PCORnet Bariatric Surgery Study Generating Real-World Data to Support Informed Decisions

Lead PI: Dr. David Arterburn, MD, MPH

SCILHS PI: Dr. Ali Tavakkoli, MBBS

This study will examine three common types of bariatric surgery currently performed on America’s severely obese population: Roux-en-y gastric bypass, adjustable gastric banding, and sleeve gastrectomy. The goal of this study is two-fold: (i) to identify the long-term outcomes of the three different procedures, and (ii) to formally test and evaluate the PCORnet’s capacity to efficiently and collaboratively produce values for observational research. The patient outcomes in this research will be measured based on several criteria, such as weight change following surgery, changes to diabetic status, and complications or further harm to patients in the years following surgery. There is also an important secondary aim of this project. The researchers are hoping to gain valuable patient input through focus groups, with the intention of investigating patient preference in terms the overall decision to undergo surgery, optimal follow-up care, and the procedure selection process. This study would provide answers with tremendous implications for America’s growing obesity epidemic, as well as answers to the functionality and readiness of the complex collaborative data system that is PCORnet.



PCORnet Observational Study: Short and Long-Term Effects of Antibiotics on Childhood Growth

PI: Jason Block, MD

This study aims to evaluate the comparative effects of different types, timing, and amount of antibiotics prescribed in the first two years of life on growth trajectories and risk of obesity later in childhood. There has been evidence of a link between childhood obesity and antibiotic use, but it is widely debated and the relationship to dosing, type and age is not well understood. It will also examine how these effects differ across demographics, as well as in relation to other medication use and maternal health measures. The goal of this study is to contribute to the understanding of antibiotic use at an early stage of life and its possible long-term effects. The secondary aim of this study involves in-depth patient and clinician interviews to assess how much this information should change clinical and parental decision-making practices. It will utilize focus groups and interviews to better determine how caregivers assess risk and how information can be disseminated responsibly to help parents share in clinical decision-making. This study will also include an assessment of the use of the PCORnet infrastructure and its readiness to support such noteworthy research in the future.



Health Systems PCORnet Demonstration Project: Variation in Case Management Programs and their Effectiveness in Managing High-Risk Patients for Medicare ACOs

PI: Maureen Smith, MD, PhD, MPH

This health systems demonstration project will answer important questions regarding best care practices for a particularly at-risk population. The focus of this project will be on older patients with multiple conditions. As these patients often require care from many physicians with complex coordination, there are several challenges to ensuring proper follow-up. Many hospitals nationwide have implemented case management programs to aid in this process, but the methodologies of these programs are quite varied in scope, and the metrics of efficacy have yet to be evaluated. The most important elements of follow-up case management remain poorly understood, in addition to speculation over which patients benefit the most from such management. In conjunction with partnering hospitals of the Greater Plains Collaborative, SCILHS seeks to represent Massachusetts General Hospital in the quest for answers. The aims of this study are two-fold: (i) to gather an understanding of the management programs currently operating in the 22 health systems comprising the SCILHS and GPC networks, and (ii) to examine the role that these systems play on reducing healthcare utilization for these patients, paying particular attention to hospital stays, observational stays, and emergency room visits. This second aim will focus particular attention to patients enrolled in Medicare accountable care organizations (ACO’s). In alignment with PCORI’s patient-driven focus, patients and clinicians will play an important role at each step of this venture, including design, interpretation, and invaluable input on how to best disseminate the generated results.



Refining and Validating an Electronic Suicide Prevention (ESP) Tool for Health Systems and Clinical Settings

PI: Jordan W. Smoller, MD, ScD

The purpose of this study is further the development of an Electronic Suicide Prevention Tool. As it stands presently, this module is able to adequately predict suicide risk months to years in advance. This instrument is not yet developed for practical purposes like clinical intervention in a real-world setting. Continuation would also allow development for predictors within a group of patients who have an increased risk of suicidal event at baseline, namely those under psychiatric care. The study also aims to further efforts to develop a prototype risk communication tool capable of alerting physicians when high-risk patients are identified. The final goal of this project would then be to validate these refined models in a diverse array of healthcare practices, thus demonstrating the wide clinical scope and utility of such models. The implementation of such a healthcare model given current suicide rates and diagnostic challenges could have remarkable ramifications, placing care into the hands of some of America’s most vulnerable populations.



