Introduction
If a health condition runs in your family, you may wonder how much of your own health is inherited. You might come across a heritability estimate and expect it to answer that question. It cannot tell you your personal risk, but it can explain something useful about differences within a group.
Heritability is an estimate of the proportion of variation in a trait within a particular population that can be attributed to genetic differences. It applies to differences between people, not to the share of one person’s trait caused by their genes. (medlineplus.gov)
This guide is for anyone trying to understand heritability in biology, psychology or health research. It explains how to read the numbers, how researchers estimate them and why a high figure does not mean a trait is fixed. For personal health concerns, family history is worth discussing with your GP alongside symptoms and other relevant information.
What Does Heritability Measure?
A trait is a characteristic that can be observed or measured. Height, natural hair colour and a score on a psychological assessment are examples. A trait can also describe whether people in a study have a particular condition.
People in a population often differ in these characteristics. Heritability asks how much of that variation between individuals can be statistically attributed to genetic variation in the group studied.
The word “variation” matters. Heritability does not measure how important genes are to the development of a single person. Nor does it identify which genes are involved.
An estimate belongs to a particular trait, population and set of circumstances. It can change when researchers study a different group, measure the trait differently or examine it at another time. MedlinePlus Genetics explains these limits and why the figure should not be interpreted as an individual result.
A Simple Example
Imagine researchers estimate the heritability of height in a particular group as 0.5. This is a hypothetical figure, not an estimate for every population.
It means that, under the study’s method and assumptions, genetic differences account for an estimated 50% of the variation in height within that group.
It does not mean:
- Half of any one person’s height came from genes.
- A child has a 50% chance of being the same height as a parent.
- Height cannot be influenced by circumstances during development.
- The same estimate applies to every population.
Researchers use a statistical measure called variance to describe how spread out the measurements are. Heritability is a proportion of that variance, rather than a proportion of each person’s height.
A heritability estimate describes differences across a group. It is not a personal score.
Heritability, Heredity and Genetics
These words are related, but they do not mean the same thing.
Heredity, or biological inheritance, describes the passing of genetic material from parents to offspring. An inheritance pattern describes how a particular genetic characteristic is passed through a family.
Genetics is the wider study of genes, genetic variation and their effects.
Heritability is a population statistic. It estimates how much variation in a measured trait is associated with genetic variation in the population studied.
There is no exact everyday substitute for heritability. “Genetic contribution to variation” can be a useful explanation, but “inheritance” alone misses the distinction between a family process and a group statistic.
A family resemblance does not always establish its cause. Relatives can share environments and experiences as well as genetic material. A family history can still be useful to a GP, but it should not be treated as a heritability calculation.
What About Rh Factor?
Rh factor is often discussed as an inherited characteristic. Whether someone is described as Rh positive or Rh negative is commonly based on the presence or absence of the D antigen on red blood cells. (nhs.uk)
That is a question about a blood-group characteristic and its inheritance. A heritability estimate would instead ask how much variation in that characteristic within a specified population is attributable to genetic differences.
Knowing that a characteristic is inherited does not, on its own, provide a heritability figure. If you have a question about your blood group and pregnancy, discuss your individual circumstances with your maternity team or GP.
Reading a Heritability Coefficient
You may see heritability called a coefficient, estimate, index or score. In this context, each usually refers to a statistic for a population, not a mark assigned to an individual.
Heritability is generally expressed from 0 to 1, or from 0% to 100%:
- Near 0: Genetic differences account for little of the measured variation under the model used.
- Between 0 and 1: Genetic differences account for some of the measured variation.
- Near 1: Genetic differences account for a large proportion of the measured variation under the model used.
A coefficient of 0.5 is the same as 50%. By contrast, 0.5% is the same as 0.005. The decimal point and percentage sign make a substantial difference.
The theoretical maximum is 1, but there is no useful universal answer to which trait has the highest heritability. The result depends on the population, environment, measurement and method.
High and Low Heritability
A high estimate means genetic differences account for a relatively large share of variation in the studied group. It does not mean every person has the trait for genetic reasons, that the trait is impossible to change or that the environment has no influence.
For example, natural hair colour is often used to illustrate a strongly inherited characteristic. The observed colour of someone’s hair can nevertheless be changed. More generally, an environmental change can influence a highly heritable trait even when genetic differences explain much of the variation in a particular population.
A low estimate means genetic differences account for a relatively small share of the observed variation in that study. It does not prove that genes play no part in developing the trait, or that they would make the same small contribution in another population.
The language someone speaks is a useful example of a characteristic shared within families through learning. Family resemblance alone does not tell us that the particular language spoken was genetically inherited.
