Two Definitions of 'Family History of Cancer'

When you ask what is considered a family history of cancer, you are actually asking two questions that people treat as one. There is a clinical definition — the one the NHS uses to decide whether you need referral, earlier screening or genetic testing — and there is an underwriting definition, the one built into the family-history question on a life insurance form. They are not the same, they are set by different people for different purposes, and you can comfortably meet one while falling outside the other.

The clinical definition of a family history of cancer, shaped by NICE, turns on how many relatives were affected, how closely related they are to you, how young they were at diagnosis, and whether the cancers cluster on one side of the family. The insurance definition is usually blunter: it asks about a parent or sibling diagnosed before a stated age. Once you see the two side by side, the confusion that surrounds this topic mostly dissolves.

This page sets the two definitions against each other and shows exactly where they disagree — including the two situations where you meet one and not the other, and why that is not a contradiction to worry about.

The short answer: two definitions, not one

The word "family" makes this feel like it should have a single, obvious meaning. It does not, because the two systems that use it want different things from it. A clinician is trying to work out whether an inherited cancer risk runs in your family and whether your care should change as a result. An underwriter is trying to price a specific product for a specific term, using a small number of questions that can be answered quickly and consistently. Those aims pull the definition in different directions.

The result is that "a family history of cancer" is really two overlapping categories. Some people sit in both, some in neither, and a fair number sit in one but not the other. Knowing which you are in — and which of the two actually matters for what you are trying to do — saves a great deal of unnecessary worry.

The clinical definition: what the NHS acts on

The clinical definition is the one that changes what happens to you medically, and it is the more sophisticated of the two. It does not simply ask whether cancer has appeared in your family — because for most families it has — but examines the shape of it. The features that push a pattern towards significance are consistent across the guidance set by NICE, which defines the clinical thresholds the NHS uses.

Four things do most of the work. The number of affected relatives: more raises concern. Their closeness to you: a parent, brother or sister — a first-degree relative — counts for more than an aunt or grandparent, and a distant cousin for little. Their age at diagnosis: cancer that struck young is more suggestive of an inherited cause than the same cancer in old age. And clustering: several cases of the same or related cancers on one side of the family is a stronger signal than scattered, unrelated cancers across both sides. The exact thresholds differ by cancer type, which is why there is no single number to quote, and why a drawn family tree — rather than a headline count — is what a service actually assesses.

The point of meeting this definition is not a label. It is access to something concrete: reassurance if you fall short of it, or earlier surveillance, a genetics referral, or testing if you cross it.

The insurance definition: what the form asks

The underwriting definition is narrower and more mechanical, because it has to be answerable in a few tick-boxes. A typical life insurance application asks whether a first-degree relative — a parent, brother or sister — has been diagnosed with certain conditions, including some cancers, before a stated age. That age is commonly set somewhere around 60, though it varies by insurer and by condition, and some forms ask about the number of affected first-degree relatives as well.

Several things follow from that design. The form generally does not care about grandparents, aunts, uncles or cousins, however striking the pattern among them. It usually does not ask you to weigh clustering or to distinguish one side of the family from the other. And it applies an age cut-off that is a blunt line rather than a clinical judgement: a diagnosis a year before the stated age is inside the question, a year after it is outside, even though the biology does not change at a birthday. None of this is the insurer being careless — it is a deliberately simple instrument for a different job — but it means the definition captures a different group of people from the clinical one.

What happens when you answer "yes" is worth knowing too, because it is rarely as dramatic as people fear. A single affected first-degree relative is a common answer and is often absorbed with little or no effect on the terms offered. A stronger pattern — two affected relatives, or a diagnosis at a notably young age — is weighted more heavily, and here the important point is that insurers do not read the same answer the same way. One company may barely adjust for a history another treats more cautiously, which is precisely why the same honest form can produce different decisions, and why comparing across the market matters more than fixating on any single insurer's verdict.

Where the two definitions disagree

Because the two are built differently, they routinely diverge. The table below sets them side by side; the rows are where the gaps open up.

Two worked examples show the divergence in each direction. A single parent diagnosed with a common cancer at 55 may sit squarely inside an insurer's question — a first-degree relative, before the stated age — while falling well below the clinical threshold, because one affected relative with no clustering and an unremarkable age does not suggest an inherited syndrome. That person answers "yes" on the form and yet needs no genetics referral at all.

