Introduction: Increased prevalence of diabetes is found in major psychosis, and poorer outcomes occur when both diagnoses co-exist. The aim of this study was to describe the association between total and cardiovascular mortality in people with both schizophrenia and diabetes, compared to diabetes alone.
Materials and methods: Data was linked from the Glasgow Psychosis Clinical Information System (PsyCIS), the Scottish national diabetes database (SCI-Diabetes), and national death certification data. The Scottish Index of Multiple Deprivation (SIMD) 2016 quintile for each patient was obtained using their residential postcode. Schizophrenia was defined using International Classification of Diseases, Tenth Revision (ICD-10) codes F20 (schizophrenia) and F25 (schizoaffective disorder). Cases (schizophrenia and diabetes) were matched by age (+/− 1 year), gender and deprivation quintile to those with diabetes but not present on the PsyCIS database (controls). Follow up 01/01/2002 to 16/8/2018. Standard descriptive statistics and survival analysis undertaken, adjusted for age.
Results: 702 individuals in each cohort. 64% male. Mean age at index 41 years. 414/702 (59%) were in the most deprived quintile. Mean body mass index (BMI) at diagnosis, or in preceding 12 months, available for 450/702 controls, and 446/702 cases. Mean (S.D.) BMI 33.53 (6.75) for controls and 33.23 (6.58) for cases, no statistical difference. At study end 63/702 controls were dead, 133/702 cases. Hazard ratio 2.34 (95% CI 1.76 to 3.21, P value<0.001). Cardiovascular diagnosis was noted in death certification in 34/702 controls and 51/702 cases. Hazard ratio 1.716 (95% CI 1.110 to 2.653, P value<0.05).
Conclusion: This study demonstrates significant excess all cause mortality and cardiovascular death over and above that expected with diabetes, despite our cohorts being matched for deprivation. Lower thresholds for starting cardiovascular prevention medications may be indicated in individuals with diabetes and schizophrenia. Using routinely collected data has limitations, and missing data can be substantial.