After the ICU: Teenagers Face Sharply Higher Risks of Depression and PTSD, Landmark Study Finds

Surviving a stay in the pediatric intensive care unit is often celebrated as the end of a medical crisis, but a sweeping new study suggests that for many children, it may be the beginning of a quieter, longer struggle. A nationwide analysis of South Korea’s health insurance records has found that children who survive an ICU admission carry substantially elevated risks of being diagnosed with psychological and neurodevelopmental disorders in the years that follow, and that the nature of those risks depends sharply on how old the child is when they are admitted.

The research, published in BMC Psychiatry by a team at Hanyang University College of Medicine, drew on the Korean National Health Insurance Service database covering the years 2015 to 2019. Because South Korea maintains a single-payer insurance system that captures virtually the entire population, the study offers something rare in pediatric research: a genuinely population-based window into what happens to children after they leave intensive care, rather than a picture distorted by the referral patterns of a handful of academic hospitals.

The investigators stratified their cohort into three age bands: young children under six years old, school-aged children from six to just under thirteen, and adolescents from thirteen to just under eighteen. They then tracked first recorded diagnoses of depression, anxiety disorders, post-traumatic stress disorder, and neurodevelopmental disorders over a follow-up period that extended as long as 8.63 years. Crucially, the team used Fine–Gray subdistribution hazard models, a statistical technique that accounts for death as a competing risk. That matters because children who die after ICU admission can never go on to receive a psychiatric diagnosis, and ignoring those deaths would bias the estimated risks of mental illness among survivors.

The headline finding is stark. Adolescents who survived an ICU admission showed dramatically higher risks of mood and trauma-related disorders than the youngest children. Compared with children under six, adolescents had an adjusted subdistribution hazard ratio of 13.81 for depression, meaning their hazard of receiving a first depression diagnosis was nearly fourteen times higher. The corresponding figures were 5.50 for anxiety disorders and 6.39 for post-traumatic stress disorder, with confidence intervals that excluded the null value in every case. In practical terms, the teenage brain emerging from critical illness appears to be a brain at greatly heightened vulnerability to disorders of mood and fear.

The pattern reverses when the researchers turned to neurodevelopmental outcomes. Young children, those under six at the time of admission, were far more likely than adolescents to receive diagnoses of language developmental disorders and autism spectrum disorders in the years after their ICU stay. When adolescents were compared directly with the youngest group, the hazard ratio for language developmental disorders was just 0.07, and for autism spectrum disorders 0.17, indicating that the burden of these diagnoses falls overwhelmingly on the youngest survivors. The likely explanation lies in developmental timing: the preschool years are precisely when language and social-communication milestones are expected to unfold, so a critical illness that disrupts development at that stage is far more likely to be detected, and to matter, than the same disruption would be in a teenager whose language systems are already mature.

One result stood apart from the age gradient. Attention deficit hyperactivity disorder showed no significant age-dependent variation in risk, suggesting that whatever mechanism links critical illness to later ADHD operates more uniformly across childhood, or that the diagnosis is detected at similar rates regardless of when the ICU admission occurred. The researchers’ overall conclusion is that follow-up care after pediatric intensive care should not be one-size-fits-all. Instead, they argue, it should be age-stratified: neurodevelopmental surveillance for the youngest survivors, and systematic screening for depression, anxiety, and PTSD in adolescents.

The study speaks directly to a growing body of clinical concern known as post-intensive care syndrome in pediatrics, abbreviated PICS-p. Clinicians have long recognized that surviving critical illness in childhood involves more than physical recovery. Prolonged sedation, mechanical ventilation, immobilization, delirium, and the sheer terror of the ICU environment are all plausible contributors to lasting psychological harm, and families frequently report behavioral changes, school difficulties, and emotional distress long after discharge. What has been missing, until now, is large-scale epidemiological evidence quantifying those risks across the whole population of survivors rather than in selected cohorts.

The methodological choices in the new study strengthen its claims. By using nationwide claims data coded with the International Classification of Diseases, 10th Revision, the researchers could capture diagnoses made across the entire Korean health system, in primary care as well as psychiatry. By adjusting for comorbidity burden, measured with the Charlson comorbidity index, they partially controlled for the possibility that sicker children are simply more likely to develop psychiatric problems for reasons unrelated to the ICU itself. And by modeling death as a competing risk rather than censoring it, they avoided one of the classic pitfalls of survivorship research. The study was approved by the Health Insurance Review and Assessment Service and the Institutional Review Board of Hanyang University Hospital, with informed consent waived owing to the retrospective design, and the authors declared no competing interests.

Still, the findings come with the caveats inherent to any observational claims-data study. Diagnostic codes reflect what clinicians record, and recording practices may differ across ages: an adolescent’s depression is more likely to be recognized and coded than a toddler’s internalizing distress, while developmental disorders are naturally diagnosed in early childhood. Some of the enormous age gap in depression risk may therefore reflect detection as well as true incidence. The authors also note that the interpretations and conclusions do not represent those of the National Health Insurance Service. Yet even with those qualifications, the direction and magnitude of the effects are difficult to dismiss, and the consistency of the age-stratified pattern across multiple diagnostic categories lends weight to the underlying biological and psychological story.

For intensive care physicians, pediatricians, and mental health services, the message is actionable. A child who leaves the ICU with recovered lungs or a healed injury may still carry invisible wounds, and the form those wounds take depends on developmental stage. The study’s authors suggest that post-discharge psychological follow-up after pediatric ICU admission may benefit from pairing neurodevelopmental surveillance in younger children with mood-disorder screening in adolescents. If such screening becomes routine, some of the nearly fourteenfold elevation in depression risk among teenage survivors could be caught early, treated, and softened. The research was funded by the research fund of Hanyang University, and the full open-access article is available under a Creative Commons Attribution 4.0 license, allowing clinicians and researchers worldwide to examine the evidence in detail.

Subject of Research: Long-term psychological and neurodevelopmental disorders in children after pediatric intensive care unit admission

Article Title: Long-term psychological disorders in children after intensive care unit admission: a population-based study

Article References: Kim, J., Yoo, K. H., Lee, J., Oh, J., Choi, N., Lim, T. H., Kang, H., Ko, B. S., & Cho, Y. (2026). Long-term psychological disorders in children after intensive care unit admission: a population-based study. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08643-4

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08643-4

Keywords: pediatric intensive care, post-intensive care syndrome, depression, anxiety disorders, PTSD, neurodevelopmental disorders, autism spectrum disorder, ADHD, adolescent mental health, population-based cohort study, competing risk analysis, child development