Background: People with serious mental illnesses (SMIs) such as schizophrenia and bipolar disorder die up to 30 years younger than individuals in the general population. Premature mortality among this population is often due to medical comorbidities, such as type 2 diabetes (T2D). Being a disease directly related to diet, adverse lifestyle choices, and side effects of psychotropic medication, an effective approach to T2D treatment and management could be non-pharmacological interventions. This systematic review and meta-analysis (1) summarise the current evidence base for non-pharmacological interventions (NPI) for diabetes management in people living with SMI and (2) evaluate the effect of these interventions on diverse health outcomes for people with SMI and comorbid diabetes. Methods: Six databases were searched to identify relevant studies: PubMed (MEDLINE), PsycINFO, Embase, Scopus, CINAHL, and Web of Science. Studies were included if they reported on non-pharmacological interventions targeted at the management of T2D in people living with SMI. To be eligible, studies had to further involve a control group or report multiple time points of data in the same study population. Whenever there were enough interventions reporting data on the same outcome, we also performed a meta-analysis. Results: Of 1867 records identified, 14 studies were included in the systematic review and 6 were also eligible for meta-analysis. The results showed that there was a reduction, although not significant, in glycated haemoglobin (HbA1c) in the NPI group compared with the control, with a mean difference of −0.14 (95% CI, −0.42, 0.14, p = 0.33). Furthermore, NPI did not significantly reduce fasting blood glucose in these participants, with a mean difference of −17.70 (95% CI, −53.77, 18.37, p = 0.34). However, the meta-analysis showed a significant reduction in psychiatric symptoms: BPRS score, −3.66 (95% CI, −6.8, −0.47, p = 0.02) and MADRS score, −2.63 (95% CI, −5.24, −0.02, p = 0.05). NPI also showed a significant reduction in the level of total cholesterol compared with the control, with a mean difference of −26.10 (95% CI, −46.54, −5.66, p = 0.01), and in low-density lipoprotein (LDL) cholesterol compared with control, with a standardised mean difference of −0.47 (95% CI, −0.90, −0.04, p = 0.03). NPI did not appear to have significant effect (p > 0.05) on body mass index (BMI), health-related quality of life (HRQL), triglycerides, and high-density lipoprotein cholesterol compared with control. Conclusions: This systematic review and meta-analysis demonstrated that NPI significantly (p < 0.05) reduced psychiatric symptoms, levels of total cholesterol, and LDL cholesterol in people with type 2 diabetes and SMI. While non-pharmacological interventions also reduced HbA1c, triglyceride, and BMI levels and improved quality of life in these people, the effects were not significant (p > 0.05).