The Big Question: Should doctors try to save extremely premature babies?

Why is the question being asked now?

Before the 1960s, most babies born prematurely died. Since the 1980s, the proportion of babies born with very low birth weights, usually premature, has more than doubled. Advances in neo-natal medicine have increased the options for doctors and parents - but, because of the risks of death or disability, have made it increasingly difficult for them to know the best thing to do. There has been an increase in disputes between doctors and parents which have ended in the courts and led to calls for more guidance.

Yesterday, the Nuffield Council on Bioethics, an independent body which examines ethical issues raised by developments in medicine, published guidelines on when to provide intensive care to extremely premature babies, defined as those born more than three months early (below 26 weeks gestation). "Doctors have gained the capacity to prolong the life of newborn babies and the question is whether they should always do so," said Professor Margot Brazier, chair of the Council.

What does the report say?

It is very rare for babies born before 22 weeks to survive. Even at 22 to 23 weeks, only 1 per cent do so according to the 1995 EPICure study, the best British research, led by the University of Nottingham. Those that do so at this gestation have a very high risk of being severely disabled. By contrast, almost half (44 per cent) of those born at over 25 weeks survive and most do so without severe or moderate disability. On this basis, the Council says babies born before 22 weeks should not be given intensive care and should be allowed to die while those born over 25 weeks should be given all necessary care to enable them to survive.

What about the grey area between 22 and 25 weeks?

The Council proposes a sliding scale, with a presumption that intensive care should not be given to babies between 22 and 23 weeks because of their very low chance of survival and very high risk of severe disability. Intensive care should normally be given for babies between 24 and 25 weeks. For those born in the middle of the period, between 23 and 24 weeks, it says the decision should be left to the parents. Those who objected strongly to the guidelines would be able to override them by agreement with their doctors. But the justification would have to be commensurate with the risk involved - in the case of saving a baby - or with the poor prognosis in the case of allowing a baby to die. One scenario presented yesterday was of a woman of 45 who had no children, a last chance at pregnancy and wanted to try to save her baby however slim the chance. "I would consider what I could do to help her," Andrew Whitelaw, professor of neonatal medicine at Bristol University and a member of the working party, said.

How many babies are involved?

About 4,000 babies a year are born between 22 and 26 weeks. Most are stillborn but 1,400 show signs of life at birth. The number has been increasing, driven in part by the increase in fertility treatment which has increased the proportion of twin and triplet births, which are more likely to be premature. Practice among the country's 250 premature baby units varies, with some striving to save babies at 23 and even 22 weeks, while others more reluctant to resuscitate even at 24 weeks. One aim of the report is to increase consistency across the NHS. The risk of disability is clearly greatest for those with the least gestation time. According to the EPICure study between 23 and 24 weeks, two thirds of the babies who survived had moderate, or severe, disabilities. By 25 to 26 weeks, two thirds had no, or mild, disabilities.

Has care not improved?

Care has improved since 1995 (the date of the EPICure study). But survival rates are still low at 16-20 per cent for babies that are born at 23 weeks, according to three recent studies (in Norway, and Trent and Bristol in England).

It is too early to tell what the disability rates will be but Professor Whitelaw said there was no reason to think they would be significantly lower than the two thirds recorded in the 10-year-old EPICure study.

"If the policy is to give intensive care to all babies at 23 weeks - most will die and we will have subjected them to painful, intrusive treatment. This is not minimal interference - it is prolonging the process of dying in the majority, who will eventually die," Professor Whitelaw said.

Which babies should be allowed to die?

When the struggle to save them results in "intolerable" suffering that outweighs the baby's interest in continuing to live. The Council said it was extremely difficult to find the appropriate language to describe the point at which the duty to prolong life is superseded by the duty to provide palliative care.

It adopted the concept of intolerability, recognising that "reasonable people" would disagree about what it meant and when it had been reached in any situation. Each case therefore had to be individually assessed. However, although the baby's interests in living or dying, or in "avoiding an intolerable life", were most important, it said the interests of parents (or carers) must be given "some" weight.

What about euthanasia?

The Council rules it out saying doctors have a professional obligation to preserve life where they can. To end life deliberately would alter the ethos of medicine, threaten the relationship between doctor and patient and suggest the infant had a different moral status from older children and adults, for whom euthanasia is banned.

Are costs a factor?

Research suggests annual health and education costs for children born extremely prematurely are two-to-three times higher at age six than those born at full term - £9,541 against £3,883 in the EPICure study. In practice the worst affected cost tens of thousands of pounds every year in extra care, while the majority who are less affected incur only modest costs above those of routine family life.

For the NHS, care given to one patient means care denied to another. Doctors should be "aware of, but not driven by," the cost implications of their decisions, the Council says. But it wants the Government to provide more support for families of disabled children. "On grounds of consistency the state should not think it permissible to enable many of the [extremely premature] babies to survive but be excused the discharge of its resultant obligation to support their care."

Is it ever justified to withdraw life-saving treatment from a baby?

Yes...

* Babies of less than 25 weeks gestation have a high risk of dying and a high risk of disability if they survive

* Putting an extremely premature baby into intensive care may merely prolong its dying and subject it to intolerable suffering

* The costs of caring for extremely premature babies who may not survive puts pressure on scarce NHS resources

No...

* There is always a chance that an extremely premature baby born alive will survive - and may go on to lead a full life

* Survival rates for extremely premature babies have almost doubled in the past decade

* Doctors and parents should not try to prevent the survival of a baby just because he or she may grow up with severe disabilities