We read with interest the article of Freeman et al. (1) reporting a higher prevalence of autism spectrum disorder in pediatric patients with type 1 diabetes in Toronto than in the general population (0.9% [95% CI 0.3–1.5 vs. 0.34–0.67]).

The finding was, however, not confirmed by Harijutsalo and Tuomilehto (2), who reported a prevalence of autism spectrum disorders in type 1 diabetic patients similar to that in the population aged <18 years in northern Finland (0.135% [95% CI 0.5–2.8] vs. 0.139% [1.2–1.57]).

In Italy, we have “our own Finland,” namely Sardinia. Italy has a peculiar epidemiology of type 1 diabetes. Sardinia has one of the highest incidences in the world (42.4/100 [95% CI 40.5–44.4]), while peninsular Italy has an overall incidence that is similar to other Mediterranean areas (8.4/100 [7.9–8.9], 11.2/100 [10.3–12.2] in North Italy, and 6.2/100 [5.8–6.7] in South Italy) (3). Only 2 of 1,373 patients aged <14 years from the Sardinian Registry of type 1 diabetes were diagnosed with autism spectrum disorders (0.1456% [0.13–0.16]), a finding similar to the Finnish data. On the contrary, a pattern similar to that observed by Freeman et al. (1) has been found in patients with type 1 diabetes aged <14 years from six Italian centers of pediatric diabetology equally distributed in the Italian Peninsula and in Sicily (0.72% [0.69–0.75]). The diagnosis of autism spectrum disorder was confirmed in all cases using the DSM-IV (Diagnostic and Statistic Manual of Mental Disorders-IV). Autism spectrum disorder includes autism, Asperger disorder, pervasive developmental disorders (not otherwise specified), Rett’s syndrome, and childhood disintegrative disorder. Its prevalence in the general Italian population is estimated at 0.1%.

Our data seem to confirm the observation that the prevalence of autism in patients with type 1 diabetes is inversely associated to the incidence trends of diabetes in the observed areas. According to Daneman (4), if the relationship between autism and diabetes is based on shared genetic influences, this finding could be explained by the fact that, in areas with higher and rising incidence of type 1 diabetes without a parallel increase of the prevalence of autism, the rise in incidence of diabetes is due to a reduced contribution of high-risk HLA haplotypes (5).

1.
Freeman SJ, Roberts W, Daneman D: Type 1 diabetes and autism: is there a link?
Diabetes Care
28
:
925
–926,
2005
2.
Harjutsalo V, Tuomilehto J: Type 1 diabetes and autism: is there a link? Response to Freeman et al. (Letter).
Diabetes Care
29
:
484
,
2006
3.
Carle F, Gesuita R, Bruno G, Coppa GV, Falorni A, Lorini R, Martinucci ME, Pozzilli P, Prisco F, Songini M, Tenconi MT, Cherubini V, the RIDI Study Group: Diabetes incidence in 0- to 14-year age-group in Italy: a 10-year prospective study.
Diabetes Care
27
:
2790
–2796,
2004
4.
Daneman D: Diabetes and autism: is there a link? Response to Harijutsalo and Tuomilehto (Letter).
Diabetes Care
29
:
485
,
2006
5.
Gillespie KM, Bain SC, Barnett AH, Bingley PJ, Christie MR, Gill GV, Gale EA: The rising incidence of childhood type 1 diabetes and reduced contributions of high risk HLA haplotypes.
Lancet
364
:
1645
–1647,
2004