U-M family medicine at the Society of Behavioral Medicine conference in March

Mixed methods research provides the opportunity to expand the breadth and depth of understanding of a phenomenon of interest. For example, the qualitative data provide information on the motivations, perceptions, and expectations of the variables of interest while the quantitative data characterizes general relationships among variables of interest. Although the application of mixed methods is unique for each specific clinical and translational research question, certain common study design paradigms are transferable. In the mixed methods research design field there are familiar "core" designs used in investigations, such as the convergent design and the explanatory sequential design. New ways to embed mixed methods are continually evolving, and recent publications indicate interest in embedding core mixed methods designs into translational research approaches, such as community-based participatory research (CBPR) and dissemination and implementation (D&I) research. Through a combination of lectures and experiential small group learning, this 2-part pre-course covers the fundamentals of embedding mixed methods core designs into translational research approaches (Part 1) and an applied overview of strategies to obtain NIH funding for translational research approaches that have embedded core mixed methods designs (Part 2). Part 1 of this 2-part Mixed Methods pre-course will have a strong methodologic focus: Dr. John Creswell will introduce the familiar “core” mixed methods designs or procedures, including the convergent design, the explanatory sequential design, and the exploratory sequential design. In addition, Dr. Creswell will present the cutting-edge approaches that allow investigators to apply these core mixed methods designs to CBPR and D&I research studies. The organizers have designed Part 1 of this pre-course to provide an introduction to mixed methods research, and also to highlight new, innovative ways to extend mixed methods into CBPR and D&I research. Part 2 of this pre-course will have a strong applied focus towards developing fundable translational research proposals that utilize mixed methods designs. While the material presented will transcend discipline, we will focus on CBPR and D&I research approaches to prevent and manage two different disease states: cancer and diabetes mellitus. NIH program officers will describe strategic approaches to obtaining funding in this field, including reviewing the R-award program announcements that often fund CBPR and D&I mixed methods work. We will also include case examples of funded K- and R-award application in CBPR or D&I mixed methods research. Finally, we will hold small group breakout sessions led by NIH K-funded and R-funded mixed methods researchers in CBPR and D&I research. In separate breakout sessions for CBPR and D&I research, respectively, attendees will discuss how to develop their mixed methods research ideas to obtain funding. This 6-hour pre-course includes both Part 1 and Part 2 with a 30-minute luncheon incorporated into the session. Participants may also choose to sign up separately for only Part 1 of the pre-course (2 hours and 45 minutes) or Part 2 of the pre-course (2 hours and 45 minutes). 

You have successfully completed your doctoral and/or postdoctoral training and have started or are starting your first (tenure-track) faculty position. Now what? This course provides targeted career development and mentoring on key issues that are of interest to junior faculty, including 1) juggling research, teaching and service obligations, 2) learning about and utilizing the resources around you, 3) identifying formal and informal mentors and mentorship opportunities, 4) planning for promotion and tenure, 5) work/life flow versus balance, and much more. The course format will include brief introductions from our expert panel (30 minutes), open question and answer from attendees (30 minutes), and three breakout roundtables on key issues that are of interest to junior faculty with panelists and other mid-career and senior experts (90 minutes). Attendees will have the opportunity to ask questions of the panelists, network with nationally recognized leaders in the fields of public health, physical activity, psychology, health disparities, cancer and chronic disease prevention, and diabetes, and develop peer support and mentoring relationships. Attendees will walk away with skills to navigate their new academic environments and to overcome challenges and succeed in academia.

With the advent of population-based health, successful translation and implementation of behavioral medicine interventions will require a clear economic understanding of healthcare delivery. Although most chronic conditions require a coordinated treatment team, healthcare is often fragmented, redundant, and inefficient. The fee-for-service payment model is a significant obstacle to improving healthcare delivery because it only reimburses for the quantity of service, not its quality, outcome, or efficiency.

This panel will provide in-depth examination of the “Integrated Practice Unit” (IPU; Porter and Lee, 2013) model for improving illness management, outcomes, quality, and costs for virtually any chronic condition. In this model, treatment units are organized around specific patient problems rather than clinical specialties.

Although many existing multidisciplinary clinical teams are structured this way, few are true IPUs. A fully-developed IPU also focuses on systematically improving care quality, coordinating patients’ transitions across inpatient and outpatient settings, and monitoring key clinical outcomes and treatment costs. Importantly, its clinical objectives are to address quality and outcomes rather than RVU generation, and its business objectives are to target improved outcomes, efficiency, and growth. Effective IPUs then expand to new geographic locations and use their data to establish their bundled care rates for defined populations.

