Cardiovascular risk management among individuals with type 2 diabetes and severe mental illness: a cohort study

Aims/hypothesis The aim of this study was to compare cardiovascular risk management among people with type 2 diabetes according to severe mental illness (SMI) status. Methods We used linked electronic data to perform a retrospective cohort study of adults diagnosed with type 2 diabetes in Scotland between 2004 and 2020, ascertaining their history of SMI from hospital admission records. We compared total cholesterol, systolic BP and HbA1c target level achievement 1 year after diabetes diagnosis, and receipt of a statin prescription at diagnosis and 1 year thereafter, by SMI status using logistic regression, adjusting for sociodemographic factors and clinical history. Results We included 291,644 individuals with type 2 diabetes, of whom 1.0% had schizophrenia, 0.5% had bipolar disorder and 3.3% had major depression. People with SMI were less likely to achieve cholesterol targets, although this difference did not reach statistical significance for all disorders. However, people with SMI were more likely to achieve systolic BP targets compared to those without SMI, with effect estimates being largest for schizophrenia (men: adjusted OR 1.72; 95% CI 1.49, 1.98; women: OR 1.64; 95% CI 1.38, 1.96). HbA1c target achievement differed by SMI disorder and sex. Among people without previous CVD, statin prescribing was similar or better in those with vs those without SMI at diabetes diagnosis and 1 year later. In people with prior CVD, SMI was associated with lower odds of statin prescribing at diabetes diagnosis (schizophrenia: OR 0.54; 95% CI 0.43, 0.68, bipolar disorder: OR 0.75; 95% CI 0.56, 1.01, major depression: OR 0.92; 95% CI 0.83, 1.01), with this difference generally persisting 1 year later. Conclusions/interpretation We found disparities in cholesterol target achievement and statin prescribing by SMI status. This reinforces the importance of clinical review of statin prescribing for secondary prevention of CVD, particularly among people with SMI. Graphical Abstract Supplementary Information The online version of this article (10.1007/s00125-024-06111-w) contains peer-reviewed but unedited supplementary material.

ESM Methods: The multiple imputation process and rationale We used a parallel version of the multiple imputation by chained equations (MICE) package 3.14.0 to perform multiple imputation.Data were missing for area-based deprivation, systolic blood pressure at baseline, total cholesterol levels at baseline, HbA1c levels at baseline, BMI, and smoking status.It was assumed that data were missing at random (MAR).
As there seemed to be an interaction between age and sex and between age and BMI, these interaction terms were included during multiple imputation as well [4,5].To deal with the interaction between age and sex, the dataset was split based on sex, whereafter multiple imputation was run on both subsets and the products were bound together.For the interaction between age at type 2 diabetes diagnosis and BMI, the interaction term was included as an additional variable by subtracting the mean and taking the product of age and BMI [5].When BMI was missing, and thus the interaction term was missing, the interaction would be calculated from the imputed value of BMI, instead of predicted based on the other available data, to prevent convergence as a result of a feedback loop between BMI and the interaction term.
Both the continuous and dichotomised outcomes were included in the multiple imputation process [6].
When the outcome was missing, the predicted continuous outcome was not used in the prediction process of the dichotomised outcome, and vice versa, to prevent convergence.Available data from individuals that miss outcomes were used during the multiple imputation process, whereafter four sets of the imputed datasets were formed, one for each primary outcome and one dataset with all individuals for the secondary outcome, statin prescription, since there was no missing outcome data for this outcome [4,7].
As a rule of thumb, the number of generated datasets should be at least equal to the percentage of incomplete cases, as 38.3% of participants had at least one of the above-mentioned covariates missing, 40 datasets were imputed [5,7].We analysed each imputed dataset separately and pooled results using Rubin's rule [8].

Table 4 :
: Odds ratios (95% CI) for total cholesterol, systolic blood pressure, and HbA1c target level achievement, comparing people with each severe mental illness versus no mental illness, stratified by sex.Odds ratios (95% CI) from the sensitivity analysis including the completecase cohort for total cholesterol, systolic blood pressure, and HbA1c target level achievement comparing people with each severe mental illness versus no mental illness, stratified by sex.
Model 1 is adjusted for age at diagnosis, area-based deprivation, NHS health board, calendar year of diagnosis, history of CVD, and history of other morbidities.Model 2 is additionally adjusted for history of an alcohol use disorder, smoking status, body mass index, and total cholesterol, systolic blood pressure, and HbA1c at time of diabetes diagnosis.CI = confidence interval; OR = odds ratio ESM

Table 5 :
Odds ratios (95% CI) from the sensitivity analysis including SIGN defined target level achievement.

Table 6 :
Odds ratios (95% CI) for receipt of a statin prescription at the time of diabetes diagnosis and one year thereafter, comparing people with each severe mental illness versus no mental illness, stratified by history of cardiovascular disease.Model 1 is adjusted for age at diagnosis, area-based deprivation, NHS health board, calendar year of diagnosis, history of CVD, and history of other morbidities.Model 2 is additionally adjusted for history of an alcohol use disorder, smoking status, body mass index, and total cholesterol, systolic blood pressure, and HbA1c at the time of diabetes diagnosis.The models estimating the association between SMI status and statin prescribing one year after diabetes diagnosis also included statin prescribing at the time of diabetes diagnosis as a covariate.