This study looked at annual rainfall in three US states and the number of children diagnosed with autism who lived in those areas in their first three years of life. It found there to be more children with autism living in wetter areas than in drier ones.

However this type of research can only demonstrate an association. It does not prove that rainfall has any connection with the development of autism and there are many other factors that may determine why some children in these areas develop autism. The causes of autism remain unclear and are thought to include genetic, environmental and medical factors.

Where did the story come from? Dr Michael Waldman and colleagues from Cornell University, Indiana University-Purdue University, and The Children’s Hospital of Philadelphia carried out this research. The study was funded by research grants from Cornell University. The study was published in the (peer-reviewed) medical journal: Archives of Pediatric Adolescent Medicine.

What kind of scientific study was this? This was an ecological study in which the authors aimed to investigate the possibility of rainfall as an environmental trigger for autism among “genetically vulnerable” children. The researchers provide no individual child data to support why the children may be considered to be genetically vulnerable. The aim of the research was to contribute towards a better understanding of the condition, which has reportedly increased from 1 in 2,500 children 30 years ago to 1 in 150 today.

The research investigated three western US states: California, Oregon and Washington, all of which have relatively high variability in rainfall across the counties within each state. In 2005, the researchers used data from government agencies to look at the prevalence of autism in children aged six to 18 years. In California, this was assessed by dividing the number of children with autism in each county by the total school aged population in each county as reported in the 2000 US Census. Slightly different methods were used in the other states where there was a different availability of data. The children in the 2005 data collection were therefore born between 1987 and 1999.

The researchers used a “relative precipitation variable”, which was defined as the difference between annual rainfall levels in each county and the average rainfall for all counties in the sample. Average daily and annual rainfall for each county was also examined. In particular, they looked at the rainfall in three-year intervals in the period 1987 to 1999, when the birth cohorts (groups) of children were in their first three years of life, the time during which autism symptoms typically emerge.

The analyses focussed on whether autism prevalence was higher in states that received high annual precipitation, and whether autism was more prevalent in birth cohorts who spent their first three years in counties that had higher rainfall than other parts of the state. The researchers also accounted for regional variables of the total populations such as income and ethnicity.

What were the results of the study? The researchers found that within counties, the prevalence of autism in school-aged children was positively associated with the county’s average annual rainfall. Autism rates were higher in the wetter counties of Oregon and Washington compared to those which experience less rainfall. In California, where there was less variability in rainfall across the counties there was a weaker relationship between rainfall and autism.

They additionally found that the amount of rainfall that cohorts of children were exposed to in the first three years of life was positively associated with rainfall levels. Autism prevalence was similar in Washington and California, but slightly higher in Oregon after adjusting for other characteristics of the county. Other demographic factors found to be associated included a higher population (which had increased autism prevalence) and higher indigenous population (which had decreased autism prevalence).

What interpretations did the researchers draw from these results? The researchers conclude that their results are consistent with the existence of an environmental trigger for autism ‘among genetically vulnerable children’ – in this case, precipitation. They suggest that further studies are needed to focus on whether such a trigger for autism truly exists, and the reasons behind this.

What does the NHS Knowledge Service make of this study? This research found there to be a higher prevalence of autism among children living in areas that experienced higher rainfall than those living in drier areas. As the authors say, this finding supports the idea of an environmental trigger to autism.

However, the causes of autism remain unclear and speculated causes include genetic, environmental and medical factors. This type of research can only demonstrate an association and only at a population level and does not prove that levels of rainfall have any connection with the development of autism, or shed any light on how rain affects childhood development.

There are many other factors that may underlie this relationship. Although certain high-level demographic information has been accounted for, the study did not look at individual children, their socio-demographic background, home and family environments, education and peer groups, family or personal medical history, all of which may or may not have an effect on autism risk.

The study also only looked at three western states of the US, and further data from other geographic areas would give support to the theory of rainfall as an environmental trigger of autism.

Additionally, if there is an association between rainfall and autism, it may be that increased rainfall is obscuring the true causative factor, for example that high rainfall causes children to stay indoors and watch more television, play computer games, etc; or that higher precipitation is increasing exposure to chemicals in the environment. Diagnostic criteria for autism may also differ across counties.

]]>