One father spent much of last week trying to find a paediatrician prepared to act as an expert witness on his behalf. The man is concerned that his ex-wife may be harming their child by dosing it with her epilepsy medication.

So far, his hunt has been unsuccessful. Around the country, the furore triggered by Professor Sir Roy Meadow is having a real impact on the standard of care and protection of children.

'What we have witnessed is state-sponsored child abduction, and it's been going on for years,' says campaigner Penny Mellor. In turn, many paediatricians say that they face a barrage of malicious complaints; damned if they fail to intervene, as in the case of Victoria Climbié, and damned if they do. One in six of the country's 2,000 paediatricians has had a serious complaint made against them concerning child protection issues in the past five years.

Of course, professionals should be accountable when families risk forfeiting a child. 'But I have seen colleagues absolutely destroyed,' says paediatrician Neela Shabdi.

Every NHS trust is supposed to have a designated doctor responsible for child protection, training and standards and liaising with other professions. One in three posts in the country is now vacant. 'It's tempting to walk away,' says one consultant. 'Who's going to object? Certainly not the child, who's too young to have a voice.'

'The pendulum is swinging,' said one forensic pathologist yesterday. 'The question is where it ends up and what that means in terms of the protection of children.'

Last week saw one of the most destructive weeks in the history of child protection in Britain. A stream of parents gave their heart-wrenching accounts of how their children were taken into care, allegedly on the flimsiest of evidence: evidence that cannot be tested because of confidentiality.

The highest court in the land has acknowledged that the cases of Trupti Patel, Angela Cannings and Sally Clark may be only the first of a string of miscarriages of justice.

Last Monday, Lord Goldsmith, the Attorney General, announced that 258 convictions of parents over 10 years for infanticide, murder and manslaughter were to be reviewed. Highest priority was to be given to 54 mothers and fathers still in jail. In addition, a further 15 pending cases could be halted. On Friday, he backtracked and said that that the number under review might be much lower.

In the Cannings appeal court ruling, Lord Justice Judge said: 'If the outcome of the trial depends exclusively on the serious disagreement between... reputable experts, it will be unwise and therefore unsafe to proceed.'

He went on to acknowledge that a small number of killers might now escape justice, but that was preferable to the greater wrong of jailing the innocent.

The ruling, according to many social care professionals, shows an alarming lack of understanding of modern child protection procedures. 'If society doesn't want to prosecute child abusers, that should be determined through Parliament, not by appeal court judgments and lynching a few professionals,' said one pathologist last week.

'What the ruling appears to mean,' said one director of Social Services, 'is that we should look away from the death of babies, but we still have an obligation to monitor for the safety of the children who come after.'

In the main, last week the media condemned the child protection system - in spite of the fact that Britain has one of the highest rates of homicide and unexplained deaths of children in Western Europe. Underlying are the traditional tensions: the sanctity of the family versus the invasiveness of the state. Little has been heard about children's rights. However, once the furore abates there is the promise of opening up a public debate that could lead to constructive reforms.

Mothers do kill their babies. In New South Wales last week, for instance, Kathleen Folbigg, 36, was found guilty of killing her four babies, aged between 19 days and 19 months, over an eight-year period. Two of the four were recorded as cot deaths - sudden infant death syndrome (Sids). Folbigg had detailed what she had done in her diary.

In the cases of Clark, Patel and Cannings, the Crown called consultant paediatrician Professor Sir Roy Meadow, now retired, as one of a number of expert witnesses. Since the 1970s, Meadow had described Munchausen Syndrome by Proxy - a condition in which a mother induces symptoms in her child to bring attention to herself. The syndrome is now referred to as fabricated or induced illness (FII). Critics say that FII is a myth.

Meadows also promoted the proposition known as Meadow's Law: that one cot death is a tragedy, two are suspicious and three is murder - unless proved otherwise. He has acted as an expert witness in more than 5,000 civil cases. Many of these cases resulted in children subsequently being adopted.

Initially, it was reported that these civil cases would also be reviewed. Now the decision has been delayed, leaving many biological and adoptive parents unsettled and anxious. Margaret Hodge, the Children's Minister, has been criticised for reacting too quickly when consultation might have been wiser.

Social services departments anticipate that a much smaller number of cases, around 100 of the 5,000, will merit re-inspection. 'No child will have been adopted or taken into care solely on the basis of expert witnesses,' says Andrew Cozens, president of the Association of Directors of Social Services. 'Many have been based on individual confessions, witness statements and credible medical evidence.'

Professionals also insist that now it is rare for a single expert's opinion to carry a case. Shabde says that in court she is required to explain the source of her views and stay within her area of expertise. (Meadows is facing a serious professional disciplinary hearing for allegedly breaching that rule). 'We have protocols. Expert evidence is only a small piece of a larger jigsaw puzzle.'

For years Meadow's views nevertheless did have a strong influence. He was one of a group of expert witnesses appearing most often for the prosecution. 'He delivered the goods,' said one lawyer.

Chris Milroy, professor of forensic pathology, argues for change in the use of experts. They should not be identified too strongly with one side and should base their opinion on solid evidence, he says.

Mothers kill for a variety of reasons, such as stress, accident, post-natal depression, a cry for help, cruelty or as an example of FII. Diagnosis is often crude, as is the collection of evidence. A situation that may now be rectified. In Britain, the number of cot deaths has dropped from around 1,000 in the 1980s to 300 a year now. Between 6 and 10 per cent of those may be due to malice or ill-treatment or neglect. Increasing medical knowledge means non-malignant diagnoses are now possi ble for what once appeared to be shaken or battered babies while genetic illnesses (as in the Patel case) are slowly also being recognised.

At the same time, some professionals believe that many murders may have gone undetected. One woman who allegedly lost five babies to Sids and was part of a medical study on cot deaths confessed two decades later to having killed them all. 'Perhaps we ought to consider whether these kinds of cases should be in the criminal courts at all,' says Dr Jean Keeling, an eminent paediatric pathologist.

A positive step would be to improve standards of investigation immediately after death. Only a third of post mortem examinations on babies who die suddenly are carried out by paediatric specialists.

Helena Kennedy QC is heading an inquiry to devise a better system of investigating babies' deaths. Milroy also argues for a change in the way such deaths are registered: 'Many are recorded as "natural" or sudden infant death syndrome,' he says. 'That conceals more than it explains. All deaths of infants must be investigated. If natural is on the death certificate, everything stops. Unascertained is more accurate. We need to protect others from potential harm.'

If the courts are calling for hard evidence, one obvious way would be by the use of covert video surveillance. In the mid-1980s, paediatrician Dr David Southall of North Staffordshire Hospital reported on the use of video cameras in two hospitals. It revealed abuse in 33 out of 39 cases, including suffocation and poisoning. Between them the 39 patients monitored had 11 siblings who had died of Sids. The parents eventually admitted deliberate suffocation in eight of the 11 cases.

The use of CVS was brought to an end because of media outrage .

'Even using CVS... there were still medical experts prepared to argue the issue,' says Dr Martin Samuels, one of Southall's team. 'There will always be some contention among experts.'

Another cause of the current furore, says Peter Marsh, professor of child and family welfare, is what's happened to social work over the past 20 years. 'Social workers are crucial to investigating families and making judgments, yet they are paid less than teachers and trained less than doctors,' he says. 'They can't act in the way they should. They have no opportunity to give another view on child protection alongside the figure of the impassioned mother and Meadow.'

'Of course, miscarriages must be corrected,' says Gretchen Precey, an independent social worker. 'But there is a need for checks and balances.

'Paediatricians are turning away from child protection because of fear of litigation. That cannot be good for the welfare of children.'