People using assistive technology may not be able to fully access information in these files. For assistance, please contact us.
Structured Abstract
Background
This is a second-phase project sponsored by AHRQ and the National Cancer Institute through the DEcIDE program. Phase 1 developed a framework for measuring patient- centered communication (PCC) in cancer care, including identifying measurement domains and subdomains for the following six functions (Epstein & Street, 2007): exchanging information, fostering healing relationships, recognizing and responding to emotions, managing uncertainty, making decisions, and enabling patient self-management.
Objectives
The goal of Phase 2 was to refine the Phase 1 framework; identify, develop, and test PCC measures; and create surveys for patients and health care professionals. These surveys were intended to assess PCC in cancer care at the organizational and population levels.
Methods
Refine Measurement Model/Scientific Evaluation Group (SEG): Refined the PCC framework and mapped the existing domains/subdomains to the refined model. This process was guided by an independent SEG, and members had expertise in clinical cancer communication, survey design, cognitive testing, and health care delivery and quality of care.
Develop and Test Measures: We developed an inventory of PCC survey items by examining published literature and soliciting expert and public input. We then cataloged the existing items, identified gaps, and developed new items as appropriate. Finally, we pre-tested the survey items with cancer patients (n=46) in various settings who have diverse cancer types and are in various stages of care.
Create Surveys: We developed self-administered surveys for both patients and health care providers. The two patient surveys (a) monitored quality of care in health care organizations and (b) captured population-level surveillance. The provider survey assessed clinicians’ PCC activities.
Results
We retained the existing PCC framework as the foundation for developing the survey items. We identified relevant survey items from a total of 83 existing instruments and mapped them to the PCC domains. The process resulted in the creation of a pool of 1,316 survey items. However, 75 percent of these items did not meet our criteria for inclusion and were not retained. The number of items retained varied widely by domain area, requiring new items to be developed. During pre-testing, the items functioned well overall. The cancer patients who participated in the testing were generally able to understand the questions, apply them to their own experiences, and use the response scales to choose an appropriate answer. Many participants reflected that the items measured important aspects of their interactions with clinicians. We made numerous recommendations about how to revise and word the items.
Conclusions
This project lays the groundwork for the assessment of PCC in cancer care. We used a systematic approach to develop PCC items for a patient survey, beginning by inventorying existing survey and other measurement instruments and items. Subsequently, we developed and tested a total of 220 items, and ultimately finalized a set of 147 candidate PCC items. Additional steps are needed to finalize the items including additional cognitive testing, conducting a large-scale field test of the items followed by psychometric analysis, and finalizing the items and developing short and long-term versions of measures.
