Secondary Logo

Journal Logo


Changes in Risk of Autism in the U.K. for Birth Cohorts 1990–1998

Jick, Hershel; Kaye, James A.; Black, Corri

Author Information
doi: 10.1097/01.EDE.0000082044.88833.c4
  • Free

To the Editor:

Recently we reported on the risk of first-time diagnosed autism for boys aged 2-5 years in annual birth cohorts of children born in 1988-1993.1 The study was based on information available in the U.K. General Practice Research Database (GPRD). Risks rose progressively throughout this study period and were 3-4 times higher in the 1993 birth cohort compared with the 1988 cohort. To determine whether the risks have continued to rise beyond the 1993 cohort, we examined birth cohorts of boys for the years up through 1998.

Like in our prior study, we identified boys in the GPRD who had a first-time diagnosis of autism. There were a few changes in the criteria for inclusion (or exclusion) of children as cases of autism: (1) We required at least 18 months of recorded information in the computerized practice record before the first diagnosis of autism, and (2) we included only boys aged 2-4 years (24-59 mo) at the time of first diagnosis for the years 1990-1998. (Boys account for approximately 80% of cases.) Risks were calculated as previously described,1 for 2-3 year olds and 2-4 year olds separately.

We identified 129 males with a first-time recorded diagnosis of autism. The risk of diagnosed autism for 2-3 year olds rose consistently from approximately 3 per 10,000 in the 1990 cohort to approximately 25 per 10,000 in the 1996 cohort, after which it either stabilized or decreased in the 1997 (19 per 10,000) and 1998 (15 per 10,000) birth cohorts. Similarly, the risk for 2-4 year olds rose consistently from approximately 9 per 10,000 in the 1990 birth cohort to approximately 33 per 10,000 in the 1996 birth cohort and did not increase in the 1997 (30 per 10,000) cohort (see Fig.).

Three-year risk of diagnosed autism among boys age 24-59 months in the General Practice Research Database by year of birth. Bars indicate 95% confidence intervals.

Substantial increases in the frequencies of diagnosed incidence of autism over the past decade have now been reported from the U.K.,1,2 California,3 and Denmark.4 The current results indicate that the diagnosed incidence of autism, which was previously found to increase for boys born in the U.K. from 1988-1993,1 has continued to rise to an apparent peak in boys born in 1996. Because the majority of children with autism are diagnosed at ages 2-4 years, it should be kept in mind that such children born in 1996 would have been diagnosed in the years 1998-2002. Thus, in considering possible explanations for the increasing frequency of the diagnosis of autism, it is important to keep in mind circumstances in those years that might have led to an increase in the frequency of diagnosis of this illness. In this regard, at least part of the more recent increase could have been related to widespread publicity surrounding the papers of Wakefield et al.5 and Taylor et al.2 (published in 1998 and 1999, respectively), which described the increasing risk of autism in the past decade.

Whether the dramatic increase in the diagnosed incidence of autism through the 1996 birth cohort is the result of changes in criteria for the diagnosis, increased public awareness of the illness, some other factor(s), or a combination of the 3 remains to be elucidated. Given the magnitude of the increased risk for the diagnosis of this serious illness over the past decade, the need for further investigation to elucidate the cause is urgent.


We are indebted to the general practitioners who contribute information to the GRPD for their continuing effort and cooperation.

Hershel Jick

Boston Collaborative Drug Surveillance Program

Boston University School of Medicine

11 Muzzey Street

Lexington, MA

[email protected]

James A. Kaye

Boston Collaborative Drug Surveillance Program

Boston University School of Medicine

Boston, MA

Corri Black

Department of Public Health Medical School

University of Aberdeen

Aberdeen, Scotland


1.Kaye JA,Melero-Montes MM, Jick H. Mumps, measles, and rubella vaccine and the incidence of autism recorded by general practitioners: a time trend analysis. BMJ. 2001;322:460–463.
2.Taylor B,Miller E, Farrington CP, et al. Autism and measles, mumps, and rubella vaccine: no epidemiological evidence for a causal association. Lancet. 1999;353:2026–2029.
3.Dales L,Hammer SJ, Smith NJ. Time trends in autism and in MMR immunization coverage in California. JAMA. 2001;285:1183–1185.
4.Madsen KM, Hviid A,Vestergaard M, et al. A population-based study of measles, mumps, and rubella vaccination and autism. N Engl J Med. 2002;347:1477–1482.
5.Wakefield AJ,Murch SH, Anthony A, et al. Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet. 1998;351:637–641.
© 2003 Lippincott Williams & Wilkins, Inc.