Winner Announced for MedCredits $10,000 Competition
We received a total of 31 submissions in the MedCredits App Competition!
Our team carefully reviewed all 31 outstanding submissions, with each team member rating each essay independently with a score of 1–5. The scores were then summed for each essay, and the top 3 highest rated submissions were sent to Kyle Samani with Multicoin Capital to select the final winner of the $10,000 reward.
The top 3 submissions are as follows:
- “Automated Triage” by Trevor Smith, MD
- “Peer-to-Peer Messaging” by Amir Ghorbani
- “Reducing Patient Readmission with Arbitrated Quests” by Robert and Mohan Tanniru
The WINNER is Trevor Smith, MD with his submission “Automated Triage.” Dr. Smith outlines an internal patient referral system that enhances quality of patient care through a streamlined triage approach that also increases adoption and retention of medical specialists/sub-specialists within the MedCredits Ecosystem. Replicated below is an excerpt from Kyle Samani’s evaluation of “Automated Triage”:
It makes perfect sense to me that a user should be able to upload an image, chief complaint, maybe some basic info to a triaging service that many types of disparate physicians are plugged into. This also is a great app of your physician registry. It seems technically feasible. The key thing is going to be ensuring data is encrypted at all times. My guess is you’re going to need something like Nucypher for that.
We would like to extend a BIG THANK YOU for all those who participated in this competition as well as Kyle Samani for his contribution as the expert judge.
When we launched this contest a month ago, the goal was threefold. Firstly, we wanted to inspire the smart and innovative people in the cryptoasset and blockchain community to explore the potential of the MedCredits health ecosystem. Secondly, we want to work more closely with the community in developing the future of global health. After all, everyone is a patient! Finally, we wanted to provide an opportunity to reward the community for their support!
As we received a number of great ideas, we look forward to discussing any of these in our social media channels! So be sure to follow us on Twitter and join us on Telegram!
With the permission of the authors, we have replicated the top 3 submissions below.
Automated Triage
By Trevor Smith, MD
Executive Summary
Using a new method of organization, patients will be directed to specific physicians based on their chief complaint, or reason for seeking care, within an automated system to connect them to the appropriate physician specialist. Simultaneously, MedCredits can collect data on physician patterns of referral to be used to further enhance the internal automation and as industry insight to monetize via consultation with the greater healthcare industry.
The Problem
Directed care is the future of medicine. The increasingly complex landscape of the business of medicine and the siloing of medical knowledge has resulted in dramatically subspecialized groups. Some areas, such as neurology, have even developed “hyperspecialists” that treat a single disease process (ex. demyelination: multiple sclerosis, NMO, Guillain barre). Though a doctor may find it to be a natural process to direct a patient to various specialists based off a primary complaint alone (and more so with additional symptoms/signs), it is often impossible for a patient to do so, even with WebMD or Google at their side. Patients do not have the knowledge or experience to know who to see for their medical concern. Additionally, they often do not have the understanding of the medical system to know which physicians to turn to, what order to see them in, or what timing will be most efficient. The value of a solution is therefore very obvious to the patient: identifying pertinent positive and negative features of a clinical history can save them significant time, money, and a disproportionate amount of frustration.
The Solution and its Benefits to MedCredits
Appropriate referrals. When a doctor signs up to see patients and explore the platform they should be set up for success. Directed care allows new physicians to immediately set to work at diagnosing and treating while retaining subspecialists along the way. This will result in increased growth and diversity of medical specialties on the physician token-curated registry (TCR).
Encourage internal referrals. The current method of TCR out-voting may inadvertently result in penalizing onboarded doctors attempting to treat outside their training rather than rewarding them for referring to the appropriate specialty within MedCredits. Instead of risking incorrect diagnoses and having their tokens appropriately voted away, doctors may move off the platform entirely. Alternatively, rewarding them for referrals and collecting the patterns of those referrals would enhance the automated directed care system within MedCredits. Avoiding physician attrition early in the process cannot be emphasized enough. Physicians love to talk negatively about their bad experiences with new technology and they will never forget a company that takes their money for just trying a new method of treating patients.
Faster response. A patient simply wants to get their questions answered and get better as soon as possible. An alert sent to the most appropriate physicians increases patient satisfaction.
This could be implemented using a “Suggested Specialist” component of the consult request form.
Based on your submission, we recommend a dermatologist. They will be the best doctor for your concerns.
Offer a group consultation. A patient could find out right away if they need multiple consultants to diagnose their issue. MedCredits can offer them treatment or assessment by (1) the most likely specialist, (2) a generalist who can direct further for more complex issues, or (3) send them to a group assessment for a higher fee:
Your chronic cough may benefit from assessment by several specialties: ENT, pulmonology, infectious disease, and allergy. Get a group opinion all at once at a reduced rate to rapidly figure out the cause and treat that nagging (chronic) cough!
