The reliability of a historical interview with physically disabled respondents

Reliability of ‘The Motor Observations with regards to Sensory Integration’: A Pilot Study

The Motor Observations (MO) with regards to Sensory Integration’ (Horowitz 1994) is based on the original Clinical Observations of Ayres (1972a). The test aims to identify children aged 5 to 12 years with a learning disorder, described in this study as a developmental coordination disorder (DCD), hypothesised by Ayres to be due to sensory integration dysfunction. This study initiated the measurement of the reliability (internal consistency and intrarater and interrater reliability) of the MO, not included in its Dutch standardisation.

An observational correlational design was employed. The subjects were 28 children aged 6 to 10 years from the Sussex Weald. The MO was administered to a random sample of 21 children without DCD and a convenience sample of 7 children with DCD. Reliability coefficienss (Cronbach's alpha and intraclass correlation [ICC] random observers) were obtained via statistical analysis of score sets generated from live testing and its video recording.

The results indicate marginal reliability of the MO. Internal consistency (N=28) at 0.62 fell below the a priori level of 0.70. Intrarater and interrater reliability for the total samples (N=28, n=15) and the DCD samples (n=7, n=5) was high, ranging from 0.98 to 0.91, above the 0.75 a priori level. However, interrater reliability for the normal sample (n=10) fell below at 0.57, which suggests a deficiency in the test to identify normal development of sensory integration reliably. As the findings are not generalisable due to small samples, further studies are recommended.

A Generalizability Study of the Occupational Performance History Interview

In a previous project the Occupational Performance History Interview was developed and its reliability was studied. This paper reports the results of a study designed to examine the interview's reliability when used by therapists using one of two different frames of reference—model of human occupation and eclectic. Videotapes of 20 psychiatrically disabled subjects were rated by three registered occupational therapists within each frame of reference. Results indicated that the interview was only moderately stable under both conditions. Implications for further development of the interview are discussed.

Construct Validity of the Occupational Performance History Interview

The Occupational Performance History Interview (OPHI) (Kielhofner, Henry, & Walens, 1989) was developed to gather reliable, valid data about individuals' past and present occupational performances. The construct validity of the OPHI was assessed in a study evaluating the long-term impact of spinal cord injury on the lives of 143 people. Subjects' scores on the OPHI were correlated with their scores on the Center for Epidemiologic Studies Depression Scale (Radloff, 1977) and the Multidimensional Pain Inventory (Turk & Rudy, 1987). There was a significant negative correlation between the OPHI present and the Center for Epidemiologic Studies Depression Scale. Significant correlations in the expected directions also were obtained between the OPHI present scores and six of the Multidimensional Pain Inventory subscales. In addition, the OPHI past and present mean scores were significantly different, indicating that the instrument is sensitive to changes in occupational performance following traumatic spinal cord injury. These results support the construct validity of the OPHI, and therefore can increase therapists' confidence in using the instrument for diagnostic or therapeutic decision-making.