How to test

To simulate logging in with your health insurance, use one of the credentials from flexpa here. Example Humana user has username HUser00001 and password PW00001!

Also check out our landing page in development:


Our team has all been impacted by the opioid epidemic in various ways, and we knew that if we were going to build something new that would make a difference we would have to focus on preventing opioid addiction in the first place. After conversations with experts in the field, data from the CMS, extensive literature review, and books like The Least of Us by Sam Quinones, we felt that the best place to start was with people who were recently prescribed opioids. Luckily for us, Flexpa recently built the infrastructure to make it possible to know when a patient has been prescribed an opioid as well as what other medications they're currently taking.

What it does

PainPal is both a patient education and monitoring platform and patient-provider communication tool. It leverages evidence-based risk prediction surveys and data analysis to offer real-time opioid addiction, misuse, and overdose predictions for patients and providers. It meets patients where they are with alternative ways to manage their pain, and incentivizes them with a cash rewards system that we plan to fund through partnerships with health plans interested in reducing cost and improving population health by preventing opioid addiction.

How we built it

Full-stack typescript (React on the frontend and Node.js on the backend), with a python stack for data analysis.

Challenges we ran into

  • Converting our python risk prediction analysis into something our API could consume

Accomplishments that we're proud of

  • Original video content to educate patients on opioid risks and alternatives

  • Implementing an evidence-based survey to predict future opioid misuse/addiction risk as well as overdose risk due to non-misuse. Future research will validate these measures in a study using patient outcomes to provide an accurately weighted assessment of risk.

What we learned

  • We learned during our extensive literature review that the components of our app are validated by peer reviewed research (e.g., the efficacy of patient education in mitigating risk).
  • Communication is key! Due to the ambitious nature of the project, we were each focused on our own deliverables and realized we could have had better alignment between our team members. There was a lack of coordination between the backend and analysis team, who discovered too late that they would have needed to code in Javascript to make our developer workflow more efficient.

What's next for PainPal

Next on our product roadmap is a feature allowing us to recommend that the patient do a trial of their prescription when the pain level and interference is rated mild (below 5 on a 10 point scale, using measures from the Brief Pain Inventory) for two consecutive days, followed by a prompt to turn them in to a safe disposal site. Seventy percent of opioid users obtained their most recent opioids peripherally from the leftover supply of a family member or friend. Research shows that patient education about disposal is highly effective, so encouraging patients to turn prescriptions in when they are no longer needed helps mitigate risk.

Additionally, adverse reactions to opioids will be tracked to alert both patients and physicians of increased overdose risk. Recommendations of alternative non-opioid therapies will be validated and provided to patients through educational materials, with the goal of implementing both synchronous and asynchronous classes providing personalized instruction that patients can complete at home, such as specific physical therapy exercises for their pain-related conditions.

Our pilot study will review survey data, insurance claims data, EHR data, zip code demographics, and patient outcomes, and compare the efficacy of different approaches to predicting risk and assigning personalized care.


Over 600,000 North Americans died of opioid overdose in the past two decades, and without corrective action, another 1.2 million are projected to die by 2029. Around 9 million patients are currently estimated to be misusing opioids, and 80% of heroin users began drug misuse with a prescription opioid.

Typical efforts to tackle the opioid crisis, like Prescription Drug Monitoring Programs, which attempt to limit drug access after the patient is already addicted, are too far downstream. Furthermore, research shows that primary care physicians often underestimate patient risk of opioid abuse. And efforts to change subscribing behavior have had limited effect.

Research also shows that patient education programs do reduce risk of developing opioid addiction, but implementation of such programs has been sporadic at best, and no coordinated effort has been made to electronically deliver patient education and monitor its effectiveness.

PainPal is a data collection and educational tool designed to intercept addiction before it starts. We begin by pulling relevant insurance claims data to inform us of each patient’s addiction and overdose risk. After signing up, patients are greeted with a welcome video explaining the purpose of PainPal and what to expect. They are then presented with a short survey to more accurately assess their risk of opioid addiction and overdose.

Daily videos guide the patient on how to manage pain and seek alternative pain management, while at the same time making patients more aware of opioid risks and side effects. Patients are provided with tracking of their pain rating and risk of addiction over time.

Short daily surveys are used to monitor progress over time. The physician also has access to the patient’s progress, and is alerted both on overdose risk, and risk of addiction.

Ideally, PainPal would integrate with Medicaid, insurance companies or PDMPs. The patient would receive a text as soon as a prescription is written, prompting sign up. Patients could be offered reward points or small cash incentives points for completing each daily task. The cash incentive model is already used successfully by Anthem Health via their partner app Sydney Health to encourage patients to get preventative care.

Here’s a little bit about our team: Joey Caffrey is a full-stack software engineer at Bold, a fitness app for older adults. His family was personally impacted by the opioid epidemic. Asna Ansari is a data scientist at Bold. Sofiane Oumaouche is a backend software engineer at Snapchat. Lauren MacIntyre, a neuroscientist specializing in addiction research, aspires to be an entrepreneur in the health and wellness space. Holly Bittinger is a marketing and communications professional at SentinelOne, a cybersecurity company. She recently lost a close family friend to heroin addiction.

Several of our team members were personally impacted by the opioid crisis. We’re grateful to Datavant for bringing this team together and for inspiring us to work on this project.

Policy changes will only go so far. If we want to beat the opioid epidemic, we need to start with effective patient education. At PainPal, we are committed to empowering patients to make smarter, healthier decisions.

+ 7 more
Share this project: