Reminiscence therapy is one of the most effective psychosocial interventions for people with dementia. To analyse the effectiveness of group reminiscence therapy in people with dementia and to check their involvement in the design and evaluation of the interventions. CINHAL, PubMed, Psicodoc and PsycINFO databases were searched for the period of 2004 to 2016. A systematic search was performed, as well as a quality assessment and a meta-analysis of the articles. The 14 studies selected had a moderate risk of bias. People with dementia who received group reminiscence therapy experienced an improvement in cognition (dMR = 0.26; 95% CI: 0.11-0.40; p = 0.0006; I2 = 0%) which was larger than the one observed in the control group (dc = 0.37; 95% CI: 0.11-0.63; p = 0.005; I2 = 0%). There was no improvement in mood, quality of life, behaviour or activities of daily living. People with dementia were not included in the design nor in the evaluation of the interventions. Group reminiscence therapy for people with dementia has favourable effects in cognition. There is a need for the inclusion of people with dementia in the design and planning of the interventions and for more randomised controlled trials.