Studies pertaining to language impairment in school-age autistic individuals underreport participant socio-demographics: A systematic review

Abstract

Autism research tends to exclude racially and ethnically minoritized autistic individuals, limiting the quality of the evidence base for characterizing and diagnosing language impairment in racially and ethnically minoritized autistic individuals. This systematic review examines reporting practices for participant socio-demographics in studies pertaining to language impairment in autistic school-age individuals using age-referenced assessments. This review was preregistered (PROSPERO: CRD42021260394) and followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses protocol. Database searches took place in August of 2021 in Linguistics and Language Behavior Abstracts, PsycINFO, PubMed, and the Directory of Open Access Journals, for publications during 1980–2021. Search terms included three essential concepts: autism, language, and age. Two coders independently screened and evaluated articles. Of qualifying studies ( n = 59), 17 (29%) reported any information on participant race and ethnicity; in these studies, participants were at least 77% white. Thirty-two studies (54%) reported gender or sex; just 10 studies (17%) reported socio-economic status using multiple indicators. These findings indicate insufficient adherence to reporting guidelines and systematic exclusion of racially and ethnically minoritized. Future work must address replicable reporting and inclusion in autism research. Lay abstract Although exclusion of racially and ethnically minoritized autistic individuals from research is a long-standing issue, we have yet to determine how exclusion impacts areas of autism research important for identifying language impairment. Diagnosis depends on the quality of the evidence (i.e. research) and is often the pathway to gaining access to services. As a first step, we examined how research studies related to language impairment in school-age autistic individuals report participant socio-demographics. We analyzed reports using age-referenced assessments in English ( n = 60), which are commonly used by both practitioners and researchers to diagnose or identify language impairment. Findings showed only 28% of studies reported any information on race and ethnicity; in these studies, most (at least 77%) of the participants were white. In addition, only 56% of studies reported gender or sex and specified what they were reporting (gender, sex, or gender identity). Just 17% reported socio-economic status using multiple indicators. Altogether, findings indicate broad issues with underreporting and exclusion of racially and ethnically minoritized individuals, which might overlay with other aspects of identity including socio-economic status. It is impossible to determine the extent and precise nature of exclusion without intersectional reporting. To ensure that language in autism research is representative of the autistic population, future research must implement reporting guidelines and broaden inclusion of who participates in research studies.