The development of a screening instrument to identify sensory integration difficulties in children from low socioeconomic environments

Van Der Linde, Janine
Journal Title
Journal ISSN
Volume Title
Developmental difficulties, such as sensory integration difficulties, in children from low socioeconomic environments are often only identified and referred for occupational therapy services once they start formal schooling at the age of six or seven years. Occupational therapy services within the public healthcare system are experiencing constraints due to the availability of appropriate, cost effective assessments and limited knowledge of occupational therapists in sensory integration; resulting in inadequate identification of sensory integration problems and inappropriate intervention. Proper and early identification of sensory integration difficulties in children from low socio-economic environments will support appropriate referral and\or provision of relevant sensory integration stimulation programmes to address the identified needs. This study set out to determine if a contextually appropriate screening instrument would provide a tool that can guide community occupational therapists in the identification of children from low socio-economic environments who are at risk of experiencing sensory integration difficulties. A single study multiphase design with some aspects of the exploratory and explanatory sequential designs within separate phases of the study was used and the study was divided into three phases to ensure a systematic process was followed. Phase 1 followed an exploratory sequential research design and aimed to develop the items for the screening instrument identifying sensory integration difficulties in children of 5 years 0 months to 6 years 11 months from low socio-economic environments. Phase 2 followed an explanatory sequential research design and aimed to field test, refine and determine the internal construct validity and clinical utility of the newly developed sensory integration screening instrument in low socio-economic environments. Phase 3 used a quantitative research design and aimed to establish additional psychometric properties, such as content validity, concurrent validity with the SIPT and sensitivity and specificity testing of the newly developed sensory integration screening instrument. Phase one focused on the development of activities to be used for observations of sensory integration. Six activities reached a consensus of 70% and was included in the screening instrument, .namely dressing and undressing (80%), walking heel-toe (80%), star jumps (80%), construction with blocks (72%), making clay shapes (75%) and cutting with scissors (72%). Following the identification of these activities, an administration format and scoring system were constructed using activity analysis. This instrument development process resulted in the South African Sensory Integration Screening Instrument (SASISI). An administration manual and training program were developed to ensure standardised assessment of the SASISI. Phase two of the instrument development phase included the validation of the SASISI. A short pilot test identified several difficulties in the administration and scoring the SAISI. Identified issues were addressed before the SASISI was field tested on a sample of 200 children. Analysis of the demographic area where the sample for construct validity lived indicated severe poverty in all three areas. Age and gender for the sample was evenly spread between boys and girls, as well as the age groups. Children spoke various languages, with Zulu being the most prevalent in Soweto (53%) and Tswana in Alexandra (43%) and Potchefstroom (63%). The internal construct validity of the SASISI was determined using the Rasch model. Although none of the domains adhered to all the criteria for the Rasch model, the results are promising in terms of the SASISI measuring the underlying constructs of sensory integration. An investigation into the clinical utility of the SASISI indicated that the instrument is appropriate for measuring sensory integration in children from low socio-economic environments. Phase three aimed to establish additional psychometric properties of the newly developed SASISI. All activities showed content validity above the recommended 0.83, with a mean ICVI of 0.98, a scale level content validity index of 0.98 and a scale-level content validity index of 0.91. Concurrent validity results indicated good content validity of the SASISI vs the SIPT and moderate significant correlations were found between the tests from the SIPT and domains of the SASISI. The sensitivity and specificity of the SASISI were high and ranged between 59.0 and 100 for sensitivity and 66.1 and 100 for specificity. Cut-off points were established for each domain of the SASISI at the points where the sensitivity and specificity were balanced. In conclusion, six activities were identified that, through the observation of a child’s actions, measures the ability to process, integrate and respond purposefully on sensory input. The results indicated that the SASISI showed provisional validity for screening of sensory integration difficulties despite the small sample size. Although further development needs to be done the SASISI is the first instrument developed to assist in identifying possible sensory integration difficulties in an underserved and vulnerable population in SA.
A Thesis submitted to the Faculty of Health Sciences, University of the Witwatersrand, in fulfillment of the requirements for the Degree of Doctor of Philosophy. Johannesburg, January 2019