Table 2 Details of the interventions used in the included studies

Rivers 2001 [4]

83 (63.8%)

27 (20.3%)

36 (27.7%)

40 (30.1%)

18 (13.8%)

1 (0.8%)

69 (53.1%)

72 (54.1%)

21.4 ± 6.9

20.4 ± 7.4

Lin 2006 [31]

39 (36.1%)

43 (37.1%)

80 (74.1%)

81 (69.8%)

13 (12.0%)

16 (13.8%)

–

–

–

–

Wang 2006 [32]

–

–

–

–

–

–

4 (25.0%)

5 (29.4%)

28.0 ± 7.0

27.0 ± 6.0

Chen 2007 [29]

–

–

–

–

–

–

–

–

16.53 ± 6.87

16.68 ± 7.14

He 2007 [30]

–

–

–

–

–

–

–

–

–

–

Yan 2010 [27]

–

–

–

–

–

–

–

–

23.5 ± 5.7

21.8 ± 6.5

Andrews 2014 [28]

16 (30.2%)

11 (19.7%)

–

–

1 (1.9%)

3 (5.4%)

–

–

–

–

ARISE 2014 [7]

108 (13.6%)

56 (7.1%)

528 (66.7%)

461 (57.9%)

122 (15.4%)

21 (2.7%)

276 (34.8%)

263 (33.0%)

15.4 ± 6.5

15.8 ± 6.5

ProCESS 2014 [6]

63 (14.4%)

37 (8.1%)

241 (54.9%)

201 (44.1%)

35 (8.0%)

5 (1.1%)

165 (37.6%)

146 (32.0%)

20.8 ± 8.1

20.7 ± 7.5

ProMISe 2015 [8]

55 (8.8%)

24 (3.9%)

332 (53.3%)

291 (46.9%)

113 (18.1%)

24 (3.9%)

179 (28.7%)

175 (28.2%)

18.7 ± 7.1

18 ± 7.1

Jansen 2010 [14]

–

–

113 (63.8%)

119 (69.6%)

58 (32.8%)

69 (40.4%)

–

–

22.7 ± 9.1

23.6 ± 8.6

Jones 2010 [15]

5 (3.3%)

11 (7.3%)

113 (75.3%)

108 (73.5%)

8 (5.3%)

5 (3.3%)

39 (26.0%)

40 (26.7%)

–

–

Tian 2012 [17]

–

–

–

–

–

–

–

–

20.9 ± 8.6

17.8 ± 6.3

Yu 2013 [13]

5 (20.0%)

4 (16.0%)

20 (80.0%)

18 (72.0%)

1 (4%)

2 (8%)

19 (76%)

20 (80%)

17.91 ± 3.77

18.18 ± 6.01

Wang 2014 [18]

–

–

–

–

–

–

–

–

19.7 ± 3.1

20.9 ± 7.6

Lv 2015 [16]

–

–

–

–

–

–

–

–

–

–