“…Four studies were clearly protocol driven and provided enough information to conclude a low risk of reporting bias (Chenoweth et al, 2014;CristanchoLacroix et al, 2015;Livingston et al, 2013;Orgeta et al, 2015). Ten studies reported small sample sizes (Alnes, Kirkevold, & Skovdahl, 2011;Beer, Hutchinson, & Skala-Cordes, 2012;Cristancho-Lacroix et al, 2015;Cruz, Marques, Barbosa, Figueiredo, & Sousa, 2011;Gentry, 2011;Haberstroh et al, 2011;Liddle et al, 2012;Prick et al, 2015;Raglio et al, 2016;van der Ploeg et al, 2013). Eight studies used unstandardised measures (Alnes et al, 2011;Bray et al, 2015;Broughton et al, 2011;Galvin et al, 2010;Judge, Yarry, Orsulic-Jeras, & Piercy, 2010;Robinson, Bamford, Briel, Spencer, & Whitty, 2010;Velzke, 2014;Weitzel et al, 2011); therefore it is not clear if these are reliable and valid and it is difficult to compare outcomes between studies.…”