Following an overview of the general IPU model, panel members will present real-world exemplars of how IPUs are being implemented to address diabetes, chronic pain, and primary care. The panel will also discuss how researchers can use IPUs’ existing data to increase the utility of clinical trials, advance the most effective treatment models, approach population care from an economically-informed perspective, and advocate for improved healthcare policies in a language understood by administrators and insurers.

Physical inactivity is associated with higher health care utilization and costs among adults with diabetes. Pedometer-based walking interventions have been shown to be effective in increasing physical activity for diabetics, but the effect on health care utilization and costs is unclear. This study examines how an incentivized pedometer-based walking intervention impacts both physical activity and health care costs for adults with diabetes.

In 2009, Blue Cross Blue Care Network enrollees with a body mass index (BMI) of 30 kg/m2 or greater were given a choice to join an internet-mediated pedometer-based walking program (Walkingspree). Participation was financially incentivized because individuals could potentially save an estimated 20% of out-of-pocket expenses by uploading their step count data at least once every 30 days to the program website and averaging at least 5000 daily steps every three months. The pedometer assessed both total daily steps and aerobic steps, measured as continuous steps for at least 10 minutes. A multiple linear regression analysis determined the impact of average daily steps and aerobic steps on the change in health care costs after one year, controlling for diabetes status, age, gender, body mass index, length of program participation and other comorbidities.

Of the 7,594 WalkingSpree participants, 15 percent had diabetes. On average, every 100 daily steps were associated with a $9.07 decrease in health care costs (p=0.049). Among participants with a mean of 5000 daily steps, predictive margins show the average person without diabetes experienced cost savings of -$441.93 (p=0.004). In comparison, those with diabetes averaged increased costs ($872.67, p < 0.001), particularly for those with diabetes with complications ($2491.88, p < 0.001). However, these costs decreased for every 100 daily steps.

Greater daily steps are associated with decreased total health care costs for all three groups, but there was a nonlinear differential effect by diabetes status. Further investigation is needed to better understand the relationship between step counts and health care costs for diabetics, especially those with more serious complications. Although people with diabetes may incur greater health care costs, daily steps may help to slow these increased costs over time.

Learning Objectives:

• Upon completion, participant will be able to discuss the potential role of group-based health behavior change programs in managing health care costs.
• Upon completion, participant will be able to describe an internet-mediated walking program.
• Upon completion, participant will be able to describe the association between daily steps and health care costs by diabetes status.

Paper presenter:  Mona AuYoung, Ph.D., M.S., M.P.H., Staff Scientist, Sr. at the Scripps Research Institute/ Scripps Translational Science Institute

There is a significant evidence base supporting the effectiveness of the Diabetes Prevention Program (DPP), a lifestyle management program to reduce incidence of diabetes, which has been translated to multiple settings. Online adaptations of the DPP may enhance reach by allowing individuals to participate in DPP at the time and location of their choosing. However, most programs have been limited to face-to-face or phone-based modes of delivery. Thus, the goal of this study was to assess the effectiveness of an online version of the DPP among Veterans with prediabetes.

Eligible Veterans with prediabetes were invited to enroll in an online DPP, which was added to an ongoing trial of two in-person programs as an additional parallel, non-randomized arm. The analytic sample included consented patients who had 12-month weight outcomes available (via wireless scales or clinical administrative data). Mixed effects models were used to assess 12-month weight change, controlling for demographic factors.

Based on preliminary analyses, on average, online DPP participants (n=268) lost 6.25 lbs (p < 0.001) from baseline to 12 months. This was comparable to the average weight loss of 7.5 lbs (p < 0.001) among participants of the in-person DPP that was part of the ongoing trial within the VA. These preliminary findings suggest that online delivery of DPP is feasible and that results may be comparable to a group-based in-person program for Veterans with prediabetes.

Learning Objectives:

• Upon completion, participants will be able to describe the importance of reach when delivering lifestyle interventions.
• Upon completion, participants will be able to describe the relative advantages of delivering online lifestyle interventions as compared to in-person programs.
• Upon completion, participants will be able to describe the challenges of delivering online lifestyle interventions and areas for future research initiatives.

Paper presenter:  Tannaz Moin, M.D., M.B.A., M.S.H.S., assistant professor at the VA Greater Los Angeles Healthcare System and the David Geffen UCLA School of Medicine