A link between the rise in autism and the MMR jab has been comprehensively rejected by a study that claims to have found "compelling evidence" the triple vaccine is safe.

The reported surge in childhood autism cases is not real and is mainly due to better diagnosis, the research shows. Scientists also found no evidence of a new type of MMR-triggered autism.

The study, in the journal Archives of Disease in Childhood, was led by Professor Brent Taylor, of the Royal Free Hospital in London. It suggested parents who blamed the jab for autism might have been influenced by the media coverage. Researchers said the number of parents who dated the start of their child's autism from their MMR vaccination began to rise only after publication of a study that first suggested a link.

Another Royal Free scientist, Dr Andrew Wakefield, published research in 1998 that claimed MMR had triggered a new variant of autism in a small group of susceptible children. This caused uptake of MMR to plummet, which has led to more outbreaks of measles and mumps. More than 1,500 families are bringing legal action against the vaccine manufacturers, claiming their children have been damaged.

Dr Wakefield says his study led to him being hounded out of the Royal Free. He now works in America. He and his supporters say they are backed by studies that have suggested a sevenfold increase in autism cases in the 10 years since MMR started in 1988.

Professor Taylor and his team studied records for five areas of north London between 1979 and 1998. They found an apparent rise in autism cases until 1992, since when figures levelled. More importantly, the age at which children were diagnosed had fallen substantially.

Professor Taylor said: "This levelling off, together with the reducing age at diagnosis, suggests the earlier recorded rise in prevalence was not a real increase but was likely due to factors such as increased recognition, a greater willingness on the part of educationalists and families to accept the diagnosis, and better recording systems."

Stuart Notholt, of the National Autistic Society, said: "Whether there has been a true increase is not known. It might be expected that with growing familiarity with autistic spectrum disorders, prevalence figures would begin to level out." The charity has remained neutral on any link between the condition and MMR.