A new Danish study does not support the hypothesis that there may be an association between thiomersal (also known as thimerosal), a vaccine preservative that contains ethyl mercury, and neurodevelopmental outcomes, including autism (1).

Findings in the field of methyl mercury have been used to suggest the hypothesis. Prenatal exposure to low doses of methyl mercury has been associated with subtle neurodevelopmental abnormalities in some studies (2) and symptoms of autism and methyl mercury intoxication have been claimed to be similar (3), and more research was called for (4). Interestingly, a recent study of the concentrations of mercury after exposure to vaccines containing thiomersal concluded that thiomersal poses very little risk to full term infants (5).

In Denmark, thiomersal was used in childhood vaccines from the early 1950s until 1992. The objective of the Danish study was to assess the incidence of autism among children between 2 and 10 years old before and after removal of thiomersal from vaccines to see if the discontinuation led to a decrease in the incidence of autism. For this study, the period of use of thiomersal vaccines was limited to 1961 until its discontinuation in March 1992 because information about the diagnosis of autism has only been obtainable from a nationwide computerised registration system, the Danish Psychiatric Central Research Register, since 1969 and only children born in 1961 or later were at risk of developing autism before 10 years of age. The children who followed the full vaccination program during the period 1961-1970 had received a total of 400 µg of thiomersal or 200 µg of ethyl mercury by the age of 15 months and during the period 1970–1992 they had received a total of 250 µg of thiomersal or 125 µg of ethyl mercury at 10 months of age. All vaccinations were given free of charge and acceptance of vaccinations in Denmark has always been very high; data on vaccination coverage are available from 1979 onwards and coverage rates have been over > 90%.

Figure. Incidence of autism in Denmark from 1970 to 2000 (2-9 year olds)

The study included a total of 956 children who had been diagnosed with autism during the period from 1971–2000, with a male to female ratio of 3.5:1. There was no trend toward an increase in the incidence of autism during the period when thiomersal was used in Denmark, until 1990. From 1991 until 2000 the incidence increased and continued to rise after the removal of thiomersal from vaccines, including increases among children born after the discontinuation of thiomersal.

The Danish study thus concluded that since the discontinuation of vaccines containing thiomersal in Denmark in 1992 was followed by an increase in the incidence of autism, the data do not support a correlation between vaccines containing thiomersal and the occurrence of autism.