You are using an outdated browser. Please upgrade your browser to improve your experience.

You are here

ACP Innovation Challenge

The ACP Innovation Challenge 2018 is celebrating its third year showcasing the best and brightest ideas from practicing physicians, fellows, residents and medical students that have the potential to transform the practice of internal medicine on a national level. One winner was chosen among four finalists who presented to a panel of experienced physicians at the Internal Medicine Meeting 2018.


2018 Winner and Finalists

GRAND PRIZE WINNER Caleb Murphy
MD/MBA Candidate, University of Minnesota Medical School Twin Cities

High-Value Technology: Developing Electronic Medical Record Capabilities to Promote High-Value Clinical Practice

The high value care movement has taken major strides over the past decade, exemplifed by national initiatives to reduce healthcare spending. One initiative receiving particular attention is the push to reduce unnecessary laboratory testing in the hospital. The American Board of Internal Medicine “Choosing Wisely” list for Hospital Medicine encourages physicians to avoid “repetitive CBC and chemistry testing in the face of clinical and lab stability.” Many high value care groups have sought to improve lab ordering practices. However, beyond the idea of “clinical and lab stability,” there have been few attempts to define with specific criteria what it means for a test to be “unnecessary.”

The high value care team I work with at the University of Minnesota has been developing a solution to this problem. With support from an American Board of Internal Medicine grant, we developed the Minnesota Lab Appropriateness (MLAB) criteria to identify inappropriate basic metabolic panels and complete blood counts. Developed with expert input, including from members of the Alliance for Academic Internal Medicine High Value Care workgroup, the MLAB criteria are specific and translatable into “yes-no” rules. We presented the criteria at the 2017 Society of General Internal Medicine annual meeting and will soon submit a manuscript detailing the development and high inter-rater reliability of the criteria. The MLAB criteria provide perhaps the best available method for identifying unnecessary BMPs and CBCs.

An important next step in exploring the utility of the MLAB criteria is to integrate them with the electronic medical record. By doing this, the criteria could be applied retrospectively to assess the historic rate of inappropriate lab orders of a given physician or department and, more importantly, as a prospective decision-support tool to inform physicians of the appropriateness of potential lab orders, thereby combining patient care and high-value care promotion into a single, time efficient act. To study the performance of the criteria in these scenarios, we need to work with EPIC programmers at our institution to develop Report (retrospective) and Best Practice Advisory (prospective) tools. Funding will be needed to facilitate this process.

An EMR tool to identify inappropriate BMPs and CBCs would reimagine the practice of internal medicine by empowering clinicians to make high-value ordering decisions in the hospital at the point of care. We believe our team has the necessary range of skills and knowledge to realize this proposal.

AUDIENCE CHOICE WINNERSumun Khetpal
MD Candidate, Yale University School of Medicine

Ride Health - Driving Access to Care

Each year, about 3.6 million low-income, elderly and disabled patients miss about 24 million appointments due to insufficient transportation access. In addition to lower health status, these missed appointments lead to $40 billion in avoidable downstream healthcare costs. These no-show appointments represent approximately $4.1 billion in lost revenue for doctors.  Medicaid, Medicare and healthcare providers spend $6 billion each year on NEMT services, but the current system fails patients and providers. Logisticare, the largest incumbent for Medicaid transportation, has an obscene track record of missed pickups and drop offs, and cab vouchers disrupt clinical workflow and are quite expensive relative to ride-sharing services.

With healthcare reimbursement trends moving toward value-based payments, providers have an increasing interest in getting patients in the door before problems arise. Ride Health accelerates the NEMT transformation and re-invents the process of coordinating transportation for such patients.  The product is a web-based, software-as-a-service (SaaS) platform that allows coordinators to schedule patient rides in advance, automatically requests rides prior to appointments, sends text message updates and reminders to patients throughout the process, and gives coordinators real-time insight on transportation spend and usage. The Company will charge providers a monthly subscription fee to access the platform. Rides are fulfilled by transportation vendors and funded by the healthcare provider or insurer, depending on patient eligibility.

Ride Health has completed development of its first product and has deployed it with a Top 10 U.S. News & World Report Hospital, an ACO with more than 1,100 providers and 83,000 members, the largest healthcare provider in southern New Jersey, and a hospital system in the greater Philadelphia area. The pipeline represents more than 38 health systems across seven states, comprising of over 160 hospitals.

