The following is a summary of “Causes of death in individuals with lifetime major depression: a comprehensive machine learning analysis from a community-based autopsy center,” published in the July 2024 issue of Psychiatry by Nunes et al.

While depression is linked to higher mortality and morbidity, specific causes of death (CoD) have not been explored through autopsy studies, which can provide detailed insights into pathological or underreported conditions.

Researchers conducted a retrospective study comparing autopsy-confirmed CoD between patients with major depressive disorder (MDD) and controls, analyzing MDD subgroups and evaluating if machine learning (ML) algorithms can distinguish MDD and subgroups from controls based on the cause.

They performed analysis for CoD in individuals who died from non-traumatic causes. The diagnosis of lifetime MDD was confirmed using DSM-5 criteria and information from a structured interview with a knowledgeable informant. A total of 11 ML algorithms were utilized to distinguish individuals with MDD from controls by concurrently examining various disease category groups to handle multiple comparisons. Additionally, the McNemar test was used to compare paired nominal data.

The result showed 1,102 individuals, with 232 (21.1%) having a lifetime diagnosis of MDD. Each patient with MDD was matched with an HC. The most prevalent CoD were circulatory (67.2%), respiratory (13.4%), digestive (6.0%), and cancer (5.6%). Despite using various ML models, no distinctive patterns related to CoD were found that could reliably differentiate individuals with MDD from HCs (average accuracy: 50.6%; range: 39-59%). The lack of distinction remained even when analyzing subgroups within MDD, such as late-life or recurrent MDD. No significant differences were found between groups for circulatory (P=0.450), respiratory (P=0.790), digestive (P=1.000), or cancer (P=0.855) CoD.

Investigators concluded that similar autopsy-confirmed CoD in individuals with MDD and controls, highlighting the need for future research on severe depression and larger samples, especially for cancer-related deaths.

Source: bmcpsychiatry.biomedcentral.com/articles/10.1186/s12888-024-05946-2