Can Heritability Be Negative?
The underlying variance proportions in standard heritability models are not negative. A statistical method can nevertheless produce a negative estimate, particularly when the true contribution is small or the data are uncertain.
That result does not mean genes have a negative share of a person’s trait. It is a reason to look at the method, uncertainty and any constraints the researchers used.
A theoretical value of 0 or 1 also needs context. A narrow-sense value of 0 means no additive genetic variance is attributed to the measured variation in that model. A value of 1 means the model attributes all the variation to its additive genetic component. Neither number describes a person’s future or proves what would happen in a different environment.
Broad-Sense and Narrow-Sense Heritability
Researchers use two main definitions. Both concern variation in a population, but they include different kinds of genetic variation.
A guide for genetic epidemiologists distinguishes between total genetic variation and the additive part of that variation. (academic.oup.com)
Broad-Sense Heritability
Broad-sense heritability is commonly written as H². It is the proportion of observed trait variance attributed to total genetic variance in the model:
H² = V<sub>G</sub> / V<sub>P</sub>
Here, V<sub>G</sub> means genetic variance and V<sub>P</sub> means phenotypic variance, the total observed variation in the trait.
The genetic component can include additive effects, dominance effects and interactions between genetic variants. A high broad-sense estimate indicates a large total genetic contribution to variation under that model. It does not tell us how much of that contribution is additive.
Narrow-Sense Heritability
Narrow-sense heritability is commonly written as h². It focuses on additive genetic variance, written as V<sub>A</sub>:
h² = V<sub>A</sub> / V<sub>P</sub>
Additive effects are the average contributions of genetic variants considered individually in the population. Narrow-sense heritability is sometimes called additive heritability.
A high narrow-sense value means additive genetic differences account for a large share of measured variation under the model. A low value means they account for a smaller share. It does not exclude other genetic effects.
Narrow-sense heritability is particularly useful in plant and animal breeding because additive effects help researchers estimate how a trait might respond to selection across generations. That use depends on assumptions about the population and conditions. It is not a formula for predicting an individual child’s characteristics.
The Stanford Encyclopedia of Philosophy’s discussion of heritability explains why even these formulas simplify the relationship between genes, environments and measured traits. (plato.stanford.edu)
How Genes and Environment Work Together
Heritability is sometimes presented as a simple contest between genes and environment. In practice, the relationship is more complicated.
Genetic differences can affect how people respond to an environment. Genetic and environmental influences can also be correlated. Families, for example, may share aspects of both their genetic background and their surroundings.
An environment can influence a trait even when everyone in a study experiences it in a similar way. Heritability concerns differences within the group. A factor that is much the same for everyone may not explain much of the variation between them, even if it matters to the development of the trait.
Imagine two groups of plants with genetic variation. In one group, growing conditions are fairly similar. In the other, conditions vary considerably. If the differences in conditions contribute to differences in growth, the proportion of variation attributed to genetics may be lower in the second group.
That does not require the plants’ genes to have changed. The proportion can change because the amount of variation associated with the environment has changed.
Heritability also cannot be calculated meaningfully when there is no observed variation in the trait to divide up. If every measured value is identical, the total variance is zero.
Why Estimates Change
An estimate can differ because of changes in:
- The population or age group studied.
- The range of genetic differences in that population.
- Environmental conditions and how much they vary.
- How the trait is defined and measured.
- The statistical method and its assumptions.
Measurement error matters too. A trait that is difficult to measure consistently may produce a different estimate from a more precise measure of the same general characteristic.
A high heritability estimate within one group also does not establish why average results differ between two groups. Those are separate questions requiring different evidence.
How Researchers Estimate Heritability
Researchers cannot read a heritability coefficient directly from one person’s DNA or blood test. They estimate it using data from groups of people and a statistical model.
Different approaches capture different kinds of information. Each has assumptions, so two estimates for the same trait may not be directly comparable.
Twin and Family Studies
Twin studies compare the similarity of identical twins with the similarity of non-identical, or fraternal, twins. Identical twins have very similar DNA. Fraternal twins share, on average, about half of the genetic variation that differs between people, as other full siblings do.
If identical twins are more alike for a measured trait than fraternal twins, that pattern can suggest a contribution from genetic differences. Researchers use the extent of resemblance to estimate heritability.
Family studies can also compare parents and children, siblings or other relatives. Shared environments matter in these studies because relatives may resemble one another for reasons that are not purely genetic.
A simple twin calculation sometimes used as an illustration is twice the difference between the correlations for identical and fraternal twin pairs. If those correlations were hypothetically 0.8 and 0.5, the calculation would give 2 × (0.8 − 0.5) = 0.6.
