Autism Diagnoses May Vary by Gender, Age, Ethnicity, and Location

The incidence of autism varies on the basis of gender, age, ethnicity, and geographical location, according to study results published in The Lancet Child & Adolescent Health.

Researchers conducted a longitudinal, retrospective school registry study and sourced from the National Pupil Database, which collected data annually between 2014 and 2017 in the United Kingdom. The incidence of autism among 23,851,455 pupils was evaluated on the basis of student demographics and school characteristics.

A total of 102,338 students received a new diagnosis of autism during the study, corresponding with an incidence rate (IR) of 426.9 per 100,000 person-years after adjusting for age and gender.

Stratified by age, the highest adjusted IR was observed in children aged 1 to 3 years (IR, 555.9 per 100,000 person-years), followed by 4 to 6 years (IR, 508.1 per 100,000 person-years) and 10 to 12 years (IR, 490.5 per 100,000 person-years). The lowest rate was observed in teenagers aged 16 to 18 years (IR, 267.6 per 100,000 person-years).

The IR of autism among boys was 4 times (IR, 668.6 per 100,000 person-years) that of girls (IR, 173.2 per 100,000).

Stratified by ethnicity, the IR of autism was highest for unclassified ethnicity (IR, 599.4 per 100,000 person-years), followed by Black (IR, 466.9 per 100,000 person-years), mixed race (IR, 436.7 per 100,000 person-years), White (IR, 431.0 per 100,000 person-years), and Asian (IR, 276.9 per 100,000 person-years) ethnicities.

By geography, there were 2338 autism hotspots in England and 881 autism cold spots. The hotspots tended to be located in the South East (10.98%), West Midlands (10.95%), East Midlands (10.56%), and London (7.6%). In general, these spatial correlations were stronger among boys (I, 0.238; P <.0001) than girls (I, 0.105; P <.0001).

After adjusting for cultural factors, family- and area-level deprivation, ethnicity, gender, year, and age, boys of any ethnicity who claimed free school meals and did not speak English as their first language were at highest odds for autism (OR range, 5.39-16.64). Girls who did not claim free school meals and did not speak English as their first language were at lowest odds for autism (OR range, 0.46-0.66).

This study may not be generalizable for other educational systems or for students enrolled in specialty schools or were homeschooled.

Study authors conclude, “The results from this work highlight how the incidence of autism differs between ethnic groups, aligning our work with existing literature on autism incidence in England, the United Sates, and Nordic countries. […] Our results challenge researchers to better understand the process of receiving an autism diagnosis and to what degree social determinants, ethnicity and level of deprivation in particular, affect autism status in the English educational system, while considering the proper support that should be available throughout this process to autistic people.”