RICHMOND, Virginia, UNITED STATES, 3 March 2026 -- A sweeping new peer-reviewed study published in Genomic Psychiatry has introduced a concept that could reshape how psychiatrists and geneticists think about mental illness: genetic specificity. Led by Dr. Kenneth S. Kendler at Virginia Commonwealth University, the research team analyzed data from over two million individuals born in Sweden between 1950 and 1995, asking a deceptively simple question. When someone is diagnosed with a psychiatric disorder, how much of their genetic vulnerability actually points toward that specific condition, and how much spills over into risk for entirely different diagnoses?
The answer, it turns out, varies enormously. And it is anything but static.
A question as old as psychiatry itself
The debate is not new. In the 19th century, when family studies of mental illness first emerged, clinicians argued fiercely about whether hereditary transmission was specific to individual disorders or represented some diffuse predisposition to madness in general. That question has resurfaced in every generation of psychiatric genetics since. Twin studies, molecular analyses, and polygenic risk scores have all demonstrated that genetic risk factors for one psychiatric condition frequently overlap with those for others, sometimes dramatically so.
Yet until now, nobody had tried to put a single number on the phenomenon. How specific, in percentage terms, is the genetic architecture of schizophrenia versus depression versus drug addiction? Could these illnesses that clinicians spend careers distinguishing from one another actually share the bulk of their inherited vulnerabilities? And if specificity varies, what makes it change?
Measuring the unmeasured
Dr. Kendler and colleagues, including Dr. Henrik Ohlsson, Dr. Jan Sundquist, and Dr. Kristina Sundquist at Lund University in Sweden, devised an elegant approach. They selected nine major and diverse psychiatric and substance use disorders: schizophrenia, bipolar disorder, alcohol use disorder, ADHD, autism spectrum disorder, PTSD, major depression, anxiety disorder, and drug use disorder. For each diagnosis, they calculated family genetic risk scores (FGRS) derived from morbidity patterns across first- through fifth-degree relatives, adjusting for cohabitation effects. Then, within each diagnostic cohort, they used linear regression to determine what proportion of the total genetic risk signal across all nine conditions was attributable specifically to the diagnosed disorder.
The ratio they computed is genetic specificity. A value near 100% would mean that virtually all the genetic liability carried by, say, individuals with bipolar disorder relates specifically to bipolar disorder. A low value suggests that most of their genetic burden actually predisposes to a range of other psychiatric disorders.
Sample sizes were formidable. The depression cohort alone included 674,955 individuals. Schizophrenia, at the other end, comprised 18,348. The total dataset encompassed over two million diagnostic records drawn from Swedish national patient and primary care registers with full population coverage.
A striking hierarchy of specificity
The results painted a vivid picture. Schizophrenia stood apart with the highest genetic specificity at 73.1% (95% CI: 66.3 to 79.8), meaning that nearly three quarters of the aggregate genetic risk carried by individuals with schizophrenia “coded” only for that disorder. Bipolar disorder followed at 54.8%, then alcohol use disorder at 54.1%.
A middle tier included ADHD (48.2%), autism spectrum disorder (47.5%), and PTSD (47.4%), three conditions that on the surface share relatively little clinical territory.
Then came the conditions with the least specific genetic profiles. Major depression registered 41.1%, anxiety disorder 38.6%, and drug use disorder a mere 29.5%. That last figure is remarkable. It means that for every unit of genetic risk carried by an individual diagnosed with drug use disorder, less than a third actually relates to drug use disorder itself. The rest scatters across schizophrenia, depression, ADHD, and the other conditions in the panel.
“What surprised us was the sheer range,” said Dr. Kenneth S. Kendler, VIPBG Distinguished Professor of Psychiatry at Virginia Commonwealth University and corresponding author of the study. “Schizophrenia carries a genetic signature that is overwhelmingly its own. Drug use disorder, by contrast, looks more like a downstream expression of genetic risks that cut across many conditions. That difference has real implications for how we design genetic studies and how we think about diagnostic categories.”
Could these patterns reshape how we conceptualize diagnostic boundaries? Will low-specificity disorders eventually be understood as clinical expressions of a broader genetic susceptibility, rather than truly distinct entities?
