DESCRIPTION (provided by applicant): The U.S. healthcare environment has changed dramatically in the last several decades. Patients are increasingly knowledge-empowered (through internet access) and "activated" (through direct advertising) to seek care from physicians who are increasingly salaried employees in large medical settings (adhering to organizational priorities). "Activated patients" often suggest specific diagnoses to their primary care physicians. Some 30 percent now request specific medications (discovered on television or by internet searches) of which 40 percent receive the requested prescription. Using complementary research methods (factorial experimentation and rigorous qualitative research) we propose to: a) describe how different primary care physicians respond to specific prescription requests from different patients (factorial experiment); and b) explain the reasons why they do so ("think aloud" qualitative techniques). Within a single, cost-efficient study, we propose two experiments focusing on common medical conditions (sciatica and osteoarthritis of the knee) which generate high levels of healthcare utilization, prescribing and costs, to address the following specific aims: 1. To estimate the independent influence of patient attributes on a physician's compliance with a request for a pain medication. Is a request from certain patients more likely to be successful? 2. To estimate the independent influence of provider characteristics on the diagnosis and management of the two pain conditions presented. Are certain physicians more likely to accede to a patient's request? 3. To understand the influence of healthcare system or organizational factors on prescribing and management. Does the size, ownership, practice setting or practice culture influence decisions? 4. A complementary qualitative component will identify the underlying cognitive reasoning processes that explain the observed variability in decisions uncovered by the experimental component of the study. Our findings will have implications for everyday clinical practice, healthcare policy and educational interventions designed to improve the quality of decision making. First, variability in the quality of care provided by primary care physicians for commonly presented pain is of widespread national concern. Second, the influence of patient requests on provider behavior remains poorly understood, but is likely to increase in the U.S. and elsewhere (DTCA of pharmaceuticals is about to be introduced throughout the countries of the European Union). Third, polypharmacy (especially pain medications) is of increasing concern in the US, particularly among older patients with multiple co-morbidities who often consult a range of specialists in addition to primary care providers. Fourth, moving from the description of healthcare variations (how?) to explanation of their cognitive origins (why?) marks a new direction in clinical decision making research and is a necessary pre-requisite for educational interventions. Public Health Relevance: We propose to use a factorial experiment to understand Clinical Decision Making (CDM) when activated patients make requests for a specific pain medication. The implications for a continual and rapid increase of direct to consumer advertising (by Internet, TV, and print), broadening health disparities in chronic pain management, and growing prevalence of chronic pain has significant health policy implications. This study aims to disentangle the patient (gender, race, SES) and physician factors (experience and gender) related to chronic pain management and opioid use, and underlying cognitive reasoning that produce the decisions we observe.

7. PROJECT NARRATIVE We propose to use a factorial experiment to understand Clinical Decision Making (CDM) when activated patients make requests for a specific pain medication. The implications for a continual and rapid increase of direct to consumer advertising (by Internet, TV, and print), broadening health disparities in chronic pain management, and growing prevalence of chronic pain has significant health policy implications. This study aims to disentangle the patient (gender, race, SES) and physician factors (experience and gender) related to chronic pain management and opioid use, and underlying cognitive reasoning that produce the decisions we observe.