That is a simplified illustration, not a reliable calculation to apply to any pair of published figures without checking the study. Twin methods rely on assumptions, including how similar the relevant environments are for the two types of twins. The genetic epidemiology guide describes these assumptions and the possible sources of bias. (academic.oup.com)
DNA-Based Studies
Researchers can also compare measured genetic variation with differences in a trait across many participants. A genome-wide association study looks for genetic variants associated with a trait across the genome.
Some methods estimate SNP heritability. A SNP, pronounced “snip”, is a variation at a single position in DNA. SNP heritability estimates the contribution captured by a particular set of measured or represented variants, usually focusing on additive effects.
It is not necessarily the same as an estimate from a family or twin study. The studies may capture different genetic effects and rely on different assumptions. Genetic similarity can also be related to environmental similarity, which can complicate interpretation.
Heritability in Psychology and Mental Health
In psychology, heritability has the same population-level meaning as in biology. Researchers may study variation in scores on a cognitive assessment, personality measures or the presence of a mental health condition.
Twin studies have been used to investigate traits and conditions including intelligence measures, autism, depression and schizophrenia. The appropriate estimate depends on the particular study, its participants and how the trait or condition was defined.
For a condition recorded as present or absent, researchers may use a statistical model that assumes an underlying range of susceptibility, sometimes called a liability scale. That extra assumption matters when comparing estimates.
None of these figures tells an individual whether they will develop a condition, what caused their experience or which support will help them. Psychological traits and health conditions also differ in how they are measured, so their estimates should not be treated as interchangeable.
Complex Traits and Missing Heritability
Many traits are polygenic: numerous genetic variants can contribute, rather than one variant providing a complete explanation. These are often called complex traits because genetic and environmental influences can both matter.
A heritability estimate may suggest that genetic differences contribute to variation without identifying the variants involved. Finding the relevant variants is a separate research task.
What Does Missing Heritability Mean?
“Missing heritability” describes a gap that can arise when one approach, such as a family study, suggests a genetic contribution that is not fully captured by the variants or methods used in another analysis.
The word “missing” does not establish one explanation. Some variants may be uncommon or poorly captured by the data. Studies can also differ in their populations, measurements and assumptions. The guide to estimation methods explains why both genomic and family-based approaches have limitations.
Is Missing Heritability Explained by Epigenetics?
Epigenetics concerns changes that can affect how genes are used without changing the underlying DNA sequence. One example is the addition of chemical marks to DNA.
Epigenetics is a distinct area of biology, not another name for heritability. Although researchers study epigenetic processes in relation to traits and inheritance, an unexplained gap between heritability estimates cannot simply be assumed to be caused by epigenetics.
What Heritability Can and Cannot Tell You About Health
Heritability can help researchers decide which questions to investigate. It offers a starting point for studying genetic contributions to complex traits and can help researchers compare methods or explore how a trait varies across populations.
It cannot, by itself:
- Identify the genes or environmental influences involved.
- Diagnose a condition or explain one person’s symptoms.
- Give an individual’s chance of developing a condition.
- Say whether a trait can be changed.
- Establish why two populations have different average results.
- Decide which test or care is appropriate for an individual.
A family history is more directly relevant to a conversation about your own health than a general heritability figure. Even then, the details matter: which relatives were affected, what condition they had and what else is known about your health.
If you have persistent, worsening or concerning symptoms, speak to your GP rather than assuming they are simply inherited. It can help to note when symptoms started, any changes over time, relevant family history and changes to medicines or supplements.
A blood test, where relevant, can provide a snapshot to support a professional discussion. It does not calculate your personal heritability or replace a clinical assessment. For more specific inherited-risk questions, a GP can discuss whether further assessment is appropriate.
When you next read a heritability figure, check the trait, population and method before applying any meaning to the number.
FAQ
Does 50% Heritability Mean a Trait Is Half Genetic?
No. It means genetic differences are estimated to account for 50% of the variation in that trait within the population studied. It does not divide one person’s trait into a genetic half and an environmental half.
Can a Highly Heritable Trait Still Be Influenced by the Environment?
Yes. A high estimate describes variation within a particular group and set of conditions. It does not mean an environmental change cannot affect the trait.
Is Heritability the Same as a Genetic Risk Score?
No. Heritability describes variation across a population. A genetic risk score is an estimate intended to describe genetic risk for an individual, using selected genetic information. The two answer different questions and should not be substituted for one another.
Can a Blood Test Tell Me My Heritability?
No. Heritability is estimated from information about a population, not from one person’s blood result. If you are concerned about symptoms or a family history, discuss them with your GP.
