Prognostic criteria are inaccurate for 6 month terminal prognosis.

Alex Schadenberg
Executive Director, Euthanasia Prevention Coalition

An editorial that was written by Dr Scott Murray and Dr Simon Etkind that was published in the British Medical Journal on March 12, 2025 explains that prognostic criteria in providing a six month terminal prognosis are inaccurate.

The article is a response to the assisted dying bill in the UK that is sponsored by Kim Leadbeater (MP). The Leadbeater bill states that a person qualifies for assisted dying if they are:

“reasonably expected to die within six months,”

The authors write:

However, the challenges of prognostication in relation to eligibility for assisted dying remain underexplored. Prognostic eligibility criteria are limited by the fact that prognosis is inherently uncertain and there are no valid tools, tests, or clinical examinations that can reliably and safely identify that a person is expected to die within six months.

The authors are concerned with the language of the bill. They write:

Clarity about what is meant by “reasonably be expected to die within six months” is vital so that clinicians can understand the intended beneficiaries of the bill. It could mean that on the balance of probabilities the patient is expected to die within six months, but the word “reasonably” potentially incorporates a wide interpretation of likely prognosis.

I suggest that the use of the term "reasonably be expected to die" cannot be defined and suggests that the bill does not intend to enforce a 6 month prognosis rather the 6 month prognosis is included to suggest that the bill has "rigorous" safeguards.

The authors explain that only 37% of predictions based on a 12 month prognosis are accurate:

Prognostic criteria in legislation on assisted dying imply that accurate prognostication in advanced illness is possible. But clinicians’ predictions are often inaccurate. Looking specifically at the “surprise question” (Would I be surprised if this patient died in the next 12 months?), a meta-analysis found a positive predictive value of only 37%. Even in cancer, where we would expect a more predictable illness trajectory, the utility of prognostic tools was low over six months.

The authors conclude by stating:

Since uncertainty is inherent in advanced illness, we must embrace it as a trigger for person centred care and planning ahead. Many of the psychological, social, and existential issues raised by uncertainty at the end of life can be addressed through skilled communication and holistic support: core components of timely palliative care. Moreover, if people understand more about the process and timescales of dying, and the potential of palliative care to complement curative care, they may feel in greater control of their situation and empowered to cope with its demands, whether they have access to assisted dying or not.

Irreducible prognostic uncertainty makes it difficult to delineate the scope of assisted dying legislation that uses prognosis based criteria. Further evidence is needed to determine whether prognosis based judgments of eligibility can be consistently and safely applied.

The authors state that based on the uncertainty of a 6 month prognosis that requests for assisted dying should be accompanied by skilled communication, holistic support and palliative care.

I agree that it is impossible to accurately predict that a person has 6 months to live but I suggest that the purpose of the inclusion of a 6 month prognosis is only to claim that the bill has rigorous safeguards. If the bill passes, the 6 month prognosis is likely to be removed through future legislation.