IT IS peak summer, and you’re dying of inoperable cancer. It’s July, and the days are long. The doctors have told you that you have “less than six months” to live.

All you can think about, as you stare out of the window into the hospital car park, is that if you die just before Christmas, the grief of your wife and three children will erase any memories of the joy the holiday season once held, and prevent it from ever again being a time of celebration.

Legalised assisted dying has just come into force. You decide that it’s better to die in the middle of the year and not ruin that holiday season for them, rather than to linger on and die in November or early December. You take the plunge and you’re dead within a fortnight.

Except that isn’t what happened. Instead, your life ran its natural course and you lived until February. You spent one last joyful Christmas and holiday season with your family. They cherished that memory forever. You were so grateful for that last holiday with them. Had assisted dying been legal, you would have missed this. Your family would have missed this. The doctors underestimated your prognosis, you see.

Spending one last holiday season together and consequently not ruining that time forevermore due to its association with grief is a common concern for the dying.

Indeed, the above narrative was based on the concerns of a real terminally ill patient whom I came across in my practice as an intensive care senior resident doctor.

If assisted dying legislation is passed, terminally ill patients will make life and death decisions up and down this country, based on potentially false information.

The assisted dying legislation being debated at Westminster by MPs requires patients to have mental capacity (though experts have expressed concerns about the suitability of the Mental Capacity Act in this regard), a clear, settled and informed wish to die (though there are fears about the difficuty of detecting coercion and the danger of vulnerable people choosing to die because they feel a burden,) a prognosis that the patient has six months or less to live, and the ability to self-administer life-ending medication (though the Bill isn’t crystal clear on what this means).

The process includes two declarations on the patient’s part, two independent doctor assessments a week apart, and a panel review. Despite over 400 pieces of written evidence, oral evidence sessions and more than 500 amendments tabled at committee stage, a fundamental problem remains: prognosis is still too unreliable to safely underpin such an irreversible choice.

Evidence submitted to the Westminster Bill Committee by organisations with considerable expertise and familiarity with these issues also makes for difficult reading for Bill supporters.

The Association for Palliative Medicine stresses that “prognostication for the last six months of life is notoriously difficult [and] highly subjective” and that it is both “problematic and yet fundamental to its safety” - not a reassuring combination.

Meanwhile, Marie Curie raises the troubling prospect of creating disparities in care, noting that “the six-month criteria could also create inequalities as it can be more challenging to prognosticate non-cancer conditions.” This creates a zone of uncertainty where eligibility decisions hinge more on guesswork than medical certainty.

The evidence on prognostication is troubling for this legislation. A 2023 UK study of nearly 100,000 patients found doctors were 74 per cent accurate at predicting death within 14 days and 83 per cent for survival beyond a year, but only 32 per cent accurate for “weeks” or “months” — the exact window relevant to assisted dying eligibility.

An Australian study showed oncologists underestimated survival by threefold in 11 per cent of advanced cancer cases, meaning patients could lose many extra months. Even with additional training, a 2024 German study found predictions for survival beyond six months remained poor, at just 25.9 per cent accuracy, despite prognostic training. Since “six months to live” is the legal threshold, relying on such uncertain estimates for life-and-death decisions should give serious pause.

Perfection in prognostication is a lofty goal and perhaps unattainable, but 32 per cent and 25.9 per cent accuracy rates for intermediate prognoses remain alarmingly low.

The claim that dual doctor assessments mitigate errors ignores ground realities: if a metastatic cancer patient’s prognosis is assessed first by an oncologist, a second ICU doctor - untrained in tumour staging, treatment response, or metastasis patterns - cannot independently validate the six-month estimate.

Without requiring specialist expertise, the second doctor risks rubber-stamping complex prognoses they lack the skills to challenge. In an overstretched NHS, time-pressed clinicians may accept initial assessments uncritically, rendering the “independent review” a procedural fiction that fails to address prognostic inaccuracy.

Putting aside the debate about whether people should have the so-called “right to die,” and whether the NHS indeed should provide it, the first question is whether the legislation is sound by its own standards.

“Six months to live” sounds authoritative and factual, but it is not. Except in the case of those who are likely to die within days to weeks, prognostication in the “months” territory is ultimately a clinical guesstimate, especially for “independent” second doctors who aren’t trained in the disease-level specifics to know better.

It is actually even more complicated than this, as the Marie Curie Palliative Care Research Department at UCL noted in their submission to the Bill committee. Their research on the accuracy of predictions of death or survival showed that “over half of the GPs were not consistent in how they applied their predictions to different patients.” Assisted dying legislation tries to create a strict, neat system, but it is one that even the very best trained medical professionals aren’t able to implement reliably. It seeks to bureaucratise and reduce the complexity of life into a tick-box death pathway. It simply can’t be done - or at least, it can’t be done safely, and there lies the rub: what risk are we willing to take as a society?

Is it truly acceptable for the gravest of human choices, life or death, to hinge on little more than a clinical guess?

Should we entrust such irreversible decisions to the shaky ground of unreliable predictions, knowing that a single misjudgment could steal precious months, memories, or goodbyes?

If we cannot answer these questions with unwavering confidence, then we must confront the uncomfortable truth: this Bill, built on prognostic uncertainty, is not fit to bear the weight of finality it demands.

Dr Tahir Nasser is an Intensive Care Senior Resident Doctor in Surrey.