Delirium is a common complication among adults. It is essential to improve the outcomes of delirium. To systematically synthesize the evidence on the effectiveness of the nurse-led non-pharmacological interventions on outcomes of delirium in adults. Electronic databases including CINAHL, Cochrane Library, MEDLINE, EMBASE, PubMed, Web of Science, PsycINFO, and Clinical Trial Registration were searched comprehensively by the authors. The authors reviewed the full text and assessed the risk of bias using the Cochrane Risk of Bias Tool 2.0. The meta-analysis was performed using RevMan and Stata software. The forest plots showed the overall effect of the included study and the I2 test was used to assess the degree of heterogeneity between studies. Random effects models were used to analyze studies with significant heterogeneity. A total of 32 studies (10,122 participants) were included in the meta-analysis. Nurse-led non-pharmacological interventions resulted in a significantly lower incidence of delirium compared with the usual care/control group (risk ratio = 0.74, p < .001) and reduced mortality in the hospital compared with usual care (risk ratio = 0.81, p = .04). However, the implementation of nurse-led, non-pharmacological interventions had no significant effect on the duration, severity of delirium, or length of hospital stay. Our findings suggest that the nurse-led, non-pharmacological strategy was effective in reducing the incidence of delirium and mortality in the hospital. Multicomponent interventions were the most effective strategy for reducing the incidence of delirium in adults.