Diabetes subtypes linked to younger death in Jiangsu schizophrenia cohort

A retrospective registry analysis of people with schizophrenia whose underlying cause of death was diabetes found that male sex and an acute metabolic diabetes subtype were associated with slightly younger age at death and higher odds of dying before age 60. The study identified 1,335 decedents in Jiangsu Province, China, from 2014 through 2022. Their mean age at death was 63.72 years, with a standard deviation of 12.34 years, and 34.1% died before the age of 60.

The investigators combined records from the Jiangsu Province Management Information System for patients with severe mental illness with entries in the provincial Mortality Registration System. Eligible decedents had schizophrenia and diabetes recorded as the underlying cause of death under ICD-10 codes E10–E14. By comparing underlying-cause diabetes subtypes within this selected group, the analysis examined how age at death varied among registered schizophrenia decedents.

This was a decedent-based retrospective study, not an analysis of all people with schizophrenia or an assessment of mortality risk across the general population. It included only selected deaths in which diabetes was recorded as the underlying cause. Consequently, the results characterize differences within the 1,335-person decedent cohort and do not establish population-wide incidence, prognosis, or clinically established mortality risks for people with schizophrenia and diabetes.

Men in the cohort had a younger mean age at death than women. This sex difference was supported by a t statistic of -5.750 with P < 0.001. Men also represented a larger proportion of deaths before age 60, with a chi-square statistic of 23.906 and P < 0.001. In the multivariable analyses, male sex was associated with both a slightly younger age at death and higher odds of being younger than 60 at death.

The acute metabolic subtype showed a similar pattern. In multivariable analyses, it was associated with a slightly younger age at death and higher odds of death before age 60. The adjusted difference in age at death was approximately 3.05 years younger than in the reference group without complications. Thus, in the statistical models used for this cohort, the acute metabolic subtype and male sex were associated with the two age-at-death outcomes examined.

The chronic renal subtype produced a less uniform result. It was associated with a modestly younger age at death in the linear model, with an adjusted difference of approximately 2.28 years compared with the no-complications subtype. However, it was not statistically significantly associated with the binary outcome of death before age 60 in the logistic model. The study therefore reported a difference between the models for this subtype rather than a consistent association across both outcomes.

The investigators characterized the overall findings as descriptive epidemiological differences among schizophrenia decedents with diabetes recorded as the underlying cause of death. The estimates for the acute metabolic and chronic renal subtypes were modest, and statistical significance varied by outcome and model. The analysis does not show that sex or diabetes subtype caused earlier death, nor does it establish that the observed patterns can be used to predict prognosis for an individual patient.

The retrospective design and restriction to decedents also limit interpretation. The analysis cannot directly estimate the incidence of diabetes-related death among all people with schizophrenia, establish temporal mortality risk for all affected patients, or determine what produced the observed subgroup patterns. Its evidence is limited to registered death information and the subgroup comparisons made within Jiangsu Province over 2014–2022. Within that defined cohort of 1,335 decedents, the clearest reported associations were between male sex or the acute metabolic subtype and younger age at death and higher odds of dying before 60, while the chronic renal subtype was associated with younger age in the linear model but not significantly with the before-60 outcome.

Subject of Research: Retrospective registry analysis of diabetes underlying-cause subtypes and age at death among people with schizophrenia in Jiangsu Province, China, 2014–2022.

Article Title: Diabetes underlying-cause subtypes and age at death among individuals with schizophrenia: a decedent-based registry study in Jiangsu Province, China, 2014–2022

Article References: Zhu, L., Zhong, Y., Yang, Y., Wang, L., Wu, C., Li, Y., Zhuang, L., Yang, X., Fang, X., & Zhang, X. (2026). Diabetes underlying-cause subtypes and age at death among individuals with schizophrenia: a decedent-based registry study in Jiangsu Province, China, 2014–2022. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08782-8

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08782-8

Keywords: schizophrenia, diabetes, age at death, premature mortality, registry study, Jiangsu Province, acute metabolic subtype, chronic renal subtype, underlying cause of death, retrospective analysis, underlying-cause, subtypes