Leverage available information to build an initial database for patient direction. Doctors will show you how they recommend redirecting consults. However, the many diagnostic websites (WebMD) can be screened for repetition of words or phrases as a starting point for the automated system and later refined by the internal referrals. An application of AI may be helpful here, but at this point a simple word recognition system would be a great place to start and is implementable early on in the platform without significant programing expertise required. In the future, AI and ML developers (i.e. Algorithmia.com) will find a growing market on the MedCredits ecosystem to develop and deploy trustless machine learning contracts to enhance automated triage algorithms. This is significant value for the time and effort.
Use the data. If automated triage and an internal referral system become a successful component of the platform, the data could be used to create value for local, centralized healthcare systems. A local University Hospital, for example, may consider a component of non-concierge medicine that still allows an elevated tier of access and directed care, both to increased access to expert physicians and to save the patient time by sending them to the appropriate subspecialty. They would benefit from consultation with MedCredits regarding our patterns of practice system developed within the MedCredits system.
Opportunities
Automated triage will grow the physician TCR by opening doors and increasing retention of specialists and subspecialists. In addition, the usability of the Hippocrates application will improve to allow patients to quickly and confidently submit healthcare request and concerns. Finally, the referral patterns and the decentralized ontology of healthcare complaints will create a rich and fertile data analytics marketplace for developers, influencers and leaders at all levels of the healthcare ecosystem.
Peer-to-Peer Messaging
By Amir Ghorbani
As of 2017, 80% of healthcare systems in the world still use an antiquated paging system for physician-to-physician communication[1]. The most widely used hospital pagers these days are produced by Spok and come programmed with 32 unique keys that can encrypt and decrypt ciphertext using the standard AES-128* algorithm. Messages are sent over the trusted Spok network. The Spok server stores a complete set of encryption keys for each pager and relays all incoming pages. The quoted price for a standard T5 pager is $25 and for $22.95/month a physician gets 500 pages. Spok provides basic free software for sending pages and allows group pages to be sent out for urgent situations such as strokes, heart attacks and urgent airway responses. As the costs are fairly reasonable and most hospital IT departments have invested significant resources into their paging systems, the desire to switch models has been fairly low.
Nevertheless, over the past several years, the physician-to-physician messaging market share has finally been giving way to mobile apps that offer secure, HIPAA-compliant messaging services. Most notable is TigerText which offers mobile and web “messaging as a service” for $10.65/month. TigerText utilizes the SSL protocol to establish secure data channels between the sender, the TigerText server and the recipient. The advantages over pagers include: requiring only 1 device, direct text responses, image support, and more efficient use of physician time.
MedCredits’ business-model is vastly different from traditional companies. The MEDX token fuels a token-curated registry of physicians. In this cryptoeconomic model, the MEDX token increases in price so long as more physicians join the network. Additionally, according to the MedCredits whitepaper, MEDX owners earn a portion of all revenue that is sent to physicians in the registry. Essentially, MedCredits can offer software for free so long as it gets physician to join the registry and/or have payments sent to physicians for various services.
MedCredits needs a “killer app” that can initially attract large numbers of physicians to apply to the registry. I propose that physician-to-physician messaging can be a honey-pot to get physicians to apply to the registry. Physicians would pay a 1-time refundable deposit to enter the registry (this fee also covers them for all other MedCredits applications) and subsequent messaging would be free. This type of application also sidesteps any regulatory issues that MedCredits may encounter when attempting to provide medical services to actual patients.
From a technical standpoint, the most basic implementation would use the public-key cryptography already native to Ethereum which uses an secp256k1 elliptic curve. A physician in the registry would send a message by encrypting it with the recipient physicians’ public key. This offers greater security than Spok and TigerText since messages would be end-to-end encrypted and stored on the local device. While, to-date, Spok and TigerText servers have not been breached, recent notable hacks have demonstrated that even the biggest industry players are vulnerable. Additionally, MedCredits’ messaging implementation would eliminate the need for a trusted third party completely. This has implications beyond libertarian philosophies because, under HIPAA requirements, all physicians using centralized messaging services must sign a Business Association Agreement (BAA) with the service provider. Because MedCredits’ messaging services would be truly peer-to-peer, there would be no such restriction. This would potentially allow MedCredits to scale their messaging platform with greater ease and potentially corner the market quickly. Finally, it is important to note that this messaging feature would be global and therefore casts a wide net to capture market share.
Though public-key cryptography has existed since the 1970s, Ethereum-type technology is now putting public/private key-pairs into the hands of millions of users. MedCredits’ token-curated registry is the first ever platform that verifiably identifies physicians by public keys. The physician-to-physician messaging industry is only now showing signs of disruption and MedCredits can outperform the competition in all fronts including cost, security and accessibility. Additionally, this feature has the added benefit of being generic enough that it could attract any physician from anywhere in the world. This would surely provide long-term benefits to future MedCredits applications as there would be immediate exposure to a pool of physicians already frequently engaging with MedCredits multiple times a day.
[1] https://www.journalofhospitalmedicine.com/jhospmed/article/141692/hospital-medicine/hospital-based-clinicians-use-technology-patient-care?utm_source=News_JHM_eNL_071917&utm_medium=email&utm_content=Hospital-level%20factors%20associated%20with%20pediatric%20emergency%20department%20return%20visits
Reducing Patient Readmission with Arbitrated Quests
By Robert Tanniru and Mohan Tanniru
Market Opportunity
Ever since the introduction of Affordable Care Act (ACA), hospitals have struggled to avoid reimbursement costs when patients return to the hospital within 30 days after discharge. To reduce these costs, hospitals have been improving their discharge planning activities such as communication of discharge care related information and patient education. Today, hospitals are spending fixed resources (e.g. staffing and technology costs) to call chronically ill patients to remind them of activities in which they need to engage (visit labs or pharmacies, schedule follow-up visits with primary care physicians, etc.), or use remote monitoring technologies to monitor patient’s vital signs, send alerts, and support consultations. This high fixed-cost hospital-directed centralized system can be ineffective in engaging patients and multiple care providers, who have varying degrees of motivation and technological maturity, to share activity information to address continuity of patient care and reduce unanticipated patient readmissions.
New Business Model
An unanticipated readmission of a cardiac patient can cost a hospital over $10k, and an average readmission of a patient with Sepsis can double their average length of stay to over 10 days and cost over $25k. The average readmission costs vary based on the diagnosis (DRG) code anywhere from $4k to $34k, with an average of $13k based on 2012 CMS (Center for Medicaid and Medicare — https://cms.gov) Data. What is needed is a trusted and distributed platform that allows a number of stakeholders (patients and external care providers) to become aware of post discharge care activities that need to be completed, incentivize these stakeholders to complete the activities, and provide hospitals flexibility in the resources they need to expend to address patient readmissions.
Many patients readmitted tend to be older, have chronic conditions (e.g. heart failure, diabetes, hypertension, etc.), and face various physical, economic and emotional challenges. Many external care providers have to spend varying degrees of effort to get reimbursed for services, as they come from multiple parties (patients, insurance providers, and government). A new business model involving incentivising patients to perform specific activities that have been proven to significantly reduce the probability of patient readmission can greatly decrease the costs beared by the hospital.
Proposal
We propose a blockchain-based gamification of the tasks a patients must perform (hereafter referred to as Quests) that encourages both patients and caregivers to complete these Quests. Given that many of these patients may not be technically savvy, this system is designed to allow patients to benefit from the rewards for completing Quests without having to directly interact with the technical details of the Quest as those interactions are performed by the physician and third party Arbitrators.
The hospital or physician will create various Quests that a patient should complete post discharge. The Arbitrator is the service provider (i.e. pharmacy, testing lab, etc.), and the Quest taker is the patient. The patients receive a reward (either in ETH or a ERC20 token such as MEDX) when the task is complete. The amount of this reward will vary depending on factors such as potential costs of readmission and likelihood that completion of this Quest will avoid readmission of this patient. The Arbitrator also receives a portion of the reward amount for participating in the Quest and their reward percentage can also be adjusted due to market factors. This Arbitrator’s public Ethereum address will be verified using the MedCredits Token-Curated Registry to verify their role (pharmacist, specialist, x-ray technician, etc.).
Benefits of using MedCredits Token-Curated Registry
MedCredits Token-Curated Registry (TCR) provides a decentralized method for verifying the credentials of a given Ethereum public address. It’s game theoretical design incentivises those who are verified by the TCR to act honorably and fulfill their role correctly to avoid negative consequences. In the Quest example, these Arbitrators will correctly verify that the patient completed the details of the Quest they were assigned before the Arbitrator sends the quest completion transaction. The MedCredits TCR is an essential piece of this process and provides security against exploits to the Quest system.
Ethereum Ecosystem
The Quest will be coordinated using the Ethereum blockchain and executed using Smart Contracts (sample code here: https://github.com/ImAllInNow/ethereum-quests). Ethereum is chosen due to its large user-base and continued development of applications to make tracking and spending coins easy for the end-user. The reward the patient and Arbitrator will receive upon completion of the Quest are written into an Ethereum smart contract. After the completion of each Quest, the Arbitrator’s role is validated using the TCR, and a reward is provided immediately. After a certain timeout, if provided, the task expires and the reward can be returned to the Quest maker via a quest failed transaction. All Quests will provide a specific role of Arbitrator that is allowed to arbitrate that Quest.
Technicals and Administration of the Quests
● Step 1: The Quest maker (physician) sits down with a patient and lists a number of tasks they need to perform (distinct tasks such as “pick up a prescription”, “get blood tests done”, “visit a rehab facility 5 times”, etc.). Each task is viewed as a separate Quest and have a specific Arbitrator role that will be validated by the TCR.
● Step 2. The physician agrees to put in some digital currency (ETH or MEDX) as a reward if the task is completed (with most of the reward going to the patient and the remainder going to the Arbitrator).
● Step 3. Each Quest is entered as a signed transaction to the smart contract on the blockchain by the physician.
● Step 4. The patient goes to the Quest Arbitrator to complete the task. A mobile/web app could be developed to help them find Arbitrators in the area and keep track of quests to perform.
● Step 5. The Arbitrator verifies that they have completed the Quest and submits a transaction indicating that the Quest is completed.
● Step 6. The reward is distributed to each party (Arbitrator and patient) automatically from the smart contract.