SCILHS Osteoarthritis Study

PI: Daniel Solomon, MD, MPH

The Scalable Collaborative Infrastructure for a Learning Healthcare System (SCILHS) project, funded in 2014 by the Patient Centered Outcomes Research Institute (PCORI), assembled clinical data research network of health care sites across the United States, with the goals of establishing a new paradigm for timely, efficient, and meaningful identification and engagement of patients for patient-centered clinical research studies. As part of this project SCILHS designed a system called mySCILHS to collect patient-reported outcomes (PROs) directly from patients that could easily be connected back and added to their medical record information. In creating this system, our goal was to make research easier and more accessible to patients across our network. The focus for this use-case was a common ailment, knee osteoarthritis (OA).  Three SCILHS sites participated in this study, Partners HealthCare (Boston, MA), Boston Medical Center (Boston, MA) and Wake Forest Baptist Medical Center (Winston-Salem, NC).

After receiving a letter signed by their doctor, patients were contacted by SCILHS staff to take a 26 question survey that asked questions about a) their demographics, b) their opinions on their willingness to participate in research and c) their quality of life and symptom burden. Participants were given the following choices of how they preferred to take the survey: 1) in-person in-clinic recruitment for paper based survey, 2) live person telephone interview, 3) interactive voice response via telephone, 4) web-based survey, or 5) paper survey mailed to their address. Providing the research participant with many options was possible in part due to the mySCILHS platform, a patient-facing semi-automated data collection tool.

The majority of the group (N = 134, 79%) opted to take the survey by live-telephone with an interviewer administering the questionnaire, followed by 19% (N=32) choosing to take the web-survey version of the questionnaire. Three individuals (2%) opted to have the survey mailed to them and one individual (0.5%) used the interactive voice response (IVR) option. The overall response rate for the knee OA study was 40%.

Our advancement in gathering patient-reported outcomes for the osteoarthritis study provided a use-case for this technology to allow easier data collection. The mySCILHS platform allows for securely collected patient-reported outcomes to be paired back with a patient’s data from the electronic medical record, with the patient holding the ultimate control over how and when they participate. The results of this survey demonstrate a path toward innovation in patient-reported outcomes research and future PCORnet endeavors. SCILHS would like to thank all the patients who participated in the study which helped improve the mySCILHS technology.


SCILHS Pulmonary Arterial Hypertension Study

PI’s: Aaron Waxman, MD, PhD, and Mary Mullen, MD, PhD

In conjunction with Phase I of SCILHS, Pulmonary Arterial Hypertension (PAH) was chosen as the rare disorder clinical condition. With the goal of testing SCILHS’ capabilities in mind, a feasibility study was conducted to query patients with PAH across four newly-constructed i2b2-SHRINE platforms. The goal was to test the use of ICD9 codes to generate aggregate patient data within these sites. The development of such precision was promising for SCILHS’ scope and future utility, theorizing that this same process could be performed (and improved) for several different disease phenotypes moving forward. The second mission of this cohort development was to collect patient-reported outcomes. This was done over the phone and through use of mySCILHS patient facing technology. An interactive voice recognition interface was developed by the SCILHS team to reduce time and labor in collecting patient responses. Data analysis is underway. During these surveys, patients were also asked whether they would like to be contacted for future projects, thus developing an active registry of patients if any opportunity to engage them should present itself. Having the ability to contact patients by phone to collect patient-reported outcomes will allow for researchers and clinicians to assess symptom burden of disease and effectiveness of associated treatments. Additionally, since pulmonary arterial hypertension is a rare disease, having an easy and effective way to engage and recruit patients in research is essential. The Pulmonary Arterial Hypertension Feasibility component of Phase I proved to be a learning experience and gave important insights into the capabilities of SCILHS.



SURE-PD3 Trial Recruitment Initiative

PI: Michael Schwartzchild, MD, PhD

SCILHS is currently aiding in recruitment for a Phase III clinical trial of urate-elevating inosine treatment to slow clinical decline in early Parkinson’s disease (PD). This study aims to consent about 700 patients with early Parkinson’s disease that have not yet received drug intervention. Patients are being recruited from 60 sites across the country. The wide scope of this trial, in addition to the small proportion of Parkinson’s patients who are eligible, create particular challenges for recruitment. As such, investigators have enlisted SCILHS technology to aid in identifying a patient cohort for this study. The strategy will facilitate enrollment of newly diagnosed PD patients who may not be referred for clinical trials, including minorities who have lower rates of referral for specialist care and PD trials, and accelerate referrals that might otherwise take so long that it renders the candidate ineligible.  The SCILHS network is one of the multifaceted recruitment techniques SURE-PD3 is implementing, and it is our hope that the information provided furthers their important efforts.


Submit a Project to SCILHS