Autistic spectrum disorders (ASD) are a disabling con- tinuum of disorders affecting between 1 and 2 per 1,000 live births (Bryson, 1996). These disorders have a core set of defining features that include impaired verbal and nonverbal communication and restricted and repetitive patterns of behavior (Filipek et al., 2000; Volkmar et al., 1999). While concerns about a recent epidemic of autism may be unfounded (Fombonne, 2001; Taylor et al., 1999), it does appear that the prevalence of autism is greater than previously thought, at least in part due to improved ascertainment (Wing, 1993). Early recognition and diagnosis of ASD are critical for at least four reasons. First, there are a number of behav- ioral and educational interventions for children with ASD that have demonstrated efficacy in improving outcomes for children, and there is increasing agreement that these interventions are more effective with younger children (Campbell et al., 1996; Committee on Children With Disabilities, 2001; Rapin, 1991; Volkmar et al., 1999). Second, parents require genetic advice regarding the asso- ciated risks to subsequent children (Baird et al., 2001; Simonoff, 1998). The probability of having a second child with autism is 1 in 20, compared with a community preva- lence of 1 in 500 (Bolton et al., 1994). The risk of hav- ing a second child with some broader communication or cognitive deficit is even higher (Le Couteur et al., 1996). Third, the Individuals with Disabilities Act (PL 94-142) specifies that starting at the age of 3 years, children with disabilities are eligible through the public education sys- tem for special education and related services. Autism was identified as a separate category of disabling condi- tion in 1990 under PL 101-476. Children with ASD must be identified early so that the school system can plan for and provide the “free, appropriate education” to which these children are entitled. Finally, ASD and its related symptoms are disabling not only to the child, but to the family system as well (Gray, 1997, 2001). Families often Race Differences in the Age at Diagnosis Among Medicaid-Eligible Children With Autism DAVID S. MANDELL, SC.D., JOHN LISTERUD, M.D., PH.D., SUSAN E. LEVY, M.D., AND JENNIFER A. PINTO-MARTIN, PH.D. ABSTRACT Objective: To examine racial differences in the age at which Medicaid-eligible children first receive an autistic disorder (AD) diagnosis and to examine time in mental health treatment until an AD diagnosis was received. Method: Philadelphia Medicaid specialty mental health claims identified 406 children who received services in 1999 for AD. Claims from 1993–1999 were used to identify the date of first mental health visit, first receipt of AD diagnosis, and number of visits occurring between those dates. Linear regression was used to examine the relationship among race, age at first diagnosis of AD, time in mental health treatment, and number of visits until the diagnosis was made. Results: On average, white children received the AD diagnosis at 6.3 years of age, compared with 7.9 years for black children ( p < .001). White chil- dren entered the mental health system at an earlier age (6.0 versus 7.1 years, p = .005); however, after adjusting for age, sex, and time eligible for Medicaid, black children required more time in treatment before receiving the diagnosis. Conclusions: Important disparities exist in the early detection and treatment of autism. These disparities may be the result of differences in help-seeking, advocacy and support, and clinician behaviors. J. Am. Acad. Child Adolesc. Psychiatry, 2002, 41(12):1447–1453. Key Words: autistic disorder, health disparities, children’s mental health services, Medicaid. J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 41:12, DECEMBER 2002 1447 Accepted July 9, 2002. Dr. Mandell is Assistant Professor and Dr. Listerud is a resident in the Department of Psychiatry, Dr. Levy is Associate Professor in the Department of Pediatrics at the School of Medicine, and Dr. Pinto-Martin is Associate Professor in the School of Nursing at the University of Pennsylvania. This work was supported by grants from the Centers for Disease Control (CCU32094) and the National Institute of Mental Health (5T32MH019292-12). Reprint requests to Dr. Mandell, Center for Mental Health Policy and Services Research, University of Pennsylvania, 3535 Market Street, 3rd Floor, Philadelphia, PA 19104; e-mail: mandelld@mail.med.upenn.edu. 0890-8567/02/4112–14472002 by the American Academy of Child and Adolescent Psychiatry. DOI: 10.1097/01.CHI.0000024863.60748.53