Assessment of the therapeutic alliance in physical rehabilitation: a RASCH analysis

Disabil Rehabil. 2012;34(3):257-66. doi: 10.3109/09638288.2011.606344.


Purpose: To investigate the psychometric properties of the Working Alliance Inventory Theory of Change Inventory (WATOCI) in physical rehabilitation.

Methods: RASCH analysis was used to examine the validity, internal consistency, reliability and discriminative properties of the 16-item WATOCI. Data was obtained from a randomized controlled trial, which employed the WATOCI to measure the alliance between physical therapists and patients with chronic low back pain.

Results: The results indicated that 7 items on the 16-item WATOCI should need to be improved for use in physiotherapy settings. The resulting 9-item version of the WATOCI was found to be a uni-dimensional tool for measuring alliance, however, demonstrated a ceiling effect.

Conclusions: There are several items in the WATOCI that require re-wording and/or re-contextualization prior to its use in clinical practice. Specifically, items relating more to the element social interaction did not discriminate well in this population and one solution may be to provide a therapy-specific context. Items regarding how well the patient understood the purpose of the treatment discriminated best between patients and this area is recommended for further development in the tool's assessment. It is recommended that future research address these issues prior to its use in clinical settings.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Aged, 80 and over
  • Australia
  • Chronic Disease
  • Female
  • Humans
  • Low Back Pain / rehabilitation*
  • Male
  • Middle Aged
  • Patient Compliance
  • Physical Therapists
  • Principal Component Analysis
  • Professional-Patient Relations*
  • Psychometrics / instrumentation*
  • Randomized Controlled Trials as Topic
  • Reproducibility of Results
  • Surveys and Questionnaires*
  • Treatment Outcome
  • Young Adult