Background: Patient-reported experience measures in cancer care ideally capture hospital performance in six domains advocated by the Institute of Medicine (IOM):
The Patient-reported Experience-Cancer (PRE-C) instrument was developed to capture these variables and inform responsive cancer service delivery.
Aims: To determine PRE-C reliability (α≥0.7); to establish PRE-C convergent and divergent validity.
Method: The PRE-C was developed with cancer patients and clinicians. Face and content validity were tested with 30 patients and 10 clinicians. Items were revised according to their feedback until a 28-item questionnaire was ready for psychometric testing. A sample size of 280 was needed to determine reliability.
Exploratory factor analysis (EFA) examined the PRE-C’s dimensionality. Maximum likelihood was used for extraction, with direct oblimin rotation examining correlation between constructs.
Results: 414 consecutively-recruited ambulatory chemotherapy patients from a single tertiary hospital participated. EFA indicated that 6 of the 28 items were problematic—some fitted better in other domains and some did not measure as intended. Particularly problematic were items that explored respect for cultural preferences and the financial impacts of treatment. Removing these items and re-running the EFA indicated reliability (α in the six domains ranging from 0.73 to 0.81).
Discussion: The revised PRE-C instrument reliably and accurately captured all IOM constructs. It is also important that the PRE-C captures the financial and cultural experiences of cancer patients. To this end, we have revised the problematic items and are currently testing the PRE-C with ambulatory patients in three large health services, anticipating N>280 for a new EFA and >300 for subsequent confirmatory factor analysis. We are also preparing the instrument for digital embedment to enable routine, responsive data capture.