RAND-University of Pittsburgh Health Institute

Translating Research into Practice (TRIP) II


The RUPHI-University of Pittsburgh CTSI TRIP Pilot Grant Program provides funding for preliminary projects that test the efficacy of moving an innovation (e.g., technology, guidelines, procedures, educational materials, etc.) into clinical practice.

Funding Period

August 15, 2009 - August 15, 2010


University of Pittsburgh Investigator RAND Investigator Title of Project
Rachel Hess, MD, MS Ron Hays, PhD Functional Assessment Screening Tablet: Patient-Reported Measures (FAST-PRM)
Gadi Wollstein, MD Claude Messan Setodji, PhD Evaluating the Glaucoma Diagnostic Skills of Non-Glaucoma Experts Using Optic Disc Photographs, Visual Fields, and Optic Imaging Devices
Robert E. Schoen, MD, MPH Ateev Mehrotra, MD, MPH, MS Using Natural Language Processing to Evaluate whether Colonoscopy Follow-up Recommendations are Consistent with Guidelines
Yuting Zhang, PhD Walid Gellad, MD, MPH Variation in the Use of $4 Generic Drugs among Medicare Beneficiaries
Jordan F. Karp, PhD Carrie Farmer Teh, PhD Preferences and Knowledge About Pain/Depression Interface and Treatment in Older Adults


Functional Assessment Screening Tablet: Patient-Reported Measures (FAST-PRM)

Rachel Hess, MD, MS
Assistant Professor of Medicine and Epidemiology
University of Pittsburgh School of Medicine
Center for Research on Health Care
230 McKee Place, Suite 600
Pittsburgh, PA 15213
412-692-2025 (office)
412-692-4838 (fax)

Ron Hays, PhD
Senior Health Policy Analyst, Social Policy Department, RAND, Santa Monica, CA
Professor of Medicine and Health Services Research
UCLA David Geffen School of Medicine
RAND Health Program.
1776 Main Street
Santa Monica, CA 90407-2138
310-393-0411, ext. 7581 (office)
310-393-4818 (fax)

Study Goals:

The Functional Assessment Screening Tablet (FAST): Patient-Reported Measures (PRM) project aims to increase doctor-patient discussions of physical activity, tobacco use, and health-related quality of life (HRQoL) in the primary care setting. The ultimate goal is to increase physical activity, reduce tobacco use, and improve health related quality of life.

Study Design and Methods:

All Division of General Internal Medicine patients currently complete the FAST. The FAST uses touch screen, tablet computers to collect, score, and report in real time PRMs (e.g., physical activity tobacco use, HRQoL using the RAND-36).

FAST-PRM uses a cluster-randomized design. Physicians are recruited and randomized to an intervention or control state. Patients are recruited to complete a survey examining the rates of discussions of PRMs during the doctor-patient encounter. A patient seeing an intervention physician receives a personalized feedback form based on her PRMs. The form places her physical activity level, tobacco use and HRQoL in the context of current guidelines and population normative data and encourages the patient to speak with her physician about these results. A patient seeing a control physician does not receive the personalized feedback form. All physicians continue to receive the current FAST report and will receive education regarding physical activity and smoking cessation and the use of the HRQoL measures.

We will compare the rates of patient-reported discussions of PRMs and, secondarily, changes in PRMs (e.g., HRQoL) between the intervention and control groups. We will assess physician perceptions of the impact of the intervention on the doctor-patient encounter.

Evaluating the Glaucoma Diagnostic Skills of Non-Glaucoma Experts Using Optic Disc Photographs, Visual Fields, and Optic Imaging Devices

Claude Messan Setodji, PhD
RAND Corporation
4570 Fifth Avenue, Suite 600
Pittsburgh, PA 15213

Gadi Wollstein, MD
University of Pittsburgh
939 Eye & Ear Institute 203 Lothrop Street
Pittsburgh, PA 15213


