Record-Based ASD Prevalence Estimates of 2004-2005 Birth Cohort in Taiwan

Background:  Previous studies in Taiwan reported the prevalence of autism spectrum disorder (ASD) among 3- to 17-year-olds was 2.1 per 1000 in 2011, with cumulative prevalence increased from 0.18 in 1996 to 2.87 in 2005 per 1000. However, these studies included children who were eligible for social assistance payments and retrospectively collected diagnostic information from clinical samples. The figures are therefore likely under-estimated the “true” prevalence. Prospective studies on ASD prevalence estimate, and the association between potential risk factors (e.g., prematurity, gender, geographical region, and urbanization) and ASD are limited.

Objectives:  The objectives of the study are 1) to calculate the prevalent rate of ASD for children ages 0-8 in 2004-2005 birth cohorts; and 2) to examine the birth and demographic factors associated with the prevalence of ASD.

Methods:  The study is a population-based cohort study with children born in Taiwan between 1 January 2004 and 31 December 2005, and had a claim documented in the National Health Insurance (NHI) database. NHI convers 96% of citizens of Taiwan. The prevalence of ASD was defined as those children who were born in the defined birth cohorts and have been diagnosed as ASD less than 8 years of age (before 2013). The denominator of calculating the prevalent rate was the total number of live births during 2004-2005. In addition, prematurity (gestational age < 37 weeks) and demographic characteristics (gender, region of residency, and urbanization) were also obtained from the database. Geographical distributions were classified into northern, central, southern, and eastern region. Urbanization was divided into urban and rural categories. Cox regression analysis was used to analyze the predictive factors for occurrence of ASD.

Results:  The total number of children who were live birth in 2004-2005 were 424,299, and 6,086 of them were diagnosed as ASD between 2005 and 2013. The prevalent rate was 14.3 per 1,000 among 1- to 8-year-old children. Higher prevalent rates were shown in males (HR, 4.58; 95% CI, 3.28-6.38), premature babies (HR, 2.11; 95% CI, 1.80-2.47) and in those who lived in northern area compared to central area (HR, 2.21; 95% CI, 2.06-2.37); and in urban areas compared to rural areas (HR, 1.72; 95% CI, 1.62-1.83) (Table 1). Southern (HR, 1.20; 95% CI, 1.11-1.30), and eastern (HR, 1.26; 95% CI, 1.10-1.46) regions showed slightly higher prevalence of ASD than central area.

Conclusions:  Using NHI claim data, our prevalence estimate for children under 8 years old in 2004/05 birth cohort is supported by reports from other countries. Factors of premature birth, male children, place of residence in northern and urban areas were associated with higher ASD prevalence. Our results indicated that the prevalence of ASD existed difference between urban and rural areas, and may be the consequence of uneven distribution of medical resources. Further researches may need to elaborate the association of distribution of medical care and ASD.