The research, published in BMC Psychiatry by Emre Han Alpay of Mersin University, Cansu Ünsal Mavi of Kartal Dr. Lütfi Kırdar City Hospital in Istanbul, and Nida Tayyibe Karagöz of Mersin University, applied a technique known as latent profile analysis to data from 583 bereaved adults. Rather than asking whether individual symptoms predict one another, latent profile analysis searches the data for hidden subgroups of people who share similar overall patterns of symptom severity. The method is a workhorse of modern psychiatric taxonomy, and it has become especially important as the field grapples with the question of whether diagnostic categories carved out in manuals like the ICD-11 and DSM-5-TR correspond to genuinely distinct populations in the real world.

The study was framed explicitly within the eleventh revision of the International Classification of Diseases, the diagnostic system that has most boldly reshaped how the world thinks about trauma and loss. ICD-11 recognizes prolonged grief disorder as a distinct condition, characterized by persistent and pervasive yearning for or preoccupation with the deceased, accompanied by intense emotional pain and functional impairment that persists well beyond what cultural norms would expect. The same classification also introduced a more complex picture of post-traumatic stress disorder, splitting it into a core PTSD diagnosis and a sibling condition called complex PTSD, which adds what the ICD-11 calls disturbances in self-organization: problems with emotion regulation, negative self-concept, and difficulties in relationships. The Turkish team measured all of these domains, along with depressive symptoms, using validated self-report instruments including the International Prolonged Grief Disorder Scale, the International Trauma Questionnaire, and the International Depression Questionnaire.

What emerged from the analysis was a four-profile solution. The largest group, comprising 231 participants or 39.6 percent of the sample, showed low levels of symptoms across the board. A second group of 171 participants, or 29.3 percent, fell into a moderate-symptom profile. A third, elevated-symptom profile captured 142 people, or 24.4 percent. And a final, high-symptom profile accounted for 39 participants, or 6.7 percent of the sample. On the surface, that looks like a familiar story: a small, severely affected minority and a larger, relatively resilient majority, with gradations in between.

But the most scientifically interesting finding lies in how those profiles differed from one another. The researchers report that the four groups were separated primarily by overall symptom severity, with only modest differences in the relative configuration of symptoms. In other words, the profiles did not behave like cleanly distinct diagnostic types, where one group is dominated by grief, another by post-traumatic stress, and a third by depression. Instead, people in the more severe profiles tended to be elevated across all four symptom domains simultaneously, while those in milder profiles were low across the board. This pattern echoes a growing body of work within the Hierarchical Taxonomy of Psychopathology initiative, which argues that common liability to general distress may explain much of the comorbidity that traditional diagnostic categories treat as coincidental overlap.

That said, the analysis did hint at some variation in relative symptom configuration, suggesting that the shape of distress, not just its intensity, might differ from person to person. The authors are careful on this point. Because the study was cross-sectional and relied on self-report, and because algorithm-based classifications were treated as probable caseness rather than clinician-confirmed diagnoses, they explicitly caution that the findings do not establish clinically distinct subgroups. The profiles are statistical portraits of a community sample, not validated diagnostic types, and the researchers call for longitudinal research with independent clinical validation to determine whether these provisional patterns are stable over time and useful in the clinic.

One of the clearest secondary findings concerned traumatic life event exposure. Participants completed the Life Events Checklist for DSM-5, and the results showed that higher exposure to traumatic events was associated with membership in more severe symptom profiles. The effect, however, was small, with an eta-squared value of 0.03, meaning that traumatic exposure explained only a modest share of the differences between groups. This is a reminder that bereavement-related distress is multiply determined, shaped by the circumstances of the death, prior adversity, social support, personality, and cultural context, and that no single risk factor tells the whole story. It is also consistent with the idea that a history of trauma may sensitize people to the psychological impact of loss, stacking one burden on top of another.

The probable caseness figures add practical texture to the statistical picture. Rates of probable prolonged grief disorder, PTSD, complex PTSD, and depression generally increased as profiles became more severe, which is what one would expect if the profiles capture a genuine severity gradient rather than arbitrary statistical partitions. For clinicians, the implication is that screening for any one disorder in a bereaved patient may warrant attention to the others. A person presenting with disabling grief is statistically likely to carry elevated trauma-related and depressive symptoms as well, and treatment planning that addresses only the most visible complaint may miss the broader architecture of distress.

The study also carries significance for the ongoing international debate over prolonged grief disorder itself. Its inclusion in ICD-11 and, more recently, in the DSM-5-TR was controversial, with critics arguing that normal grief was being pathologized and defenders countering that a distinct, impairing condition had long gone unrecognized. Findings like these occupy a middle ground. They confirm that a substantial minority of bereaved people, here roughly one in fifteen in the most severe profile, experience intense, multi-domain distress, while also suggesting that this distress is not neatly separable from depression and trauma responses. Whether future longitudinal work reveals genuinely distinct grief-dominated subgroups, or whether severity remains the dominant axis of differentiation, will shape how the next generation of diagnostic manuals draws its boundaries.

For now, the message for the public is both sobering and reassuring. Grief is heterogeneous, and a meaningful fraction of bereaved people carry heavy, entangled burdens of grief, trauma, and depression that deserve professional attention. Yet the largest group in the study, nearly forty percent of participants, showed low symptom levels across every domain measured, a statistical testament to the resilience that most people bring to one of life’s hardest experiences. The Turkish team’s work, conducted without dedicated funding and published open access, offers a rigorous template for the kind of person-centered research that will be needed to turn these provisional profiles into clinically actionable knowledge, and it underscores that the science of loss is finally catching up with the complexity of the human experience of bereavement.

Subject of Research: Latent profiles of prolonged grief, PTSD, and depressive symptom co-occurrence in bereaved adults

Article Title: Profiles of symptom co-occurrence in bereavement: a latent profile analysis of grief, trauma-related, and depressive symptoms

Article References: Profiles of symptom co-occurrence in bereavement: a latent profile analysis of grief, trauma-related, and depressive symptoms. (n.d.). https://doi.org/10.1186/s12888-026-08599-5

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08599-5

Keywords: bereavement, prolonged grief disorder, PTSD, complex PTSD, depression, latent profile analysis, ICD-11, symptom co-occurrence, trauma exposure, psychopathology, Türkiye, mental health