There are four key problems with the definition of “terminal illness” in the Bill:
- (i) that term is necessarily vague and a wide range of conditions could be brought within it,
- (ii) the level of confidence for the 6 months prognosis is necessarily low and will mean that people will be eligible even if they could have lived for much longer,
- (iii) patients could refuse treatment to bring themselves within 6 months of death and therefore become eligible (this has happened in Oregon),
- (iv) despite the attempt of the Bill’s promoters, mental health and disability remain within scope.
Some of those points might seem like nitpicky drafting points but it is important to remember that this Bill has been drafted by the recently retired most senior legislative drafter for the Government.1 Therefore, it can safely be assumed that the legislation she wrote reflects the policy and legal instructions she was given. This means that the flaws pointed out below are design flaws rather than simply drafting flaws. MPs should therefore not assume that the proponents of the legislation would agree to amendments to ‘tweak’ them as these reflect deliberate policy choices. In the section below we will endeavour to explain why we believe the terms were drafted that way.
What does ‘terminally ill’ mean?
This is defined in clause 2 of the Bill
(1) For the purposes of this Act, a person is terminally ill if—
(a) the person has an inevitably progressive illness, disease or medical condition which cannot be reversed by treatment, and
(b) the person’s death in consequence of that illness, disease or medical condition can reasonably be expected within 6 months.
(2) For the purposes of subsection (1), treatment which only relieves the symptoms of an inevitably progressive illness, disease or medical condition temporarily is not to be regarded as treatment which can reverse that illness, disease or condition.
(3) For the avoidance of doubt, a person is not to be considered to be terminally ill by reason only of the person having one or both of— (a) a mental disorder, within the meaning of the Mental Health Act 1983; (b) a disability, within the meaning of section 6 of the Equality Act 2010.
Taking a step back, the colloquial understanding of ‘terminal illness’ is necessarily context dependent. Conditions which in the past might have been considered terminal and not so considered anymore, for example diabetes prior to the discovery of insulin. Conversely medicine may come to classify conditions as terminal which most people would not regard as such, for example recent moves to talk about ‘terminal anorexia’.
In this context coming up with a robust definition was always going to be difficult. Nonetheless, the definition adopted (“inevitably progressive” and “cannot be reversed by treatment”) is too wide and would include conditions that colloquially would not be understood as terminal:
- diabetes for example cannot be reserved by treatment (whilst the treatments given relieve the symptoms and keep the patient alive they do not cure it): a great deal of weight would then be placed on what the word “temporarily” means in subclause (2).
- The position is the same with HIV. Antiretroviral drugs do not “cure” HIV but control its effects. Given that HIV can become drug-resistant, does that mean that ART drugs only work “temporarily” even though they can work for many years?
- As a recent article in a scientific journal argued, certain forms of severe and enduring anorexia cannot be cured and will lead to death. So, the authors argued, it should be classified as a terminal illness known as ‘terminal anorexia’.2
In the other direction, even Stage 4 cancer can be treated and the treatments do not always fail: does this mean that terminal cancer patients will not, generally, have an “inevitably progressive” disease?
So the first limb of the definition is likely both over and under inclusive.
Is the exclusion of mental illness and disability effective?
Subclause (3) does not exclude mental illness and disability from the ambit of the definition of terminal illness. Rather it says that the test would not be met “by reason only” of mental illness and disability. But if the patient had a physical condition as a result of a mental illness, they could qualify. So, for example, even if anorexia itself does not mean that a patient falls within the meaning of “terminally ill”, the anorexia combined with being in a state of physical starvation could mean that they qualified.
How sure must a doctor be about the 6 months prognosis?
The Bill says that a “person’s death in consequence of that illness, disease or medical condition can reasonably be expected within 6 months.” That phrase already appears in the context of special rules for benefits for terminally ill people and a UK court has already considered it to mean ‘would you be surprised if the patient died within 6 months?’ (Department for Communities v Cox [2021] NICA 46 at [42]-[47]) When Parliament uses a phrase which has already been interpreted by a Court it is presumed to want to adopt this meaning.
This surprise question is a very low bar but it is already widely used in medicine and the evidence is that doctors get it wrong 50% of the time.3 This means that by design the Bill will implement a safeguard which will only work half the time. MPs should consider whether that is an error rate they are happy with?
6 Months cut-off does not require treatment
Whilst the first limb of the definition of terminal illness has a requirement of treatment (albeit one qualified by subclause 2: treatment which only relieves the symptoms temporarily does not count) there is no such reference in the requirement of death within 6 months.
This means that the question can be answered on the basis that the patient is not receiving treatment. In other words the test is not “even with treatment, would death be reasonably expected within 6 months”.
This definition is the same as was used in Oregon and it has changed its practice to mean that patients who have illnesses which can be cured could qualify (by refusing further treatment).4
This would mean that many people who are not, in any ordinary sense, likely to die within six months, could bring themselves within the provisions of this legislation if they were suicidal, by refusing treatment.
- https://www.theguardian.com/society/2024/nov/20/diane-abbott-edward-leigh-urge-commons-reject-assisted-dying-bill ↩︎
- Gaudiani JL, Bogetz A, Yager J. Terminal anorexia nervosa: three cases and proposed clinical characteristics. J Eat Disord. 2022;10(1):23. https://pubmed.ncbi.nlm.nih.gov/35168671/ ↩︎
- Orlovic, M., Droney, J., Vickerstaff, V. et al. Accuracy of clinical predictions of prognosis at the end-of-life: evidence from routinely collected data in urgent care records. BMC Palliat Care 22, 51 (2023). https://doi.org/10.1186/s12904-023-01155-y ↩︎
- See communications from the Oregon Health Department to a Swedish researcher: https://drive.google.com/file/d/1xOZfLFrvuQcazZfFudEncpzp2b18NrUo/view ↩︎