Self-monitoring of blood glucose among adults with serious mental illness and type 2 diabetes: Sociodemographic, clinical and psychiatric characteristics

People with serious mental illness (SMI) are twice as likely as the general population to have type 2 diabetes. Management of diabetes requires a complex self-care regimen that typically includes self-monitoring of blood glucose (SMBG). Rates of SMBG are suboptimal among both the general population and adults with SMI, but little work has been done to identify predictors of SMBG among adults with SMI. The purpose of this retrospective study was to explore predictors of SMBG among adults with SMI and type 2 diabetes. Seventy-nine subjects (mean age 52.8, SD = 8.96) were selected from the baseline data of a diabetes self-management program for adults with SMI and metabolic disorders. A majority of participants were Caucasian (n = 52, 66%), male (n = 40, 51%), diagnosed with schizophrenia (n = 51, 65%), and were not prescribed insulin (n = 55, 70%). We analyzed sociodemographic, diabetes regimen, and psychiatric characteristics, including psychotic symptoms, depression, and cognitive functioning. Self-monitoring behavior was dichotomized into any-SMBG or no-SMBG groups. Results of bivariate analyses show that those in the non-SMBG group had fewer years of education, more severe negative psychiatric symptoms, and lower memory functioning. They also had higher rates of schizophrenia and lower rates of insulin prescription. A stepwise multiple logistic regression of variables that were significant at the bivariate level yielded a three-predictor model: insulin prescription (OR = 19.06, p = .009), schizophrenia diagnosis (OR = 5.38, p = .044) and memory functioning (OR = .09, p = .015). We also searched for relationships between psychiatric symptoms and long-term blood glucose control, and for differences in long-term blood glucose control between SMBG groups, but did not find significant results in those areas.