Now reverse it. Someone with several relatives affected by related cancers on one side of the family — a grandmother, an aunt and a cousin, say, with breast and ovarian cancer — can clearly meet a genetics referral threshold, because the clustering and pattern are exactly what the clinical definition looks for. Yet a form that asks only about parents and siblings may capture none of it, and that person could honestly answer "no" to the narrow question while carrying a family history the NHS would take very seriously. Neither answer is dishonest; they are answers to different questions.

Which definition should you worry about

Given the two can pull apart, it is worth being clear about which deserves your attention, and the answer is the clinical one. It is the definition that changes your care — whether you are assessed, screened earlier, offered surveillance or tested. The underwriting definition only affects the terms of a policy you may or may not take out. One is about your health; the other is about a product.

So the order of priority is simple. Take the family tree to your GP first and let the clinical definition do its work, because that is where a real inherited risk gets found and acted on. Treat the insurance question as secondary — important to answer accurately, but not a verdict on your health. Ranking the two correctly, rather than letting the form's blunt question stand in for a clinical judgement, is the single most useful thing to take from all this.

It also helps to keep the underlying odds in view, because a "yes" on a form is not a diagnosis and a family cluster is not a fate. People routinely overread both family patterns and mortality statistics; the same misreading that makes a pedigree look like destiny is the one we untangle in why a period life table will not tell you whether you will reach 85. If your worry runs wider than cancer, it is worth seeing what the genuinely leading risks are at a given age in what actually kills men in their fifties in the UK, where the biggest causes turn out to be the ones you can most influence.

What is in neither definition: the genetic result itself

There is a third thing people fold into "family history" that belongs to neither definition: the result of a predictive genetic test. Whether it came from the NHS or a shop-bought kit, a predictive result — taken when you have no symptoms, to estimate a future risk — is treated separately from a family history, and on the insurance side it is protected. Under the industry Code on Genetic Testing and Insurance, a UK life insurer cannot ask for or use a predictive genetic result. It can still ask the ordinary family-history question, because your relatives' actual diagnoses are not your gene test.

That distinction is worth getting right, and we explain it in full in our piece on whether a 23andMe result affects UK life insurance. The short version is that the family-history question asks what happened to your family; it does not ask what a laboratory found in your DNA. And where you genuinely cannot answer the family-history question — because you were adopted, or the information was never passed down — "not known" is an honest and accepted answer, as we set out for anyone adopted with no family medical history to give.

If, having sorted the clinical question, you decide cover is overdue, answer the family-history question accurately and let an FCA-authorised broker compare life insurance quotes, since insurers weigh the same honest answers differently from one another.

Scotland and Northern Ireland

The clinical definition is broadly consistent across the UK, because the underlying guidance and the biology are the same, but the services that apply it are run separately by NHS Scotland, Health and Social Care in Northern Ireland, and NHS England and Wales. Referral routes and waiting times differ, so ask your GP how the pathway works where you live. The insurance definition does not vary by nation at all: the family-history question on an application, and the Code that protects a predictive genetic result, apply across the whole United Kingdom.

Whichever nation you are in, the sequence holds: sort the clinical question first, because it can change your care, and treat the insurance definition as the secondary matter it is. Our guide to family history, inherited risk and the odds pulls the whole subject together if you want to see how the pieces fit.

Frequently asked questions

What counts as a family history of cancer?

There are two answers. Clinically, NICE thresholds turn on how many relatives were affected, how closely they are related, their ages at diagnosis and whether the cancers cluster on one side. On an insurance form it usually means a parent or sibling diagnosed before a stated age. The two capture different groups of people, so it is worth knowing which one you mean.

Can I meet one definition and not the other?

Easily. A single parent diagnosed at 55 may sit inside an insurer's question yet below the NHS referral threshold, while several affected relatives on one side can trigger a genetics referral and still fall outside a form that asks only about parents and siblings. Answering "yes" to one and "no" to the other is not a contradiction; they are different questions.

Which one should I worry about?

The clinical one, because it changes your care. The underwriting definition affects the terms of a policy; the NICE one decides whether you are assessed, screened earlier or offered testing. Take the family tree to your GP first, and treat the insurance question as secondary.