“…In total 28 original studies were identified that tested an intervention comprising solely of peer support or had a peer support component. We examined the extent, nature and distribution of the studies included in this review and determined the following: the majority of studies ( n = 12) used a trial design including randomised controlled trials, pragmatic randomised trials, randomised controlled evaluations, randomised controlled two group design or randomised pilot studies (Charlesworth et al, ; Finkel et al, ; Gaugler et al, , , , ; Gaugler, Reese, Reese, & Mittelman, ; Hayden, Glynn, Hahn, Randall, & Randolph, ; Laakkonen et al, , ; Lai et al, ; Martindale‐Adams, Nichols, Burns, Graney, & Zuber, ; Mittelman et al, ; Núñez‐Naveira et al, ; Torkamani et al, ; Wai Tong & Lee, ; Wang & Chien, ; Wang et al, ; Winter & Gitlin, ), eight studies used a pre/post‐test design (Boots, de Vugt, Withagen, Kempen, & Verhey, ; Chenoweth et al, ; Chu et al, ; Dam, van Boxtel, Rozendaal, Verhey, & de Vugt, ; Gaugler, Hobday, et al, ; Hsu et al, ; Javadpour, Ahmadzadeh, & Bahredar, ; O'Connor, Arizmendi, & Kaszniak, ), four a quasi‐experimental design (Andren & Elmståhl, , ; Easom, Alston, & Coleman, ; Marziali & Garcia, ; Pagán‐Ortiz, Cortés, Rudloff, Weitzman, & Levkoff, ), two a quasi‐experiment with mixed methods design (Baily, Kingsyon, Alford, Taylor, & Tolhurst, ; Küçükgüçlü, Akpınar Söylemez, Yener, & Işık, ) and the final two studies used mixed methods (Hattink, Droes, Sikkes, Oostra, & Lemstra, ; McKechnie, Barker, & Stott, ).…”