Schizophrenia in Children: How It Differs From Adult Forms

Schizophrenia can and does occur in children, though it is rare enough that many clinicians go an entire career without encountering a case. Roughly 8% of all people eventually diagnosed with schizophrenia receive that diagnosis before age 18, and about 18% report that their first symptoms appeared before that age.¹ When it strikes before puberty, the condition is typically called childhood-onset schizophrenia (COS), and it tends to look different, hit harder, and prove more difficult to manage than the adult form most people picture when they hear the word.

How It Differs From the Adult Form

Schizophrenia in adults usually surfaces in the late teens or early twenties. COS, defined as onset before age 13, is far less common and tends to develop more gradually. Rather than a sudden psychotic break, children with COS often show a slow unraveling across several years, with developmental problems appearing long before anyone suspects psychosis. More than half of children later diagnosed with COS have had developmental difficulties in language, motor coordination, or social behavior well before their first hallucination or delusion.² Compared to people who develop schizophrenia as adults, children with COS show greater delays in language development, more speech and language disorders, more learning difficulties, and more disruptive behavior before the psychosis begins.³

Transient symptoms that look like autism, such as motor stereotypies (repetitive movements like hand-flapping or rocking), are also common before psychotic symptoms emerge.⁴ This overlap between early schizophrenia and autism-like features is one of the reasons getting the right diagnosis takes so long. For boys in particular, premorbid motor abnormalities are especially pronounced, though early language and social impairments do not differ meaningfully between boys and girls.⁵

Why It Is So Hard to Diagnose

One of the central challenges with COS is that young children who are hallucinating or confused are often assumed to have something else. The conditions most commonly mistaken for COS include mood disorders with psychotic features, autism spectrum conditions, anxiety disorders, and trauma responses. At the National Institute of Mental Health, researchers running a large COS study found that many children referred to them with clinical diagnoses of schizophrenia actually had psychotic mood disorders once they underwent a thorough research evaluation with structured diagnostic interviews.⁶ The reliable diagnosis of COS requires a time-consuming evaluation process that most community clinics are not equipped to provide.

The overlap with autism is particularly tricky. In two large systematic studies, between 30% and 50% of children with COS also met criteria for a pervasive developmental disorder (an older term that encompassed what is now called autism spectrum disorder), either as a preceding condition or as a co-occurring one.⁷ That means a child might carry both diagnoses simultaneously, or the autism-like features might be a developmental precursor that later shades into psychosis. Teasing the two apart is not just an academic exercise; the treatments are quite different, and misidentification delays appropriate care.

Historically, the confusion ran even deeper. From the 1930s through the late 1970s, “childhood schizophrenia” was used in the United States as a broad catch-all diagnosis that included what we would now recognize as autism. It was only with the publication of the DSM-III in 1980 that autism was explicitly introduced as a separate diagnosis, and the older, more expansive notion of childhood schizophrenia was abandoned. Understanding that history matters because some older studies on “childhood schizophrenia” were describing a far more heterogeneous group of children than the diagnosis captures today.

The Genetic Picture

COS appears to sit at the more severe end of the schizophrenia spectrum genetically. Families of children with COS show a higher clustering of schizophrenia-spectrum disorders compared to families where schizophrenia first appears in adulthood.⁸ In other words, relatives of a child with COS are more likely to have some form of the condition themselves than relatives of an adult who develops it later.

Genetic scans have also turned up a higher rate of rare copy number variants, which are small deletions or duplications of stretches of DNA, in children with COS compared to what would be expected by chance and even compared to rates seen in adult-onset schizophrenia.⁹ This suggests that COS may involve a heavier “genetic loading” than the adult form, which could help explain both why it strikes earlier and why it tends to be more severe. None of this implies that a single gene causes the disorder; like adult schizophrenia, COS involves many genetic risk factors interacting with each other and with environmental conditions.

What Brain Imaging Shows

Some of the most striking findings in COS research come from longitudinal brain imaging studies, many of them conducted at the NIMH. Children with COS lose gray matter, the outer layer of the brain where most processing happens, at a markedly faster rate during adolescence than healthy children do. One study tracking 15 children with COS over time found a significant decrease in total brain volume, driven by an 8% reduction in gray matter, with the frontal, parietal, and temporal regions losing between 7% and 10% of their volume.¹⁰ The occipital region, which handles vision, was largely spared.