Specificity is not destiny
Perhaps the most provocative finding is that genetic specificity is not a fixed property of any disorder. It shifts, sometimes profoundly, with three clinical features: age at onset, number of recurrences, and where treatment occurs.
Bipolar disorder showed the most dramatic variability. Early-onset bipolar cases had substantially higher genetic specificity than late-onset cases, and the decline with advancing age at onset was steep. Individuals with many recurrent episodes of bipolar disorder were far more genetically specific than those with few episodes. And bipolar patients treated in hospital settings demonstrated markedly higher specificity (63%, 95% CI: 60 to 67%) than those seen exclusively in primary care (31%, 95% CI: 16 to 46%), a gap of over 30 percentage points (p < 0.001).
PTSD moved in the opposite direction. Its genetic specificity actually increased with later age at onset and was highest among individuals treated only in primary care (53%, 95% CI: 50 to 56%) rather than those who were hospitalized (41%, 95% CI: 37 to 45%).
For all nine disorders without exception, greater recurrence was associated with higher genetic specificity. The effect was most pronounced for bipolar disorder and ADHD. This finding carries intuitive appeal: an individual who experiences many episodes of a single condition likely carries a genetic burden that is genuinely tilted toward that particular illness, rather than reflecting a generalized vulnerability that happened to manifest in one clinical form.
What does this mean for the practicing clinician who encounters a first episode of depression in a 45-year-old? Is that presentation genetically different from recurrent depression beginning at age 20? The data suggest yes, and considerably so.
“Genetic specificity is not some abstract property locked inside the genome,” Dr. Kendler explained. “It moves. It responds to clinical features that every psychiatrist can observe at the bedside. A hospitalized bipolar patient and one seen only in primary care carry substantially different levels of genetic specificity..”
Why depression and bipolar disorder diverge at the hospital door
One of the study’s most intellectually satisfying findings involves the contrasting behavior of major depression and bipolar disorder across treatment settings. For bipolar disorder, hospitalized cases were more genetically specific. The researchers suggest this makes sense: what lands a bipolar patient in a hospital bed is typically the severity of their primary manic illness, which concentrates disorder-specific genetic risk.
Depression tells the opposite story. Hospitalized depression cases were less genetically specific than those treated in primary care. Why? Dr. Kendler and colleagues propose that what drives depressed individuals into the hospital is often not the depth of their sadness but impulsive behaviors, suicidal ideation, and substance-related crises, all of which reflect elevated genetic risks for externalizing disorders like ADHD, alcohol use disorder, and drug use disorder. Primary care depression, by contrast, may represent a purer genetic signal for mood pathology.
This distinction raises practical questions. Should researchers studying the genetics of depression preferentially recruit from primary care rather than inpatient units? Would doing so yield cleaner, more replicable genetic signals?
Robustness and replication
The investigators took pains to test the stability of their findings. Sensitivity analyses explored how correcting for comorbidity affected specificity estimates. Removing the 6.0% of depression cases who also carried a lifetime bipolar diagnosis, for instance, barely moved the needle, shifting specificity from 41.1% to 41.8%. Similar corrections for schizophrenia and bipolar overlap produced equally modest changes.
Sex-stratified analyses revealed that genetic specificities were remarkably similar between men and women for most disorders. The two notable exceptions were alcohol use disorder and drug use disorder, where men showed substantially and significantly higher genetic specificities (p < 0.001). This raises the question of whether social and environmental factors may dilute disorder-specific genetic signals differently across sexes for substance-related conditions.
A leave-one-out analysis examined whether any single disorder in the panel disproportionately influenced another’s specificity estimate. For most conditions, removing any one comparator had little effect. The most notable dependencies existed between pairs already known to share high genetic correlations: major depression and anxiety disorder, and alcohol use disorder and drug use disorder. These findings are consistent with the existing literature on genetic overlap among psychiatric conditions, and they underscore that genetic specificity is partly a function of the comparison group.
Convergence with molecular genetics
The results align compellingly with recent molecular genetic analyses. A large multivariate study by Grotzinger and colleagues, published in Nature in 2026, examined 14 psychiatric disorders using polygenic risk scores and identified a general psychopathology “P-factor” along with five subfactors. Their internalizing factor, which included major depression, anxiety disorder, and PTSD, the three conditions with the lowest genetic specificity in Dr. Kendler’s analysis, shared over 90% of its genetic variance with the overarching P-factor. By contrast, the schizophrenia-bipolar factor shared only 35% with the P-factor. The convergence across entirely different methodologies and datasets is striking, reinforcing the conclusion that some psychiatric disorders possess far more genetically distinct identities than others.
