (74.1) 74 (69.2)

49 (67.1)
.50


Deteriorated 16 (14.3) 17 (15.9)

11 (15.1)
.50

a3 missing values (no general self-efficacy score at baseline).

bOR: odds ratio.

cComposite score dichotomized into improved vs deteriorated or unchanged.

dSignificant at P<.05.

eComposite score improved vs unchanged vs deteriorated.

fThe statistical test used (Mantel-Haenszel chi-square test) does not generate odds ratio or 95% CI.