Glaucoma is the second leading cause of blindness worldwide. Diagnosis of this disease remains challenging, relying on the detection of structural changes in the eye accompanied by typical functional deficits. The purposes of this study are to identify diagnostic weaknesses of non-glaucoma expert eye care professionals (optometrists, trainees in ophthalmology, and general ophthalmologists) in comparison with glaucoma experts and to determine the added diagnostic benefit of visual fields (VF) and glaucoma imaging device (GID) data. Using a computer program, 5 glaucoma experts and 60 non-experts will evaluate a set of 100 stereoscopic optic nerve head photographs (SP). This set will include eyes of healthy subjects, glaucoma suspects and glaucoma subjects. Each participant will use a 4 points glaucoma risk scale and majority opinion of the experts will be used as the diagnostic gold-standard. When experts identify glaucomatous abnormalities, they will be prompted to identify the signs that led to their decision. This process will be repeated after sequentially providing sets of SPs accompanied by visual fields (VF) and glaucoma imaging device (GID; optical coherence tomography, confocal scanning laser ophthalmoscopy and scanning laser polarimetry) printouts. Reproducibility of the assessment will be determined by reevaluation of a subset of the data. Eyes that were often misdiagnosed by the non-experts will be evaluated according to the experts’ decision justification to identify common area of misdiagnosis. Information from this pilot study will be used to create a learning module tailored for non-glaucoma experts to enhance their capabilities in identifying glaucomatous changes.

Variation in the Use of $4 Generic Drugs among Medicare Beneficiaries

Yuting Zhang, PhD
University of Pittsburgh
Department of Health Policy and Management
130 De Soto Street
Crabtree Hall A664
Pittsburgh, PA 15261

Walid Gellad, MD, MPH
RAND Corporation 4570 Fifth Avenue, Suite 600
Pittsburgh, PA 15213

Generic drugs are bioequivalent to brand-name drugs and meet the same standards of the Food and Drug Administration for safety and effectiveness, but generic drugs cost 60-90% less than brand-name drugs. Since 2006, several retail pharmacies have made heavily-discounted generic drugs available to consumers at $4 per 30-days supply or $10 per 90-days supply. Use of $4 generic drugs could potentially save patients and payers billions of dollars, but no data exist on how the program is used by seniors, what factors affect the use of $4 generic drugs, and how the program might affect prescription drug use, spending, and adherence. The program represents an innovation in care whose efficacy has not yet been tested.

We will obtain national pharmacy data for a 5% random sample of 17.6 million Medicare beneficiaries enrolled in stand-alone Part D plans. We will determine the association between the use of $4 generic drugs and patient characteristics, geography of residence, and generosity of drug coverage. This data will form the basis for future proposals to study how effective these low cost generics are in improving adherence among Medicare beneficiaries compared to current benefit plans. Additionally, should we find different patterns in use of $4 generic drugs, we can design interventions to promote their use by targeting specific groups of patients, providers, or regions. Our pilot study would be the first to evaluate the pattern of use of $4 generic drugs, and our findings will serve as a foundation for successful future research proposals.

Preferences and Knowledge About Pain/Depression Interface and Treatment in Older Adults

Jordan F. Karp, PhD
Western Psychiatric Institute and Clinic/UPMC
3811 O’Hara Street, 766 Bellefield Towers
Pittsburgh, PA 15213

Carrie Farmer Teh, PhD
RAND Corporation 4570 Fifth Avenue, Suite 600
Pittsburgh, PA 15213

The proposed project responds to the RAND-University of Pittsburgh Health Institute and the University of Pittsburgh Clinical and Translational Science Institute pilot grant program: Translating Research Into Practice. Primary care patients living with chronic musculoskeletal pain comorbid with subsyndromal depression present unique treatment challenges. These patients are at high risk of conversion to major depressive disorder and exacerbation of pain-related disability. In addition, treatment non-adherence (associated with sub-optimal outcomes) may result if non-preferred treatments are prescribed. To reduce the risk of side effects associated with traditional analgesic medications, and address comorbid symptoms of depression, it is critical to know which non-analgesic treatments are preferred by both patients and PCPs. We propose to collect qualitative data about patient and PCP attitudes towards these linked conditions, and both patient and PCP preferences for non-analgesic management of chronic pain with subsyndromal depression. We will also explore the methods variability surrounding assessment of treatment preference for these patients. We will identify the two or three treatments which 1) are deemed most preferable by patients, and 2) have high levels of physician buy-in and acceptability. Our scientific approach will utilize a sequence of qualitative interviews with 10 PCPs and 20 older patients living with chronic pain and subsyndromal depression. This iterative approach will result in a rich dataset of how PCPs and patients approach treatment of these conditions and the best method to assess treatment preference. These preferred non-analgesic treatments will be tested in a comparative effectiveness study to be submitted as a R34 to NIMH.