Skip Navigation
Grant Details

Grant Number: 5R21CA124913-02 Interpret this number
Primary Investigator: Shields, Cleveland
Organization: Purdue University
Project Title: Racial Differences in Physician-Patient Communication for Cancer Pain Management
Fiscal Year: 2008
Back to top


DESCRIPTION (provided by applicant): BACKGROUND: Racial disparities exist in both the health and healthcare of Americans. We propose to assess patient-physician communication and indirectly assess physician bias through analysis of audio and videotapes of physician visits with unannounced standardized patients (SPs) who are randomly assigned to the physician on the basis of the SP race. SPECIFIC AIMS: (1) train black and white SPs with high degree of fidelity (2) to ensure SP realism in role and (3) to develop estimates for effect sizes for difference by race in physician prescribing for pain control and in physician-patient communication. METHODS: We will randomly assign participating physicians to unannounced black or white SPs. The SPs will be trained to portray lung cancer patients with severe pain who are new to the physicians' practice. Office visits will be recorded using hidden cameras. We will compare visits between white and black SPs in terms of communication patterns, medical decision-making, and non-verbal behavior. PARTICIPANTS: Forty oncologists and primary care physicians in the local community will be recruited to participate. MEASURES: We will measure physician dosage and use of long acting pain medications, thoroughness of pain assessment, patient-centered communication, and autonomy support. SIGNIFICANCE: This study will advance research methodology, inform communication theory, and guide future research to improve communication and eliminate disparities in care. Use of SP methodology to study bias is an innovative adaptation of a methodology proven in other contexts. SP methodology will allow us to address limitations of prior studies such as confounding by patient behavior, illness severity, insurance status and symptoms. No previous study has directly observed physician behavior when interacting with white and African-American patients discussing requests for pain medication or any other request of which we are aware.

Back to top