|
Outcome
|
Intervention group
|
Comparison group
|
Effect (difference between the change in the two groups)
|
p-value a
|
|---|
|
Number of observation periods, n
|
Total count
|
Median (IQR)
|
Change in median (from baseline)
|
Number of observation periods, n
|
Total count
|
Median (IQR)
|
Change in median (from baseline)
|
|---|
|
Walking (secondary)
|
Baseline
|
12
|
48
|
4 (1.75)
|
–
|
24
|
64
|
2 (1.75)
|
–
|
–
|
–
|
|
7 months
|
12
|
135
|
12.5 (8.5)
|
8.5
|
24
|
116
|
5 (3)
|
3
|
5.5
|
.005*
|
|
12 months
|
12
|
140
|
11.5 (10)
|
7.5
|
24
|
88
|
3 (3)
|
1
|
6.5
|
< 0.001*
|
|
24 months
|
10 b
|
141
|
12 (9.5)
|
8
|
24
|
98
|
4 (3.75)
|
2
|
6
|
< 0.001*
|
|
Vigorous (secondary)
|
Baseline
|
12
|
3
|
0 (0.75)
|
–
|
24
|
1
|
0 (0)
|
–
|
–
|
–
|
|
7 months
|
12
|
13
|
0.5 (2)
|
0.5
|
24
|
1
|
0 (0)
|
0
|
0.5
|
.009*
|
|
12 months
|
12
|
7
|
0 (0.75)
|
0
|
24
|
1
|
0 (0)
|
0
|
0
|
.96
|
|
24 months
|
10 b
|
13
|
1 (2.25)
|
1
|
24
|
0
|
0 (0)
|
0
|
1
|
.002*
|
|
Sedentary (exploratory)
|
Baseline
|
12
|
2
|
0 (0)
|
–
|
24
|
4
|
0 (0)
|
–
|
–
|
–
|
|
7 months
|
12
|
24
|
1.5 (4)
|
1.5
|
24
|
17
|
0 (1)
|
0
|
1.5
|
.07
|
|
12 months
|
12
|
31
|
2 (1)
|
2
|
24
|
9
|
0 (0.75)
|
0
|
2
|
< 0.001*
|
|
24 months
|
10 b
|
24
|
2.5 (2.25)
|
2.5
|
24
|
8
|
0 (0)
|
0
|
2.5
|
< 0.001*
|
|
Connect (exploratory)
|
Baseline
|
12
|
17
|
1.5 (2)
|
–
|
24
|
24
|
0.5 (2)
|
–
|
–
|
–
|
|
7 months
|
12
|
49
|
4 (5.25)
|
2.5
|
24
|
41
|
2 (2)
|
1.5
|
1
|
.42
|
|
12 months
|
12
|
40
|
3 (5.75)
|
1.5
|
24
|
21
|
0 (2)
|
−0.5
|
2
|
.03*
|
|
24 months
|
10 b
|
49
|
3.5 (5.25)
|
2
|
24
|
25
|
0 (2)
|
−0.5
|
2.5
|
.006*
|
|
Take Notice (exploratory)
|
Baseline
|
12
|
3
|
0 (0.75)
|
–
|
24
|
5
|
0 (0)
|
–
|
–
|
–
|
|
7 months
|
12
|
11
|
1 (1.75)
|
1
|
24
|
6
|
0 (0.75)
|
0
|
1
|
.07
|
|
12 months
|
12
|
8
|
0 (1)
|
0
|
24
|
7
|
0 (0)
|
0
|
0
|
.58
|
|
24 months
|
10 b
|
15
|
1.5 (1)
|
1.5
|
24
|
3
|
0 (0)
|
0
|
1.5
|
.001*
|
- IQR interquartile range
- a Mann-Whitney U tests were carried out to determine if there were significant differences in the change in counts of behaviours from baseline to each follow-up between intervention and comparison groups; b Two observation periods were cancelled at the intervention site due to ethical concerns regarding observer safety
- * Statistically significant at p < 0.05 (z-test, two-tailed)
