EPV1524
Type 2 diabetes among people with psychosis in Finland – prevalence, treatment balance, and mortality
J. Keinänen 1 2,*, M. Holm 1, K. Suokas 3, J. Haapakoski 1, A. Tahkola 1, T. Kauppala 1, E. Liukko 1, L. Schildt 1, S. Vanhamäki 1, S. Metso 1, J. Suvisaari 1
1National Institute for Health and Welfare, 2Psychiatry, Helsinki University Hospital, 3University of Helsinki, Helsinki, Finland
Introduction: Individuals with psychosis experience markedly reduced life expectancy and a high burden of cardiometabolic comorbidities. Multiple risk factors for type 2 diabetes (T2D) accumulate in people with psychosis. Despite this high burden, real-world data on T2D management and outcomes among people with psychosis remain limited, underscoring the need for integrated approaches that address both psychiatric and metabolic health.
Objectives: Aim of the study was to compare baseline cardiometabolic profiles, treatment balance and mortality between those with psychosis and T2D and those with T2D alone.
Methods: We identified individuals younger than 60 years with type 2 diabetes (T2D) from the Finnish diabetes register who were also recorded in the psychosis care quality register. The diabetes register includes all Finnish residents who were alive and had a documented diabetes diagnosis in or after 2018. We compared T2D treatment balance and all-cause mortality between those who developed T2D after a prior diagnosis of a psychotic disorder and those with T2D only, with follow-up through the end of 2023.
Results: Table 1 shows the baseline characteristics of the study population. Overall, the psychosis plus diabetes group demonstrated a more favorable cardiometabolic profile at baseline. Individuals with a history of psychosis prior to diabetes diagnosis had a significantly increased risk of mortality compared to individuals without prior psychosis (Table 2, Figure 1). In the Cox proportional hazards model, prior psychosis was associated with more than a two-fold increase in the hazard of death (HR = 2.08, 95% CI 1.90–2.27, p < 0.001). Older age and high clinical risk status were also associated with elevated mortality risk, whereas female sex was associated with lower mortality.
Table 1. Baseline characteristics of the study population |
| |||
|
Variable |
Psychosis + T2D (n=8034) |
T2D (n=134844) |
All (n=142878) |
p-value |
|
Age (mean (SD)) |
46.36 (9.50) |
48.82 (8.50) |
48.68 (8.58) |
<0.001 |
|
Gender Female (%) |
3658 (45.5) |
54932 (40.7) |
58590 (41.0) |
<0.001 |
|
HbA1c |
47.97 (12.90) |
51.56 (13.20) |
51.36 (13.21) |
<0.001 |
|
Low-density lipoprotein cholesterol |
2.57 (0.78) |
2.67 (0.82) |
2.66 (0.82) |
<0.001 |
|
Glomerular filtration rate |
96.19 (17.74) |
95.36 (16.52) |
95.41 (16.59) |
<0.001 |
Table 2. Adjusted Cox regression results for all-cause mortality |
|
|
|
|
Predictor |
Hazard Ratio |
95% CI |
p-value |
|
Psychosis before diabetes vs only T2D |
2.08 |
1.90 – 2.27 |
<0.001 |
|
Age (per year increase) |
1.03 |
1.02 – 1.03 |
<0.001 |
|
Female sex (vs male) |
0.65 |
0.61 – 0.69 |
<0.001 |
|
High medical risk (vs low) |
2.66 |
2.51 – 2.81 |
<0.001 |
Image:
Conclusions: In this nationwide cohort, individuals with psychosis and type 2 diabetes exhibited lower HbA1c and LDL levels at baseline compared with those with T2D alone, suggesting better glycemic and lipid treatment balance in the psychosis group. Despite the superior cardiometabolic markers, prior psychosis was associated with a increased risk of all-cause mortality.
Disclosure of Interest: None Declared