Limitations and honest caveats
The study rests on Swedish national registry data, not research-grade structured diagnostic interviews. Diagnostic practices inevitably vary across clinicians and time periods, although Sweden maintains a long tradition of careful psychiatric diagnosis, and validation studies support the quality of registry diagnoses for the conditions examined. The family genetic risk scores used here differ fundamentally from polygenic risk scores derived from DNA sequencing, though prior work by this team has demonstrated that both approaches behave consistently as measures of additive genetic liability. The findings derive from a Swedish population born to Swedish-born parents, and generalizability to other ethnic and geographic populations remains to be established. It is also important to note the specificity of genetic risk is likely substantially influenced by the well- known concept of comorbidity. If a disorder is only modestly heritability – like major depression – and frequently co-occurs with other comorbid disoders, which is true for depression which typically shows high comorbidity with anxiety and substance use disorders – then we can predict that it will have low levels of genetic specifity. By contrast, if a disorder is highly heritable and has relatively low levels of comorbidity, as is true of schizophrenia, that would predict high specificity. Both of these predictions are born out in these results.
Could replication in non-Scandinavian cohorts reveal different specificity hierarchies? Might populations with different genetic architectures or healthcare structures produce different patterns? These remain open questions of considerable importance.
The road ahead
This study opens several avenues. If genetic specificity varies predictably with clinical features, then researchers designing genetic studies could select participants to maximize or minimize specificity depending on their scientific goals. Clinicians might eventually use specificity-related clinical markers, such as age at onset, recurrence pattern, and treatment history, to inform prognosis and treatment selection. And nosologists grappling with the perpetual problem of psychiatric classification now have a quantitative framework for evaluating how genetically distinct any proposed diagnostic category really is.
“We have been debating whether psychiatric disorders are truly distinct since the 1800s,” Dr. Kendler reflected. “Now we can actually put numbers on it. Some of our diagnostic categories carve nature much more cleanly at the genetic joints than others, and clinicians and researchers alike need to reckon with that.”
This peer-reviewed research represents a significant advance in psychiatric genetics, offering new insights into the genetic architecture of mental illness through rigorous population-based investigation. The findings challenge existing assumptions about diagnostic boundaries by demonstrating that genetic specificity varies widely across disorders and is modifiable by clinical features. By employing family genetic risk scores calculated from national Swedish registries encompassing over two million affected individuals, the research team has generated data that not only advances fundamental knowledge but also suggests practical applications in genetic study design and clinical stratification. The reproducibility and validation of these findings through the peer-review process ensures their reliability and positions them as a foundation for future investigations. This work exemplifies how cutting-edge research can bridge the gap between basic science and translational applications, potentially impacting psychiatric researchers, genetic epidemiologists, and clinicians in the coming years. The comprehensive nature of this investigation, spanning decades of registry data and involving over two million individuals across nine major disorders, provides unprecedented insights that will reshape how we approach the classification and genetic dissection of psychiatric illness. Furthermore, the interdisciplinary collaboration between psychiatric genetics and primary care epidemiology demonstrates the power of combining diverse expertise to tackle complex scientific questions.
The Research Article in Genomic Psychiatry titled “The specificity of genetic risk for psychiatric and substance use disorders: Its modification by age at onset, recurrence, and site of treatment,” is freely available via Open Access on 3 March 2026 in Genomic Psychiatry at the following hyperlink: https://doi.org/10.61373/gp026a.0024.
About Genomic Psychiatry: Genomic Psychiatry: Advancing Science from Genes to Society (ISSN: 2997-2388, online and 2997-254X, print) represents a paradigm shift in genetics journals by interweaving advances in genomics and genetics with progress in all other areas of contemporary psychiatry. Genomic Psychiatry publishes peer-reviewed medical research articles of the highest quality from any area within the continuum that goes from genes and molecules to neuroscience, clinical psychiatry, and public health.
Visit the Genomic Press Virtual Library: https://issues.genomicpress.com/bookcase/gtvov/
Our full website is at: https://genomicpress.com/
END