LAST week the British government was forced to launch an advertising campaign
designed to reassure parents that the combined measles, mumps and rubella
vaccine (MMR) is safe for their children. Vaccination has been falling to
dangerously low levels because of continuing fears that MMR can trigger
autism.
The main source of these fears is the work of Andrew Wakefield, a
gastroenterologist at the Royal Free Hospital in London who has been
investigating autistic children whose parents blame their problems on MMR. “For
MMR, autism, and inflammatory bowel disease, a significant index of suspicion
exists without adequate evidence of safety,” he wrote in a paper published last
month that has fanned the flames of the controversy.
The furore started in 1998, when The Lancet published a study by
Wakefield of 12 children who had all apparently developed autism within 14 days
or so of being given MMR. They also had an inflammatory bowel disease. Wakefield
thought the evidence was suggestive enough to warrant further research into
whether MMR triggers bowel disease, in turn leading to autism.
Advertisement
Health authorities have repeatedly rejected Wakefield’s claims. But many
parents are now trying to find doctors who will give their children the more
expensive individual vaccine for measles, or they have simply refused MMR.
Do parents have anything to worry about? Wakefield thinks the bowel
inflammation might make the gut leaky, allowing “rogue peptides” from food to
poison the brain
(see “Looking for answers”).
But even if the virus does cause inflammation, that doesn’t prove the link to
autism. In fact, the bowel problems in many of the children Wakefield looked at
actually came after the behavioural changes.
That still leaves 12 children who apparently became autistic just after being
given the triple vaccine. For parents who see the first signs of autism soon
afterwards, it must be impossible not to blame MMR. Indeed, some are so
convinced they are now suing the British government for compensation.
Yet autism is usually detected when children are around 18 months old. The
MMR vaccine is first given when children are 12 to 15 months old. So it’s
inevitable that some children will be vaccinated just before the first signs of
autism become obvious.
Wakefield says he now knows of 170 children whose autism emerged shortly
after the MMR jab. But in Britain, 600,000 children receive their first dose
every year. Over the 10 years that these cases represent, calculations suggest
that at least 455 children should have been diagnosed as autistic in the two
months after their MMR.
As persuasive as such cases might seem when looked at individually, they fall
well within what would be expected by chance. As Wakefield himself wrote in his
original paper: “We did not prove an association.”
Since 1998, the results of a number of studies have been published, some
looking specifically at his claims. For example, a study of 1.7 million primary
schoolchildren in Australia vaccinated in 1998 found no link with inflammatory
bowel disease or autism (Communicable Disease Intelligence, vol 24, p
27).
But healthcare professionals and parents were asked only to report any side
effects developing within 30 days of vaccination. Wakefield has criticised such
studies for having such short follow-up times.
In doing so, he has moved the goalposts. Almost all the children in
Wakefield’s original study developed behavioural symptoms within about 14 days
of vaccination. But now he says that inflammatory bowel disease and autism could
take much longer to appear.
One study in Finland traced children for much longer. Heikki Peltola at
Helsinki University and his colleagues followed the fate of children given
Merck’s MMR vaccine
(see “What’s in a shot?”).
In Finland, nearly all
vaccinations are given at 1000 public health centres. Just before MMR
vaccination began in 1982, nurses and doctors at these centres were asked to
report any potential side effects. No time limits were imposed.
Last month, the team published a full report of the study (The Pediatric
Infectious Disease Journal, vol 19, p 1127). Peltola concludes that the
risk of serious side effects is tiny
(see “It’s your choice”). What’s more, no cases of
inflammatory bowel disease or autism were reported to be linked with the 1.8
million individuals vaccinated over the 14 years until the end of 1996.
To check Wakefield’s claims of a link between bowel disease and autism,
Peltola’s team traced 31 children who had had gastrointestinal problems such as
diarrhoea, vomiting and abdominal pains within 15 days of their first dose of
MMR. The mean time from when the children’s symptoms were reported to when the
team checked up on them was 9 years—ample time for any effects to emerge.
But none had developed autism or long-term inflammatory bowel disease.
Critics of the study have argued that it relies on passive
surveillance—that is, it relies on doctors to report possible side
effects. And at the time, no autism link was suspected. But Peltola is adamant
that the study is valid. “This is the most active approach that can be
accomplished nationwide,” he says. “So far, I have found no hints of a link to
autism.” Elizabeth Miller, head of vaccination at the Public Health Laboratory
Service in London, agrees. She says it’s inconceivable that a link between the
vaccine and autism wouldn’t have been spotted in the Finnish study.
Another criticism of these follow-up studies is the lack of comparison with
non-vaccinated children. But Peltola and his team have done a small study to
tackle this issue. They gave 581 pairs of twins aged between 14 months and 6
years two shots, spaced three weeks apart. One was MMR, the other a placebo. But
neither researchers nor parents knew which was which.
The parents completed a detailed daily questionnaire for three weeks after
each shot. “It was amazing what we found,” says Peltola. There was no evidence
for any gastrointestinal problems connected with MMR (Pediatrics, vol
106, p e62). In fact, there was a higher incidence of diarrhoea in children
after the placebo. Again, no cases of autism were reported.
Could all these studies have missed a link between MMR and autism? In a paper
published in The Lancet in 1999, Miller and her colleagues took another
approach. Instead of looking at children after vaccination, they traced 498
autistic children in one area of England and then checked to see if their
condition could be linked to MMR. They found no association between the timing
of MMR vaccination and the onset of autism. Nor was there was any difference in
the age at which vaccinated and unvaccinated children were diagnosed as
autistic.
If MMR does trigger autism, it’s rather surprising that none of these studies
threw up any evidence of a link. At the very least, this research suggests that
if there is a risk, it must be absolutely minuscule. Most experts go rather
further. “Epidemiologically, this is just ridiculous,” says Peltola.
“[Wakefield] has not been able to show any scientifically solid data for his
.”
Indeed, many think that further studies are pointless. “It’s pretty sterile
to go down this research path when there are so many big questions that need to
be addressed on autism,” says Miller. But in the wake of the BSE scandal, such
attitudes could backfire. Whether the government can convince the public that
MMR is safe remains to be seen.
