Free
Research Article  |   October 1997
The Test of Sensory Functions in Infants: Test–Retest Reliability for Infants With Developmental Delays
Author Affiliations
  • Tracy L. Jirikowic, MS, OTR/L, is Staff Therapist, Kwiat Child Development Center, Seattle, Washington, and Staff Therapist, Fetal Alcohol Syndrome Clinic, Center for Human Development and Disability, University of Washington, Seattle, Washington. (Mailing address: 727 Bellevue Avenue E., Seattle, Washington 98102)
  • Joyce M. Engel, PhD, OTR/L, FAOTA, is Assistant Professor, Department of Rehabilitation Medicine, Occupational Therapy Division, University of Washington, Seattle, Washington
  • Jean C. Deitz, PhD, OTR/L, FAOTA, is Associate Professor and Graduate Program Coordinator, Department of Rehabilitation Medicine, Occupational Therapy Division, University of Washington, Seattle, Washington
Article Information
Pediatric Evaluation and Intervention / Research
Research Article   |   October 1997
The Test of Sensory Functions in Infants: Test–Retest Reliability for Infants With Developmental Delays
American Journal of Occupational Therapy, October 1997, Vol. 51, 733-738. doi:10.5014/ajot.51.9.733
American Journal of Occupational Therapy, October 1997, Vol. 51, 733-738. doi:10.5014/ajot.51.9.733
Abstract

Objective. The Test of Sensory Functions in Infants (TSFI) is one tool that occupational therapists use to identify sensory processing disorders among infants. However, data on the reliability of TSFI scores with infants with developmental delays are lacking.

Method. Test–retest reliabilities for TSFI total test and subtest scores were determined with a sample of 26 infants with developmental delays. All infants were between 10 months and 18 months of age. The test–retest interval ranged from 5 days to 10 days. Magnitudes of difference between test and retest scores and percentages of agreement among the TSFI classification categories (i.e., normal, at-risk, deficient) also were determined to examine relationships between test scores.

Results. Reliability for the total test score was borderline, with an intraclass correlation coefficient of .78. Reliability coefficients for the five subtests ranged from .54 to .74. Percentage of agreement for the total test classification categories between test and retest was adequate (81 %). Percentages of agreement for subtest classification categories were low, ranging from 58% to 68%.

Conclusion. TSFI scores should be interpreted cautiously and used only in conjunction with findings from additional developmental assessments and clinical observations for infants with developmental delays.