The most comprehensive study into the MMR vaccine has found no evidence that it is linked with the development of autism in children.

Researchers who examined the vaccination records of more than 1,000 children with autism were unable to find any association with MMR.

The Medical Research Council (MRC), which funded the study, said the finding added weight to existing evidence that MMR is not a cause of autism.

Public concern about safety has led to less than 82 per cent of children in the UK having the triple vaccine against measles, mumps and rubella. Levels have fallen from about 92 per cent in the mid-1990s and are now below the level needed to avoid epidemics of measles. Outbreaks have already been seen in areas where uptake of the vaccine is low.

The MMR scare was triggered in 1998 by the publication in The Lancet of a small study by Andrew Wakefield suggesting a possible link between the vaccine and bowel problems and autism in children.

Efforts to confirm Dr Wakefield's findings failed and a series of large-scale studies found no link between MMR and autism. Serious flaws in Dr Wakefield's research were exposed earlier this year and the major part of his 1998 paper was retracted.

The latest study, also published in The Lancet, looked at the vaccination records of 1,294 children diagnosed with autism or other pervasive development disorders (PDDs) between 1987 and 2001 in England and Wales.

These youngsters were compared with a control group of 4,469 children of the same sex and similar age who were registered with the same general practices but did not have a recorded diagnosis of autism.

The study actually found that fewer children with autism or other PDDs had received the MMR vaccination (78 per cent) than among the control group without autism (82 per cent). But the difference was not statistically significant and likely to be due to chance.

The team, led by Dr Liam Smeeth, tested their findings by looking at children who had MMR before their third birthday - autism is not normally diagnosed before the age of three. They also looked at children vaccinated in the period before the suspected link between MMR and autism was first reported.

The researchers finally compared children who joined the practices before their first birthday with those joining after this age. Again they found no link.

The researchers also reviewed previous studies that showed any increased risk of autism associated with MMR. They concluded that their combined results increased the strength of the assessment, also held by the Department of Health and medical experts, that there was no evidence of a link between the combined vaccination and autism.

Dr Smeeth said: "We hope the results of this study, the most robust and comprehensive undertaken to date, will reassure parents that MMR is not associated with an increased risk of developing autism.

"Our findings are consistent with evidence from a growing body of high-quality scientific studies. It is time to move on and focus on research into other potential causes of autism." Dr Smeeth added that the absence of knowledge about autism may have contributed to "misplaced emphasis" on MMR as a cause.

The MRC said measles was not the mild illness it was commonly perceived to be. One child in 15 with the infection suffers serious complications, such as pneumonia and convulsions, and one in 5,000 will get encephalitis (inflammation of the brain).