More and More Autism Cases, Yet Causes Are Much Debated By ERICA GOODE Published: January 26, 2004 No one disputes it. Cases of autism, the baffling and often devastating neurological disorder that strikes in early childhood, are rising sharply. In California alone, the number of children receiving special services for autism tripled from 1987 to 1998 and doubled in the four years after that. National figures tell a similar story. The upsurge has lent urgency to calls for more research on autism and more government spending to educate autistic children and has inspired federal officials, who late last year held an "autism summit" meeting in Washington, where they presented a 10-year plan of action. But what lies behind the increase in cases is sharply debated. To some, the upswing has all the hallmarks of an epidemic and indicates that autism itself is increasing rapidly. To others, the rise can in large part be explained by increased public awareness of autism in recent years, changes in the way the disorder is diagnosed and the incentive of tapping into federally mandated services for autistic children. Neither side can prove its argument, because the types of studies that could tease out a true increase have not been done. But the question is crucial, experts say, because its answer has significant implications for how federal money is spent, how afraid parents should be and how much effort scientists should devote to tracking down environmental factors in addition to genetic influences. Advocacy groups, many of them founded by parents of autistic children, have tended to line up on the side of an epidemic. And some autism experts also believe the illness is increasing. "To me, it's a huge public health emergency, a crisis," said Portia Iversen, a founder of Cure Autism Now, an organization based in Los Angeles that finances research. Ms. Iversen said she was certain that the number of children with autism was rising sharply. But epidemiologists cluster on the other side of the debate. They do not rule out the possibility of a true increase in autism. But they point to flaws in the way that the rising numbers ? especially those in California ? have been presented to the public. And they say the small size and widely varying findings of epidemiological studies of autism make it impossible to say what is going on. For example, Dr. Eric Fombonne, an epidemiologist and a professor of child and adolescent psychiatry at McGill University, said most of the increase was probably a result of diagnostic changes and statistical anomalies. What everyone agrees on is that autism is being diagnosed more frequently than in the past. The disorder, which is believed to be strongly influenced by genes, is marked by a profound impairment in the ability to relate to other people, a delay in language development, or repetitive behaviors. Before the mid-1980's, most studies estimated the prevalence of autism at fewer than 5 cases for every 10,000 children. Over the last decade, epidemiological studies have come up with wildly disparate estimates, from 5.2 cases per 10,000 (in a large Norwegian study) to 72.6 per 10,000 (in a small Swedish study). But the trend has been upward, with most experts agreeing that at least 10 children out of every 10,000 are autistic. Last year, in a review of all available studies of autism rates, Dr. Fombonne concluded that the findings "point toward an increase in prevalence over the last 15 years." But predictions for the future differ along with views of what is responsible for the increase. If autism itself is increasing rapidly, the rates can be expected to keep rising, and with them, the projections of how much money will be required for services. But if most of the increase reflects more accurate diagnosis, then rates should level off as the number of previously overlooked children diminishes. The different interpretations also point researchers in different directions. "If you accept the fact that the numbers have increased, you must seek an environmental cause," said Mark Blaxill, of Cambridge, Mass., a member of the board of SafeMinds, one of several advocacy groups that view some vaccines given to toddlers as a likely cause of the increase in autism. Over the years, a host of other environmental factors have also been nominated as culprits, including a variety of infections, like German measles in pregnant mothers; the sedative drug thalidomide; the drug Pitocin, used to induce labor; synthetic compounds like plastics and PCB's; and food additives. Yet so far, said Dr. Fred R. Volkmar, an autism expert and professor of psychiatry at the Yale Child Study Center, "hard scientific evidence to support any specific environmental cause has been lacking." "Many of the purported environmental causes," Dr. Volkmar went on, "have been proposed on the basis of a single case, or a handful of cases, and the observations have not held up in larger samples." A number of studies, including a large Danish trial, have found no link between autism and the so-called MMR vaccine for measles, mumps and rubella, though some advocates remain unconvinced. The possible role of a mercury-based vaccine preservative, thimerosal, is still being investigated. But many investigators express skepticism for such a connection. For the last few years, vaccines sold in the United States for the routine immunization of children either have not contained thimerosal or contain only a trace, according to the Centers for Disease Control and Prevention. The agency and the American Academy of Pediatrics have stated that no credible scientific evidence connects thimerosal with autism or other childhood neurological disorders. If autism itself is not increasing markedly, the role of environmental influences diminishes. Epidemiologists say the wide variance in estimates from the prevalence studies that do exist support the idea that the disorder has not increased hugely. In 2000 and 2001 alone, various published studies in the United States, Britain and Scandinavia put autism rates per 10,000 children at 30.8, 7.8, 12.2, 40.5, 26.1, 13.2 and 16.8. A study in Brick Township, N.J., published in 2001, found 36 cases among 8,896 children studied, a rate of 40.5 per 10,000. Each study uses different methods and defines autism slightly differently. Complicating matters further, some studies include what are called autism spectrum disorders, a category that includes conditions like Asperger's syndrome. Another reason for caution, the epidemiologists say, is that the numbers that have received the most publicity ? the California findings, for instance, reported by the state's Department of Developmental Services ? are not based on scientific studies. Instead, such reports are simply tallies of the number of children enrolled in state programs who carry an autism diagnosis. They do not take into account changes in methods of diagnosis or shifts in population over time. Even so, the scientific support that would clinch the skeptics' arguments is lacking. "We don't have a lot of evidence one way or another, sadly," said Dr. Craig Newschaffer, an associate professor of epidemiology and mental health at Johns Hopkins University. Dr. Newschaffer said he believed "a large chunk" of the increase was a result of heightened public awareness and other such factors. "The devil is in how big a chunk is that big chunk," he said. Dr. Robert S. Byrd, an associate professor of clinical pediatrics at the University of California at Davis, has tried to clarify the issue by examining the California data to see whether methods of diagnosis or other statistical anomalies could account for the increase. Joined by colleagues at the university's MIND Institute, Dr. Byrd scrutinized 684 children enrolled in 21 regional centers for developmental disorders, including autism and mental retardation. About half were born from 1983 to 1985, the other half a decade later. The study was based on questionnaires given to the children's parents. It concluded that the same proportion of children in each age group ? about 88 percent ? met the diagnostic criteria for autism, suggesting that changes in diagnosis did not explain the increase in cases. The researchers also say they excluded two other possibilities: that parents were moving to California to obtain autism services and that some children with autism were in the past labeled as mentally retarded. In a report to the California Legislature in 2002, Dr. Byrd and his colleagues concluded that "some, if not all, of the observed increase represents a true increase in cases of autism in California." Yet their study, which has not been published in a peer-reviewed journal, itself became an immediate magnet for controversy. Critics found serious flaws in it, including the small sample of responses. They also cited a variety of other factors, including public awareness, that were not examined in the study and so could not be ruled out. The Centers for Disease Control, which in 2000 began a surveillance program to track autism cases in 18 states, may have a better chance of deciphering the issue. Dr. Marshalyn Yeargin-Allsop, the medical epidemiologist who is overseeing the effort, said the agency was studying the prevalence of autism, demographic factors like race and ethnicity, an array of possible causes of the disorder and whether autism increases over time. In the meantime, if the numbers keep rising, the debate over what they mean will continue. "When do you say some of it has to be real?" Dr. Newschaffer asked. "I don't think anybody knows the answer to that question."