This accelerated gray matter loss in COS is not a sign of an entirely different disease process from adult schizophrenia. Rather, it appears to be an exaggerated version of the normal cortical thinning that happens during adolescent brain development. As children with COS grow into young adults, their brain scans eventually come to resemble those of adults who developed schizophrenia later in life.¹¹ The gray matter loss was significantly greater in children with COS than in children with other psychotic conditions or in healthy controls, suggesting that this pattern is somewhat specific to schizophrenia rather than psychosis in general.¹²

One counterintuitive finding: children who showed greater clinical improvement actually had a higher rate of gray matter loss, not a lower one.¹³ The researchers did not have a clear explanation for this, though one possibility is that some of the tissue loss reflects a healthy pruning process that just happens too aggressively in COS. Reassuringly, the rate of cortical loss appeared to plateau during late adolescence, suggesting the brain is not simply deteriorating indefinitely.¹⁴

Cognitive Effects

Children with COS tend to experience more severe cognitive difficulties than people whose schizophrenia begins later. Their IQ scores typically fall one to two standard deviations below the general population average, and these scores often drop between the premorbid period and the time after psychosis begins, suggesting an active decline rather than a lifelong stable deficit.¹⁵ Attention, working memory, processing speed, planning, and mental flexibility are all affected. Children with COS show longer reaction times, more errors on attention tasks, and slower processing speeds compared to healthy peers.

These cognitive deficits carry obvious consequences for school performance. A study looking at whether individualized education plans (IEPs) helped children with early-onset schizophrenia found that having an IEP did not improve absenteeism or academic performance, and dropout rates were the same regardless of whether a child had one.¹⁶ That finding should not be taken as evidence that educational support is pointless; it more likely reflects the fact that standard IEPs are not designed for the specific challenges posed by a psychotic illness in a child. More tailored approaches are needed, though the research on what those approaches should look like remains thin.

Medication Treatment

Antipsychotic medications are the primary treatment for COS, just as they are for adult schizophrenia, but the side-effect profile in children is concerning. The second-generation (atypical) antipsychotics that are most commonly used have been consistently linked to weight gain and moderate elevations in prolactin when used in children. The risk of weight gain and elevated prolactin may actually be higher in younger children than in adolescents or adults.¹⁷ Beyond weight, metabolic problems emerge quickly. After starting these medications, children and adolescents with early-onset schizophrenia showed significant increases in body mass index, triglycerides, and cholesterol, with about a third developing abnormal triglyceride or cholesterol levels.¹⁸ Some medications appeared worse than others: olanzapine and quetiapine produced greater increases in triglycerides than other options.

These metabolic side effects are not minor inconveniences. For a child who may need lifelong antipsychotic treatment, significant weight gain and dyslipidemia in the first year can set the stage for cardiovascular problems and diabetes decades earlier than they would otherwise appear. This makes careful metabolic monitoring essential from the start, and it forces difficult conversations between clinicians and families about balancing symptom control against long-term physical health.

When Standard Medications Fail

A substantial portion of children with COS do not respond adequately to first-line antipsychotics, which brings clozapine into the conversation. Clozapine is widely considered the most effective antipsychotic for treatment-resistant schizophrenia, and a systematic review and meta-analysis found that it outperformed other antipsychotics in both short-term (six-week) and long-term (two to nine years) management of childhood and adolescent-onset schizophrenia.¹⁹ Clozapine also reduces aggression and suicidality, which are real concerns in this population.²⁰

The trade-off is a side-effect profile that requires constant vigilance. Sedation and excessive salivation are reported in up to 90% of young patients on clozapine, and constipation occurs in roughly 13% to 50%.²¹ More dangerously, neutropenia (a drop in infection-fighting white blood cells) occurs in 6% to 15% of cases, though full-blown agranulocytosis, the most feared blood-related side effect, is very rare. Children on clozapine also face an increased risk of seizures compared to adults.²² The drug is used off-label in this population, and prescribing guidelines for pediatric use remain inconsistent, which means the quality of monitoring a child receives can vary widely depending on where they are treated.

Therapy and Psychosocial Support

Medication alone does not address many of the challenges children with COS face: navigating school, maintaining friendships, understanding their illness, and managing the daily fear and confusion that psychosis can cause. Cognitive behavioral therapy adapted for young people with psychosis is a growing area of interest, though the evidence base is still young. A randomized pilot study of modified CBT in adolescents with early-onset psychosis found that the approach was safe, well tolerated (no dropouts, and over 80% session attendance), and showed moderate effect sizes favoring CBT for delusions, negative symptoms, functioning, and quality of life. Notably, the group receiving CBT plus standard care had one serious adverse event over the study period compared to six in the group receiving standard care alone.²³ The sample was too small to draw firm conclusions, but the direction of the findings is encouraging.