Jin Sol Lee
MD/MPH Candidate, David Geffen School of Medicine at UCLA

Fruitful Health: A Smartphone App to Personalize Healthcare for Every Individual

Every year, $300 billion of US healthcare dollars are lost as a result of medication non-adherence. As a result, numerous interventions have been attempted including those that have increased patient education, follow-up, face to face time with physicians to address non-adherence and reduced copayments. Despite these efforts, approximately 50% of patients were still non-adherent to medications in 2013.

We believe the most important contributing factor to non-adherence is due to individual-level barriers that cannot be effectively addressed with one blanket solution. Thus, interventions should be patient centered. However, bridging patient and provider shared decision making has become increasingly difficult with growing administrative burdens. Thus, if we equip both patients and providers with technology that reduce workload while delivering the same quality of information necessary to practice patient-centered medicine, we can change the practice of internal medicine.

As a result, we developed a user-friendly decision aid smartphone application tool called Fruitful Health. Our mobile health platform allows patients to utilize non-value added time in the waiting rooms of clinics to support patient centered care. Our clients engage in self-care in two distinct ways. First, before they see their doctor they complete two tasks: 1) identify, prioritize, and limit their visit specific health needs to three items that will be discussed with their provider; and 2) create a rank order list of preferences that impact their decision-making process (e.g. cost, side effects, time, etc.). Fruitful Health visually displays these individualized, visit-specific patient care aids to their physician promoting efficient and effective communication. Second, Fruitful Health follows up with our clients by 1) asking patients to document if their visit needs were addressed and if not, will save and auto-populate for their next clinic visit; 2) recommending patients to reflect on and commit to one health behavior change; and 3) recommending patients to create a To Do List that must be completed prior to their next clinic visit. Additionally, Fruitful Health will utilize gamification to encourage our patients to 1) utilize our product regularly and 2) track patient adherence to their own self-reported goals (e.g. To Do List, Commitment) along with objective clinical outcomes (e.g. A1C). In other words, patients will be rewarded with virtual points which can be cashed out for prizes that will be geared towards healthy investments such as gym memberships.

Susan Zelasko
MD/PhD Candidate, University of Wisconsin, Madison School of Medicine and Public Health

Detecting Catheter-Associated Biofilms at the Bedside

Just as the advent of pulse oximetry enabled bedside monitoring of a patient’s oxygen saturation, a detector of catheter related bloodstream infections (CLABSIs) could change how we address the care of patients with such infections. Central venous catheters (CVC) serve as a direct port into a patient’s bloodstream for easy clinical access and are indispensable in the care of gravely and chronically ill patients. Unfortunately, CVCs also provide a common route for hospital-acquired infections.  Each year, CLABSIs result in an estimated 250,000 infections, with an attributable mortality rate near 18% [1]. Treatment of such infections is becoming increasingly difficult due to the precipitous rise in antimicrobial drug resistance. Moreover, CLABSIs are associated with prolonged hospital stays and an additional $32,000 in healthcare costs per infection [2].

Currently our group is exploring an innovative approach to develop a reusable medical device that would assist in the early detection of CVC contamination. Microbial communities growing on contaminated CVCs develop into biofilms capable of releasing planktonic cells into the patient’s bloodstream. Mature biofilms become encased in extracellular matrix that offers protection against antimicrobial agents, making their eradication challenging. Our device will be mounted on an external segment of the catheter and will detect biofilm formation through spatial and temporal changes in the ultrasound frequency within the lumen of the catheter. We hypothesize that biofilm formation will alter the fluid mechanics of the catheter in a manner that can be detected, as normalized to the catheter on insertion, via ultrasound technology.

Early detection and subsequent early treatment of CLABSIs could be highly beneficial for the treatment of patients with CVC borne bloodstream infections. At present, the initial detection of CLABSIs relies on the identification of non- specific symptoms in catheterized patients, followed by extensive testing before care can be initiated. The advantage of an external monitor used to alert healthcare staff to the development of a biofilm may prove crucial in initiating care earlier than is currently possible.


[1] Umscheid CA et al. Infect Control Hosp Epidemiol. 2011 Feb;32(2):101-114.

[2] Stevens V et al. Clin Microbiol Infect. 2014 May;20(5):O318-24.