RESUMO
PURPOSE: Diagnostic accuracy of autism spectrum disorder (ASD) is crucial to track and characterize ASD, as well as to guide appropriate interventions at the individual level. However, under-diagnosis, over-diagnosis, and misdiagnosis of ASD are still prevalent. METHODS: We describe 232 children (MAge = 10.71 years; 19% female) with community-based diagnoses of ASD referred for research participation. Extensive assessment procedures were employed to confirm ASD diagnosis before study inclusion. The sample was subsequently divided into two groups with either confirmed ASD diagnoses (ASD+) or unconfirmed/inaccurate diagnoses (ASD-). Clinical characteristics differentiating the groups were further analyzed. RESULTS: 47% of children with community-based ASD diagnoses did not meet ASD criteria by expert consensus. ASD + and ASD- groups did not differ in age, gender, ethnicity, or racial make-up. The ASD + group was more likely to have a history of early language delays compared to the ASD- group; however, no group differences in current functional language use were reported by caregivers. The ASD + group scored significantly higher on ADI-R scores and on the ADOS-2 algorithm composite scores and calibrated severity scores (CSSs). The ASD- group attained higher estimated IQ scores and higher rates of psychiatric disorders, including anxiety disorder, disruptive behavior, and mood disorder diagnoses. Broadly, caregiver questionnaires (SRS-2, CCC-2) did not differentiate groups. CONCLUSION: Increased reported psychiatric disorders in the ASD- group suggests psychiatric complexity may contribute to community misdiagnosis and possible overdiagnosis of ASD. Clinician-mediated tools (ADI-R, ADOS-2) differentiated ASD + versus ASD- groups, whereas caregiver-reported questionnaires did not.
RESUMO
Epidemiological surveys have increased in number and complexity. This paper focuses on a few selected issues that persist in the field of autism epidemiology today. Recent surveys have improved case ascertainment by including a regular school survey component, and they have consistently identified cases which were undiagnosed. However, the logistic of these surveys is complex, and the data analysis of these complex survey designs can lead to serious over- and sometimes under-estimation of overall population prevalence. Other issues discussed concern techniques used to confirm and validate caseness in surveys. Surveys relying on mechanical use of diagnostic algorithms or single questionnaire answers are plagued with measurement error. The impact of the repeated changes in nosographical systems and the recent changes of DSM-5 are discussed. Issues specific to surveys of adults and preschoolers are briefly reviewed. Finally, contributions from international studies and the need for further cross-cultural comparisons are considered.
Os estudos epidemiológicos aumentaram em número e complexidade. Este artigo tem como foco alguns problemas que ainda persistem no campo da epidemiologia do autismo. As pesquisas mais recentes aprimoraram os métodos de averiguação de casos, principalmente incluindo coleta de dados em escolas regulares. Essa estratégia tem, consistentemente, permitido a identificação de casos que anteriormente ficavam sem diagnóstico. Entretanto, a logística dessas pesquisas é complexa e a análise de dados desses estudos pode levar a uma sobrestimação e, às vezes, a uma subestimação da prevalência global da população. Outra temática discutida neste artigo diz respeito às técnicas usadas para confirmar e validar pesquisas. Estudos baseados no uso mecânico de algoritmos ou em pontuação de um único questionário podem levar a erros de medição. O impacto das frequentes mudanças nos sistemas nosográficos e as recentes alterações no DSM-5 são discutidas. Questões específicas para pesquisas com adultos e crianças são brevemente revisadas. Finalmente, são consideradas contribuições de estudos internacionais e a necessidade de mais comparações interculturais.
Estudios epidemiológicos han aumentado en número y complejidad. Este artículo se centra en problemas que persisten en el campo de la epidemiología del autismo. Investigaciones recientes han mejorado la detección de casos, incluido el componente de investigación en la escuela. Esto ha identificado casos que no fueron diagnosticados. Sin embargo, la logística de encuestas es compleja y el análisis de los datos puede llevar a sobreestimación o subestimación de la prevalencia de la populacion. Otros temas discutidos son las técnicas utilizadas para confirmar y validar la investigación. La pesquisa basada en el uso mecánico de algoritmos o cuestionario está plagada de errores de medición. Se discute el impacto de los cambios repetidos en los sistemas nosográficos y los cambios recientes en el DSM-5. Las preguntas para la investigación con adultos y niños se revisan brevemente. Se consideran las contribuciones de los estudios internacionales y la necesidad de más comparaciones interculturales.