Family psychoeducation, where families learn about the illness, its management, and communication strategies, is considered a best practice in adult schizophrenia care. For children and adolescents, however, the evidence base is thinner than many clinicians would like. Research on family psychoeducation has been conducted primarily with adults, and reviews have pointed out that models specific to younger patients remain largely untested. That gap matters because the dynamics of a family caring for a child with psychosis are fundamentally different from those of a family supporting an adult relative. Parents are simultaneously managing a psychiatric crisis, making treatment decisions on behalf of someone who cannot fully consent, and grieving the developmental trajectory they expected for their child.

Long-Term Outlook

The honest picture here is sobering. A systematic review of long-term outcome studies in early-onset schizophrenia found that only about 15% of patients experienced what researchers classified as a “good” outcome, roughly a quarter had a “moderate” outcome, and about 60% fell into the “poor” category.²⁴ Those numbers are worse than what is typically reported for adults with schizophrenia. The review also identified male sex and longer follow-up periods as predictors of poorer outcomes, though it acknowledged that patients diagnosed in earlier decades, when treatments were less sophisticated, dragged the averages down.

More recent data from a six-year follow-up study in China offers a somewhat more encouraging snapshot. At the end of follow-up, about half of the patients were still in school and roughly 40% were employed, with only about 7% having dropped out of both school and work entirely.²⁵ How quickly a child received their first hospital admission and the presence of depressive symptoms both predicted better educational and occupational outcomes. That aligns with a broader theme in psychosis research: early, aggressive treatment tends to improve trajectories. One curious finding from the same study was that co-occurring obsessive-compulsive symptoms appeared to act as a protective factor against readmission, a result that the authors found somewhat unexpected but that has been observed in some adult schizophrenia research as well.

Immune and Inflammatory Markers

A relatively new area of investigation looks at whether children and adolescents with psychotic disorders show differences in immune function that could serve as biomarkers or point toward new treatment targets. A meta-analysis of studies comparing youth with psychotic disorders to healthy peers found elevated levels of several inflammatory markers, including tumor necrosis factor, C-reactive protein, interleukin-6, total white blood cell count, and the neutrophil-to-lymphocyte ratio.²⁶ The effect sizes were small to moderate, so these markers are nowhere near precise enough to serve as a diagnostic test. But they add to a growing body of evidence that schizophrenia involves the immune system as well as the brain, and that this inflammatory component is already present in youth, before decades of medication and lifestyle changes could explain it.

What this means practically is still unclear. No one is recommending anti-inflammatory treatments for COS based on this research. But the findings are shaping how researchers think about the disease, moving it away from a purely “brain wiring” model and toward one that incorporates the body’s inflammatory system. For families, the most immediate takeaway is that children with COS likely benefit from the same cardiovascular and metabolic monitoring that adults with schizophrenia receive, and probably should get it earlier and more consistently than they often do.

The Role of Childhood Trauma

The relationship between early adversity and schizophrenia risk is real, but it is easy to overstate. A study examining childhood trauma and premorbid adjustment in first-episode schizophrenia found significant correlations between trauma history and functioning across all life stages, in both social and academic domains, though the effect sizes were small to medium rather than dramatic.²⁷ The correlations were present across all categories of trauma assessed, suggesting that it is cumulative adversity rather than one specific type of trauma that matters most.

For parents of a child diagnosed with COS, this research is easily misread as blame, which is not what it shows. The genetic loading in COS is substantial, and no amount of perfect parenting prevents a genetically predisposed brain from developing psychosis. What the trauma research does suggest is that stressful environments can worsen the developmental trajectory in someone who already carries biological vulnerability, and that reducing stress and increasing stability around a child with emerging symptoms is a worthwhile goal alongside medical treatment. It also means that clinicians evaluating a child for possible psychosis should screen for trauma history, because untreated trauma responses can mimic or exacerbate psychotic symptoms and complicate the diagnostic picture further.

Ethical Challenges in Research and Treatment

Studying COS is uniquely difficult, and not just because the condition is rare. Research protocols that are routine in adult psychiatry become ethically complicated when the patient is a child. Medication-free research periods, for example, have been used at the NIMH to study children with COS with parental consent, but the idea of withdrawing antipsychotic medication from a psychotic child, even briefly and under close supervision, raises questions that do not arise as sharply with adult volunteers who can consent for themselves.²⁸

The rarity of COS compounds the problem. With so few cases available for study, sample sizes are small, replication is difficult, and many treatment decisions are extrapolated from adult data rather than directly tested in children. Clozapine’s use in children, as noted above, is entirely off-label. Most of the antipsychotics prescribed to children with COS were tested and approved in adults, and pediatric dosing guidelines are often based on clinical experience rather than rigorous trials. Families navigating this territory deserve to know that the evidence guiding their child’s treatment is thinner than they might assume, and that honest uncertainty from a clinician is often a better sign than false confidence.