Determining the Psychometric Properties of a Bank of Items Related to an Individual's Psychological Condition

The disturbing prevalence of depression-related suicide reveals the life-and-death importance of proper measurement in detecting symptoms of depression and associated somatization. Additionally, the high incidence of depression among those with impairments to the spinal region make it imperative that lumbar spine patients are assessed using valid and reliable tools. Depression, along with related symptoms of somatization, are treatable, however, and 60-80% respond well to treatment. If a patient’s symptoms associated with depression are accurately assessed and evaluated, potentially life-saving treatment can occur. Therefore, the correct identification, assessment, and evaluation of depression and somatization cannot be over-emphasized, especially in the context of occupational therapy. This insight into a client’s condition is especially relevant within the practice of occupational therapy. Occupational therapists should be aware of the mental health of their clients in order to facilitate proper treatment planning and appropriate interventions. Equipped with this understanding, occupational therapists may better provide holistic care for their patients.

The purpose of this study is to examine the psychometric properties of the items on the SCL-90-R. Specifically, we will examine the item-level characteristics of the SCL-90-R using Rasch analysis and factor analysis, including item difficulty measures, person ability measures, item fit/dimensionality, and item-person match. We hypothesize that the SCL-90-R items will represent two constructs: depression and somatization. Separated into two groups of items, one representing items measuring depression and one representing items measuring somatization, we hypothesize that these two item banks will have excellent item characteristics, hierarchy of difficulty will be logical, and items will be appropriate for the sample of interest (individuals with lumbar spine disorders). Furthermore, based on identified item characteristics, we will create data collection forms to assist occupational therapists in goal setting and treatment planning. Data collection forms created will provide optimal guidance for clinicians, motivating increased utilization of the SCL-90-R.

DESIGN: This study involves analysis of survey question data that has been previously collected by the Focus On Therapeutic Outcomes, Inc. (FOTO) (Knoxville, TN, USA). FOTO is a web based patient assessment system that reports functional status measures. It is used nationwide in 3,400 clinics by 15,000 clinicians. FOTO will provide us with de-identified data from patients.

METHOD: This project will: 1) use item response methodologies, specifically Rasch analysis, to link person ability and item difficulty, 2) using the link between person ability and item difficulty create data collection forms allowing for easier goal setting and treatment planning, and 3) examine various other characteristics of the items in the bank (e.g., person separation).

RESULTS: Preliminary results indicate that the item bank does contain items that represent two distinct constructs but not with exactly the same items as proposed by the creator of the assessment. When divided into these two constructs, preliminary analyses suggest that these two item banks will have excellent item characteristics, hierarchy of difficulty will be logical, and items will be appropriate for the sample of interest (individuals with lumbar spine disorders). All analyses will be completed by August 2018.

CONCLUSION: Accurate assessment and consideration of psychological factors is important in rehabilitation settings. Additionally, linking person ability to item difficulty will enable data collection forms to be created to guide goal setting and treatment planning.

References

Depression and suicide risk. (2014). American Association of Suicidology. Retrieved from https://www.suicidology.org/portals/14/docs/resources/factsheets/2011/depressionsuicide2014.pdf

Lehman, L., Woodbury, M., Shechtman, O., Wang, Y., Pomeranz, J., Gray, D., & Velozo, C. (2011). Development of an item bank for a computerised adaptive test of upper-extremity function. Disability and Rehabilitation, 13(1). doi:10.3109/09638288.2011.560336

Li, C-Y., Velozo, C., Hong, I., Li, C., Newman, J., & Krause, J. (2017). Generating Rasch-based activity of daily living measures from the spinal cord injury longitudinal aging study. Spinal Cord, 8(1). doi:10.1038/sc.2017.99

Wright, B. D. (1967). Sample-free test calibration and person measurement. ETS Invitational Conference on Testing Problems. Retrieved from https://www.rasch.org/memo1.htm