Presentation of ASD in Females: Examining DSM Criteria in a Clinical Sample

Background: Prevalence rates of a 4:1 male to female ratio for autism spectrum disorder results in narrow population of study, with females representing less than 15% of participants in ASD samples across published intervention studies (Watkins et al., 2014). Early research suggested girls with ASD were more likely to have an intellectual disability as compared to boys with ASD. However, recent research suggests that girls and women with ASD may present with different symptoms as compared to boys and men, resulting in a delayed diagnosis for females or missing high functioning females (Head et al, 2014; Lai et al, 2011; Begeer et al, 2013; Niller, 2014).To date, much research focuses on sex differences between males and females with ASD on specific measures with less emphasis on characteristics that may influence diagnosis or changes in symptoms over time.

Objectives: The primary objective is to examine the presentation of ASD in a clinic sample of females referred for an autism evaluation. The secondary objective is to consider potential sex differences in diagnostic criteria and to identify variables for further study and hypotheses regarding manifestation of ASD in females.

Methods: We examined characteristics of ASD through a retrospective analysis of a clinical sample, including potential sex differences in diagnostic variables (age at diagnosis, co-existing symptoms) as well as a descriptive analysis of DSM criteria in females with ASD as compared to males. Our sample includes data from our clinical database (n > 1300) of patients referred for an autism evaluation in our interdisciplinary clinics across the past three years. Analysis includes descriptive statistics, frequencies, and t-tests to compare groups. DSM criteria is examined in a subsample of males and females.

Results: Females are less likely to receive a diagnosis of ASD as compared to males with a significantly lower rate of diagnosis. Although we did not find a significant difference for age of ASD diagnosis in our clinical sample, females that did not receive a diagnosis were significantly older compared to males that did not receive a diagnosis. Similar to Hiller et al. (2014), females in our sample demonstrated low rates of restricted interests. We also found fewer symptoms related to deficits in nonverbal communication, play, or stereotyped behaviors.

Conclusions: Females in our clinical sample demonstrate some differences in diagnostic criteria as compared to males, with some symptoms predicting an increased likelihood of diagnosis in females. These data provide direction for further study and potential hypotheses related to how ASD manifests differently in females.