Background: Many psychosocial interventions, such as cognitive training, physical exercise, or music therapy, are showed to be effective to reduce cognitive decline in people with dementia. Poster Presentations: Sunday, July 14, 2019 P468 The objective of this study was to compare the effectiveness of different types of psychosocial interventions. Methods: A network meta-analysis of RCTs was performed to compare the treatment effect of different psychosocial interventions for people with mild-to-moderate dementia. Literature searches were performed in MEDLINE, EMBASE, CINAHL and PsycINFO. RCTs with the comparison of the psychosocial interventions, such as cognitive training, physical exercise, musical interventions, interventions with general social activity, or usual care were included. The pri- mary outcome was the change of the score in the Mini-mental state examination (MMSE) cognitive screening. Pairwise comparisons in the original trials were obtained for direct comparison. When two interventions were compared through a common reference across the trials, indirect comparison was used to estimate the rela- tive effectiveness. A network meta-analysis of direct and indirect comparisons was conducted across the trials. The Bayesian random effects model with Markov chain Monte Carlo method was used to account for the heterogeneities across trials. Mean difference (MD) with 95% credible interval (CrI) were calculated and the ranks of effectiveness were assigned for each intervention. Results: A total of 43 RCTs with 3,112 participants with mild-to-moderate demen- tia formed a network comparison for MMSE. People received cognitive training showed better cognitive functions than social ac- tivityandusualcaregroup(MD¼1.68,95%CrI¼0.85–2.51,and MD ¼ 1.23, 95% CrI ¼ 0.71 – 1.76, respectively). People received music intervention also showed better cognitive functions than social activity and usual care group (MD ¼ 2.18, 95% CrI ¼ 0.48 –3.88,andMD¼1.73,95%CrI¼0.25–3.21,respectively).How- ever, exercise did not show improvement in cognitive functions. Besides, there was no significant difference among cognitive training, exercise, music and combined interventions. Conclusions: Cognitive training and music interventions can enhance cognitive functions than the usual care or general social activity. With a limited sample size, physical exercise did not show significant benefit from social activity or usual care. P1-521 FROM BARRIERS TO BEHAVIOR CHANGE: EXERCISE PROVIDERS MAY BE KEY TO ENABLING EXERCISE AMONG PEOPLE LIVING WITH MCI/DEMENTIA Laura E. Middleton 1 , Shannon Freeman 2 , Lauren Bechard 1 , Jacqueline A. Pettersen 3 , Jennifer R. Tomasone 4 , Sherry L. Dupuis 1 , Rebecca Ferris 2 , Kayla Regan 1 , 1 University of Waterloo, Waterloo, ON, Canada; 2 University of Northern British Columbia, Prince George, BC, Canada; 3 University of British Columbia–Northern Medical Program, Prince George, BC, Canada; 4 Queen’s University, Kingston, ON, Canada. Contact e-mail: laura.middleton@uwaterloo.ca Background: Exercise improves physical and mental health among people with mild cognitive impairment (MCI) or dementia, yet many remain inactive. To implement exercise in clinical manage- ment, we must address barriers and facilitators in implementation. Our objectives were to: 1) understand barriers and facilitators for exercise among people with MCI/dementia; 2) use this knowledge toidentify behavior changes strategiesusing the Behaviour Change Wheel (BCW). Methods: Focus groups and interviews were con- ducted with community-dwelling people with MCI/dementia (n¼14), care partners (n¼12), exercise providers (n¼35), and health care professionals (n¼15) in Ontario and Northern BC (Canada). Audio recordings were transcribed and thematically analyzed. Themes were mapped to capability, opportunity, and motivation domains of the BCW and mapped to interventions. Results: Seventy-six people participated in interviews and focus groups. Barriers and facilitators were identified by BCW domains: 1) Capability: cognitive deficits, physical health, fatigue; 2) Oppor- tunity: access to programs meeting needs of people with MCI/de- mentia, poor understanding of dementia by exercise providers, stigmaofdementia;3)Motivation:apathy/lowintrinsic motivation, encouraging exercise providers, care partner support, social engagement.The primary intervention selected asmostappropriate was enablement of exercise through education and training of exer- cise providers, which would expand group exercise opportunities with the added benefits of modelling and incentivization by social engagement. Conclusions: Training exercise providers to under- stand, engagewith, and meet the needs of peoplewith MCI/demen- tia could be key to implementation of exercise in clinical management,byincreasingthequantityandqualityofexercisepro- grams accessible to people with dementia.