Basic Neurosciences · Paper A
Heritability / familial risk
Family, twin, and adoption studies each measure genetic contribution to psychiatric illness in a different specific way, and one common exam trap involves misreading what heritability actually means. A condition with 90% heritability does NOT mean 90% of any one person's illness is 'caused by genes'. That population-level, not individual-level, meaning is the single most tested trap here.
In schizophreniaA long-term mental illness involving episodes of psychosis such as hallucinations and delusions.Read more →, general population risk is about 1%, rising to about 6-10% for a first-degree relative, 10-15% concordance for dizygotic (non-identical) twins, and 40-50% concordance for monozygotic (identical) twins. In bipolar disorder, population risk is about 1%, first-degree relative riskThe risk of an outcome in one group divided by the risk in another, showing how many times more or less likely it is.Read more → is 10-15%, and monozygotic concordance is high, around 40-70%. Autism has one of the highest heritability estimates in psychiatry, around 80-90%. The underlying logic is that identical twins share nearly 100% of their genes while non-identical twins share about 50%, so a higher concordance in identical than non-identical twins signals a genetic contribution. Adoption studies separate genetic from shared-environment effects by comparing adoptees with their biological versus adoptive relatives. e.g. identical twins where only one develops schizophrenia despite sharing all their genes: that incomplete concordance is itself proof that environment plays a real role, even in a highly heritable condition.
Exam questions on heritability almost always hinge on one misreading: treating a population-level statistic as if it described a single person's illness. Once you can hold "population variance," not "one person's cause," in your head, the whole topic (family studies, twin studies, adoption studies, and the genetic counselling that follows from them) falls into place.
Heritability is the proportion of variation in a trait ACROSS A POPULATION that can be attributed to genetic differences between people. It is not a statement about any one individual. Three study designs are used to estimate it. Family studies measure how a disorder clusters in relatives, without separating genes from shared environment (the upbringing, diet, and social conditions relatives happen to share). Twin studies compare monozygotic (identical) twins, who share nearly 100% of their genes, against dizygotic (non-identical) twins, who share about 50%, like ordinary siblings. Adoption studies separate genetic from environmental effects by comparing adoptees with their biological relatives, who share genes but not upbringing, versus their adoptive relatives, who share upbringing but not genes.
A common mistake is assuming an 80% heritability figure means a person's own illness is "80% caused by genes." That is always wrong. Heritability is a population-variance statistic: it tells you how much of the DIFFERENCE between people in a population is explained by genetic differences, not how much of any one case is genetic versus environmental. An individual's illness could still depend heavily on non-genetic factors even inside a highly heritable condition.
Imagine a condition with 90% heritability in the scientific literature. This does not mean any one person's illness is 90% genetic and 10% environmental. It means that, across a population, 90% of the variation in who develops the condition and who does not is statistically attributable to genetic variation between people. Now consider identical twins where only one develops despite sharing every gene. That incomplete concordance (the rate at which both twins in a pair share the same outcome) is itself proof that environment plays a genuine, measurable role, even in a highly heritable condition: if genes alone determined outcome, monozygotic concordance would approach 100%, and it clearly does not.
In , general population risk is roughly 1%, rising to about 6-10% for a first-degree relative (a parent, sibling, or child), around 10-15% concordance for dizygotic twins, and 40-50% concordance for monozygotic twins. In bipolar disorder, population risk is also about 1%, first-degree is 10-15%, and monozygotic concordance is higher still, around 40-70%. Autism has one of the highest heritability estimates in psychiatry, around 80-90%. The underlying logic behind twin studies is straightforward: because identical twins share essentially all their genes while non-identical twins share only about half, a higher concordance in identical twins than in non-identical twins signals a real genetic contribution to that condition.
Heritability is easy to confuse with three neighbouring ideas. Genetic determinism (the false idea that genes alone fix an outcome) is not what heritability claims: incomplete twin concordance disproves it directly. A concordance rate (the raw percentage of twin pairs who share an outcome) is a simple observed number, whereas heritability is a modelled statistic derived by comparing MZ and DZ concordance; the two are related but are not the same figure and should never be quoted interchangeably. And family clustering alone, seen in a family study, cannot distinguish genetic from shared-environment causes the way twin and adoption designs can: a disorder can run in families for entirely environmental reasons, which is exactly why family studies are the weakest of the three designs on their own.
There is no single "heritability test" for an individual patient; instead, risk is built from structured family-history taking, typically a three-generation pedigree (a family tree marking who has been affected), cross-referenced against the published relative-risk and concordance figures above. For polygenic conditions like and bipolar disorder, no single genetic test confirms or excludes risk: the assessment stays statistical and population-based. This differs sharply from single-gene (Mendelian) conditions, such as Huntington's disease, where a specific genetic test can give a near-certain individual answer; conflating the two is a common error.
Families frequently ask what a relative's diagnosis means for their own risk. Genetic counselling for conditions meeting formal diagnostic criteria, such as those in ICD-11 or DSM-5, explains risk using population-level figures like the ones above, not an individual causal percentage. NICE guidance on supporting families of people with psychosis reflects this same population-based framing, emphasising early intervention and support rather than deterministic genetic prediction. Where a relative meets diagnostic criteria for or bipolar disorder, early-intervention services increasingly offer preventive monitoring and low-threshold access to treatment for at-risk young relatives, on the reasoning that early management improves prognosis even though genetic risk alone cannot predict who will become unwell.
The single most tested trap is a question implying that an 80% heritability figure means a person's illness is "80% caused by genes." The correct answer always distinguishes population-level variance from individual-level cause. A close second trap is a stem describing a monozygotic twin pair, one affected by bipolar disorder, and asking for the co-twin's risk: the correct approach uses the published concordance figure, around 40-70%, as an informed estimate, while remembering that concordance is never 100%, because environment and chance keep shaping outcome even in identical genomes.
Heritability measures how much variation ACROSS a population is genetic, never how much of one person's illness is: and incomplete twin concordance is the proof that environment still matters even in the most heritable conditions.