Background: The number of people with neurocognitive disorder is increasing, and the majority of them are cared for by informal caregivers in the community. Mental health problems are common among caregivers, however, professional support for them is often limited. Non-pharmacological self-help interventions, such as bibliotherapy, may improve mental well-being and has the potential for being integrated into clinical or social services. Objectives: To explore what types of bibliotherapy have been used for improving the mental well-being of informal caregivers of people with neurocognitive disorders, and the effect on mental well-being outcomes. Design: A systematic review and meta-analysis. Review methods: Six databases were searched for relevant articles on July 1, 2019. Clinical trial registries and the reference lists of included studies were also searched. Both randomized controlled trials and quasi-experimental studies were included. The Cochrane Collaboration risk of bias tool for randomized controlled trials was used to assess the quality of studies. Review Manager 5.3 was used to analyze data, standardized mean difference (SMD) and 95% confidence interval (CI) were used to estimate the pooled treatment effect. Random effects models were used for meta-analyses. Funnel plot was not performed due to the limited number of studies. This systematic review was registered at PROSPERO (CRD42019129152). Results: Nine randomized controlled trials with 1036 informal caregivers were included. Most of the included studies had some aspects of bias. Three types of bibliotherapy were used. Bibliotherapy had a significant pooled medium to large effect on reducing depression at Z = 1.99 (SMD = -0.74, 95%CI = -1.47 to -0.01, p = .05), however, the heterogeneity was high (I2 = 94%). For the subgroups, only the video-based bibliotherapy significantly reduced depression at Z = 2.78 (I2 = 83%, SMD = -2.11, 95%CI = -3.6 to -0.62, p = .005). Bibliotherapy had a significant small to medium effect on caregiver's self-efficacy for dealing with problem behaviours at Z = 2.44 (I2 = 0, SMD = 0.36, 95%CI = 0.05 to 0.67, p = .02), however, the effect on self-efficacy for obtaining respite was not significant (I2 = 0, SMD = 0.17, 95%CI = -0.16 to 0.49, p = .32). The effect on decreasing state anxiety was significant at Z = 2.30 (I2 = 22%, SMD = -0.22, 95% CI = -0.41 to -0.33, p = .02). Conclusions: Bibliotherapy showed positive effects on reducing depression, improving self-efficacy for dealing with problem behaviors and reducing anxiety among informal caregivers. The effects on reducing depression should be viewed with caution due to high heterogeneity. The effects on other mental well-being outcomes are inconclusive due to limited number of studies and this underscores the need for further research.