Background <p>Neuropsychiatric disorders are among the most common diseases worldwide and are characterized by complex pathogenic mechanisms. Cathepsins (CTS) are crucially involved in the pathogenesis and treatment of numerous diseases. Increasing evidence suggests a relationship between cathepsins and neuropsychiatric disorders. However, the causal associations remain unclear.</p> Methods <p>We performed a bidirectional two-sample Mendelian randomization (MR) analysis, applying univariable (UVMR) and multivariable MR (MVMR) to evaluate the causal association between nine cathepsins and five neuropsychiatric disorders. Data for this study were derived from genome-wide association studies (GWAS). The MR analysis primarily used five methods: Inverse Variance Weighted (IVW), Weighted Median Estimator (WME), MR-Egger regression, Simple mode, and Weighted mode. Additionally, sensitivity tests were employed to assess the robustness of the MR results.</p> Results <p>MR analyses indicated that CTSH is associated with an increased risk of Alzheimer’s disease (AD) (UVMR: OR, 1.041; 95% CI, 1.013–1.069; <i>p</i> = 0.004; MVMR: OR, 1.040; 95% CI, 1.014–1.066; <i>p</i> = 0.003; replication UVMR: OR, 1.046; 95% CI, 1.014–1.082; <i>p</i> = 0.011). Findings that were significant in only one MR approach, such as the putative causal effects of CTSF on AD and bipolar disorder (BIP), of CTSL2 on major depressive disorder (MDD), and of CTSH and CTSE on Parkinson’s disease (PD) should be interpreted with caution.</p> Conclusion <p>CTSH can be considered a plasma biomarker for AD, offering new insights and potential directions for the prevention and treatment of AD. Additionally, during the treatment of BIP and PD, attention should be paid to CTSF and CTSE expression levels to maintain physiological homeostasis.</p>