To evaluate the effectiveness and safety of interventions targeting sensory challenges in children with autism spectrum disorder (ASD).
We searched MEDLINE, Embase, the Cumulative Index of Nursing and Allied Health Literature, and PsycINFO from January 2010 to September 2016.
We included studies comparing interventions incorporating sensory-focused modalities with alternative treatments or no treatment. Studies had to include at least 10 children with ASD ages 2–12 years. Two investigators independently screened studies and rated risk of bias. We extracted and summarized data qualitatively because of the significant heterogeneity. We also assessed strength of the evidence (SOE).
We identified 24 unique comparative studies (17 newly published studies and 7 studies addressed in our 2011 review of therapies for children with ASD). Studies included 20 randomized controlled trials (RCTs), 1 nonrandomized trial, and 3 retrospective cohort studies (3 low, 10 moderate, and 11 high risk of bias [ROB]). Populations, intervention approaches, and outcomes assessed varied across studies. Relative to usual care or other interventions, sensory integration–based approaches improved measures related to sensory and motor skills in the short term (3 RCTs with high, moderate, and low ROB and 1 high ROB retrospective cohort study). Environmental enrichment improved nonverbal cognitive skills in treated children compared with standard care in two small RCTs (low and moderate ROB). Four small RCTs (2 moderate and 2 high ROB) of auditory integration–based approaches reported mixed results. Studies of music therapy (4 RCTs—1 low, 2 moderate, and 1 high ROB—and 1 high ROB nonrandomized trial) used different protocols and addressed different outcomes, precluding synthesis. Massage improved ASD symptom severity and sensory challenges versus a waitlist control condition (7 studies, 5 with likely overlapping participants, 3 moderate and 4 high ROB). Additional RCTs (moderate and high ROB) of interventions with sensory-related components (tactile stimulation exercises, weighted blankets) reported few significant differences between treatment groups.
Some interventions targeting sensory challenges may produce modest short-term (<6 months) improvements, primarily in sensory-related outcomes and outcomes related to ASD symptom severity; however, the evidence base for any category of intervention is small, and durability of effects beyond the immediate intervention period is unclear. Sensory integration–based approaches improved outcomes related to sensory challenges (low SOE) and motor skills (low SOE), and massage improved sensory responses (low SOE) and ASD symptoms (low SOE). Environmental enrichment improved nonverbal cognitive skills (low SOE). Auditory integration–based approaches did not improve language outcomes (low SOE). Some positive effects were associated with other approaches studied (music therapy, weighted blankets), but findings in these small studies were not consistent (insufficient SOE). Data on longer term results are lacking, as are data on characteristics that modify outcomes, effectiveness of interventions across environments or contexts, and components of interventions that may drive effects. In sum, while some therapies may hold promise and warrant further study, substantial needs exist for continuing improvements in methodologic rigor in the field.
