Abstract
Purpose
The fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) introduced severity level specifiers for autism spectrum disorder (ASD) with minimal description of the criteria for categorizing three levels of severity (1 to 3, with 3 being the most “severe”). The objective of the analysis was to assess the prevalence of ASD severity levels using population-based surveillance data.
Methods
We analyzed severity level data on children with ASD ages 4- and 8-years-old in 2018 and 2020 in the multisite Autism and Developmental Disabilities Monitoring (ADDM) Network. Prevalence of any documented severity level and of each individual level were calculated overall and by demographic characteristics. Prevalence ratios adjusted for sex, race/ethnicity, age, intellectual disability, ADDM surveillance year, and site (aPRs) and 95% confidence intervals (CIs) were used for comparisons.
Results
Less than half (40.4%) of children with documented ASD diagnoses had any severity level specified in their records, with wide variation by site (4.8%-73.2%). Severity levels were more common in records of children aged 4, in surveillance year 2020, and more often missing in non-Hispanic Black children and from records also missing information on intellectual disability (ID). Higher prevalence of more severe (level 3) ASD was observed among non-Hispanic Black children, children aged 4 years, children in 2020, and children with ID.
Conclusion
Utilization of the DSM-5’s severity levels by community professionals varied widely, limiting their potential utility in identifying needed services and supports for children with ASD.
From the article:
Wide variation in use and assignment of severity levels across sites suggest a lack of standard of practice for determining a child’s level of severity. This may be due to differences in diagnostic training, clinical protocols, or documentation practices. The DSM-5 specifies that “the descriptive severity categories should not be used to determine eligibility for provision of services. Indeed, individuals with relatively better skills overall may experience different or even greater psychosocial challenges. Thus, service needs can only be developed at an individual level and through discussion of personal priorities and targets” (APA, 2013). Providers may therefore question the utility of severity levels defined by support needs, when the DSM also advises that service needs be determined on an individual basis. There is at least one documented instance of severity levels being used to determine eligibility for services: Australia requires at least a level 2 designation to receive their National Disability Insurance Scheme (NDIS) (National Disability Insurance Agency, 2022). It is unknown whether and how often severity levels are used by U.S. service systems.
The degree to which autistic people find utility in the severity levels is unclear. Some who had been diagnosed under the DSM-IV with Asperger’s disorder found the Asperger’s descriptor helpful in understanding themselves and in describing their needs, and they reported concern when it was removed from the DSM-5 (Kapp & Ne’eman, 2020). The severity levels introduced in the DSM-5 could provide a similar type of descriptor that some autistic people find useful. However, other autistic people have expressed concern about the use of severity levels because they might be used to limit care access or to inappropriately group people with very different types of support needs (Kapp & Ne’eman, 2020). Similar concerns have been expressed in other attempts at grouping functioning in ASD, notably with the introduction of the term “profound” ASD, coined by the Lancet Commission on the Future and Care and Clinical Research in Autism (Kapp, 2023; Kripke-Ludwig, 2023; Lord et al., 2022; Pukki et al., 2022).
