Symcat: Data-Driven Symptom Checker

Problem

symcat_logo_simple_900x260When people get sick, they have several options for obtaining health care. These include going to the emergency room, urgent care center, or calling a doctor or nurse. However, 80% of people experiencing symptoms start with an Internet search. Unfortunately, searching on Google offers spotty results and frequently leads to undue concern. For example, one is 1000x more likely to encounter “brain tumor” in web search results for “headache” than they are to ever have the disease. Undue concern is a contributor to the 40% of emergency room visits and 70% of physician visits that are considered to be inappropriate.

In many cases it is clinically appropriate to pursue care at any of several types of facilities. One can choose, for example, to get a flu shot at a pediatrician’s symcatoffice or a retail clinic. Patients frequently have little insight into the cost or quality of the different options available to them and must make these decisions blindly.

Existing solutions have failed to overcome these challenges. The average physician is overwhelmed with 24 phone calls per day, forcing him or her to provide hasty advice. Nurse advice lines, while theoretically able to alleviate this problem, have not been widely adopted. Meanwhile, emergency rooms are more crowded than ever as they become the simplest option for confused healthcare consumers. There is a clear need for a scalable, reliable, trusted information source that helps patients decide where and when to seek care.

Solution

Symcat is a free web and mobile service that helps consumers figure out what to do when they get sick. Users start by entering their symptoms and demographic information. The Symcat algorithm asks additional questions and matches users to the most relevant, high-quality provider based on their profile. Providers may, for example, be at physician offices, retail clinics, urgent care centers, or emergency rooms.

In order to do this, Symcat analyzes a large amount of data from the Centers for Disease Control and Prevention (CDC). This is nationally representative, visit-level medical record data. Each piece of information the user inputs (eg a symptom, demographic information, symcat bmjetc.) updates the likely conditions and triage routes for that user. Instead of seeing the possibilities that arise in a medical textbook and being told simply “go see a doctor if you are concerned,” users see what o
ther people with their clinical profile actually had and did.

Though it is not a true diagnostic service, Symcat more accurately points users to
relevant conditions using this data. Recent analysis by Harvard Medical School researchers placed Symcat at the top of a list of 23 symptom checkers for diagnostic accuracy and a 42% improvement over WebMD.

Challenges

  • Creating a unique ontology that maps medical concepts from ICD9/10, symcat graphSNOMED, RFV and more to easy-to-understand consumer health concepts
  • Combining over 40+ datasets and performing Bayesian analysis of likely diagnoses
  • User-centered design of a web and mobile clinical triage experience
  • Exploring the market among providers, employers, and insurers for more effective, patient-guided triage

Honors

  • Winner of $100k from the Robert Wood Johnson Foundation Aligning Forces for Quality Developer Challenge
  • Winner of $25k from the Cigna Health Innovation Challenge
  • Venture-backed
  • Selected among over 300 teams to participate in the Blueprint Health startup accelerator
  • See Awards and Honors

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s