Table 3.

Assessment of risk of bias for included RCTs

Author	Randomization process	Deviations from intended interventions	Missing outcome data	Measurement of outcome	Selection of the reported result
Ballard et al. [62]	o	o	o	o	o
Chenoweth et al. [64]	v	o	o	o	o
Chenoweth et al. [63]	o	o	o	o	o
Cohen-Mansfield et al. [65]	o	o	v	v	o
Eritz et al. [66]	v	o	o	v	o
Fossey et al. [67]	o	o	o	o	o
Lawton et al. [68]	o	x	v	v	o
Rokstad et al. [69]	v	o	o	o	o
Sloane et al. [70]	o	o	o	o	o
Testad et al. [71]	v	o	x	o	o
van Bogaert et al. [72]	o	v	v	o	o
van de Ven et al. [73]	o	o	v	v	o
van der Ploeg et al. [74]	o	o	v	o	o
van Haitsma et al. [75]	o	v	o	v	o
van Weert et al. [76]*	o	x	v	o	o

Note: Low risk of bias (o), moderate risk of bias (v), high risk of bias (x). Abbreviations: RCTs, randomized controlled trials.

^*van Weert et al. (2005) applied a quasi-experimental pre- and post-test design including randomization, hence this study was analyzed with Rob2 for risk of bias of included RCTs.