Fifty-three percent of occupational therapists lack confidence when working with the autism population and 60% consider autism spectrum disorders (ASD) to be more challenging to work with than other diagnoses (Ashburner, Rodger, Ziviani, & Jones, 2014). One possible reason for this is the heterogeneity in clinical presentation in ASD, as interventions may work for some children with ASD but not others (Hoddenbach et al., 2012). To narrow intervention focus and support clinical reasoning, there is a widespread effort to identify homogenous subtypes of children with ASD who may respond similarly to intervention (The Interagency Autism Coordinating Committee, 2014). One such subtyping schema was proposed by Lane et al. (2014) and consists of four sensory subtypes of children with ASD based on parent responses to the 38-item Short Sensory Profile (SSP): 1) Sensory Adaptive; 2) Taste/Smell Sensitive; 3) Postural Inattentive; and 4) Generalized Sensory Difference. Differences between sensory subtypes have been documented in three out of five health-related domains of the International Classification of Functioning, Disability and Health (ICF), including body functions and structures, activities, and personal factors (Lane et al., 2010, 2011, 2014a; Tanner and Lane, unpublished manuscript). This information may be valuable in the development of decision-making tools for clinicians working with children with ASD and their families. However, differences between the subtypes in the two remaining ICF health-related domains of environmental factors, such as caregiver burden, and child participation in valued activities have yet to be explored. Moreover, our pilot data indicate that not all SSP items are relevant to subtype determination, suggesting that reduced respondent burden and increased efficiency via item reduction may be possible. Thus, prior to widespread clinical use of sensory subtyping, it is necessary to determine differences between subtypes in caregiver burden, determine differences between subtypes in child participation, and improve the effeciency of the sensory subtyping algorithm via item reduction. The objectives of this dissertation are to: 1) determine differences between subtypes in level of burden that caregivers report; 2) determine differences between subtypes in child participation of valued activities; and 3) reduce the number of SSP items needed for sensory subtyping.
Method: To achieve these objectives, a national survey of 378 caregivers of children with ASD aged 5-13 years was conducted. Caregivers completed the SSP as well as valid and reliable measures of caregiver burden and child participation. Caregiver burden was measured by: 1) the SF-12v2, which is a measure of health related quality of life, and 2) the Caregiver Strain Questionnaire, which is a measure of perceived strain. The Participation and Environment Measure- Children and Youth was used to measure child participation in meaningful activities. Canonical correlation analyses were used to achieve objectives 1 and 2, while a multinomial logistic regression with lasso penalization was used to achieve objective 3.
Results: Results from Chapter 2 revealed that caregivers of children with ASD in different sensory subtypes did not differ in perceived mental and physical health related quality of life, but did differ in level of perceived caregiver strain. When compared with caregivers of children in the Generalized Sensory Difference subtype, caregivers of children in the Sensory Adaptive subtype reported the lowest levels of strain, followed by caregivers of children in the Taste/Smell Sensitive subtype. Caregivers of children in the Postural Inattentive and Generalized Sensory Difference subtypes did not significantly differ in level of strain. Similarly, results from Chapter 3 indicate that children in different sensory subtypes were found to differ in their participation in home, school, and community activities. When compared with children in the Generalized Sensory Difference subtype, children in the Sensory Adaptive subtype had the highest levels of participation, followed by children in the Taste/Smell Sensitive subtype. The Postural Inattentive and Generalized Sensory Difference subtypes did not significantly differ in level of participation. Findings from Chapter 4 identified a subset of 18 SSP items that determine membership to the Taste/Smell Sensitive, Postural Inattentive, or Generalized Sensory Difference subtypes with 94.6% agreement with the existing algorithm based on the 38-item SSP. This subset does not, however, identify members of the Sensory Adaptive subtype.
Conclusion: Findings of this dissertation were combined with existing literature to create a decision-making framework that focuses the scope of intervention for each sensory subtype by linking deficits with evidence-based interventions. This body of work improves the clinical and research utility of sensory subtyping in ASD by: 1) linking subtypes to areas of caregiver burden and child participation to target for intervention, and 2) reducing respondent burden by determining sensory subtypes with fewer SSP items. This research contributes much needed empirical evidence to guide clinical decision-making in intervention selection for children with ASD. Future studies are necessary to further these findings and effectively build upon this evidence by examining the degree to which this decision-making framework improves treatment outcomes for children with ASD. Increasing evidence in support of this framework will aid in its clinical translation, where it will likely have a direct positive impact on the quality of care provided for children with ASD.