Sensory Subtypes in Children with ASD and Associated Outcomes: Latent Profile Transition Analysis Using a National Survey of Sensory Features

Background: Sensory features are prevalent in children with ASD and will soon be associated with the core characteristics. The heterogeneity of sensory features has long been discussed but lacking in the literature is the identification of homogeneous sensory phenotypes as well as assessing the stability of such identified subtypes overtime. 

Objectives: This study uses latent profile transition analysis (LPTA) to identify sensory subtypes, assess their stability overtime, and presents association of the subtypes to child characteristics and functional outcomes (i.e.,Vineland Adaptive Behavior Scale-II (VABS) and Parenting Stress Index Short Form (PSI)).

Methods:  Data were collected from participants with ASD, ages 2-12, at two time points (Time 1, n= 1307, Time 2, n=884), one year apart as part of a national online survey. A confirmatory factor analysis (CFA), of the Sensory Experience Questionnaire-3.0 (SEQ-3.0) yielded four factors of sensory response patterns (i.e., hyporesponsiveness; HYPO, hyperresponsiveness; HYPER, sensory interests, repetitions, and seeking behavior; SIRS, and enhanced perception; EP).  These scores were exported for an LPTA.  Previous literature, latent profile analysis (LPA) from both time points, LPTA with multiple profile solutions, and assessing statistical fit (AIC, BIC, Lo-Mendell-Rubin test, entropy, and the Bootstrap Likelihood Ratio Test) were used to determine the appropriate number of distinct subtypes overtime. The final LPTA was run with select child and family covariates as well as outcome measures.

Results:  Four distinct sensory subtypes were supported by statistical measures and theoretical/clinical models. Participants (n=971, 91%) remained stable in their sensory subtype across one year. The first subtype (n=297, 31%) described children who scored low on all sensory patterns, while the second subtype (n=189, 19%) showed the opposite profile, with high scores in all four sensory patterns. The third subtype (n=294, 30%) scored very close to the mean on all patterns, with a tendency to score low on HYPO and SIRS, but higher on HYPER and EP. The fourth subtype (n=191, 20%) had the opposite pattern of the third subtype with scores more extreme on HYPO and SIRS.  The four sensory subtypes related differentially to outcome measures.  The first subtype (81.38) had the highest VABS Adapted Behavior Composite score, followed by the third (79.84), second (70.65), and fourth (61.56) subtypes. The first and third subtypes were not significantly different.  However, the second and fourth subtypes were significantly different from each other as well as the other two subtypes.  The first (89.05) subtype had the lowest PSI total score followed by the third (96.64), fourth (103.15), and second (108.97) subtype. The second and fourth subtypes were not significantly different from each other.  However, the first and third subtypes were significantly different from each other as well as the other two subtypes.

Conclusions:  The LPTA identified four distinct sensory subtypes that were stable over one year in a population of children ages 2-12 with ASD. The identification of homogenous sensory subtypes and characterization of children within their subtype to functional outcomes and demographic variables will lead to improved assessment, treatment and potentially inform biological mechanisms.