By ZEESHAN SYED
Tens of millions of Americans rely on consumer experience apps to help them find the best new restaurant or the right hairdresser. But while relying on customer opinion might make sense for figuring out where to get dinner tonight, when it comes to picking which doctor is best for you, AI might be more trustworthy than the wisdom of the crowd.
Consumer apps provide us with rich data categories that often take into account preferences, from location to free wi-fi, to help users narrow down choices. Navigating your health insurer’s network of physicians is a different proposition, and some of the popular ranking systems reportedly have significant limitations. Doctors are often categorized by specialty, insurance, hospital, or location, which may be effective for logistics, but fail to take into account a patient’s unique health conditions and say very little about what an individual patient can expect in terms of health outcomes. Research from my company Health at Scale shows that 83 percent of Medicare patients seeking cardiology care and 88 percent of cases seeking orthopedic care may not be choosing providers that are highly rated for best predicted outcomes based on each patient’s individual health conditions.
Deep personalization is exactly what physicians, health systems, and insurers need to offer patients to improve outcomes and lower costs across the board. A study using our data recently published in the Journal of Medical Internet Research sought to quantify how consumer, quality and volume metrics may be associated with outcomes. Researchers analyzed data from 4,192 Medicare fee-for-service beneficiaries undergoing elective hip replacements between 2013-2018 in the greater Chicago area, comparing post-procedure hospitalization rate, emergency department visits, and total costs of care at hospitals ranked highly by popular consumer ratings systems and CMS star ratings as well as those ranked highly by a machine intelligence algorithm for personalized provider navigation.
The results showed that patients treated by hospitals ranked highly by the machine intelligence-based algorithm experienced better health outcomes and lower total costs of care than those treated in hospitals rated highly by the other approaches. Not only did machine intelligence outperform the field on all three metrics, but in some cases the hospitals ranked highly by other approaches had worse outcomes.
The machine intelligence algorithm employed here solves a problem long believed to be intractable: modeling how physician outcomes vary from patient to patient across a broad set of health factors. Using anonymized health record data from over a hundred million lives in the U.S., the machine intelligence algorithm constructs a detailed profile for each provider in a health insurance network and their history of optimal outcomes with specific patient profiles relative to one another. The model uses this information and a richly detailed profile of a patient to create a personalized ranking of providers for the patient. Using a nationwide dataset enables rigorous evaluation of the model across specialties and geographies, ensuring that the model is as accurate for assisting a heart patient in Houston as it is for the hip patient in Chicago. In short, by developing highly detailed profiles of both provider and patient, machine intelligence can apply big data solutions to a small data problem.
So what does all of this mean? The results show that relying on general, sometimes arbitrary metrics may be of limited utility when considering provider options relative to a personalized and outcomes-based approach. If insurers or care managers employ more precise machine intelligence tools to inform these patient decisions, they may take a step closer to care that is highly personalized and highly effective, based on selecting the right physicians based on each patient’s unique medical needs. Yet there is still room to grow: just 26% of patients in the study attended the hospital that machine intelligence determined was top rated for them.
To improve the health care system for patients, care managers and insurers need to use the best decision-making tools to guide their search for care, focusing on technologies that account for the health variables that make each patient unique and providing suggestions that prioritize measurable health outcomes. Machine intelligence is proving its ability to make care navigation simple and precise, demonstrating that we can make selecting a doctor both less like a drudge through the phonebook and more reliable than advice from strangers on an app.
Zeeshan Syed, CEO of Health at Scale, was a Clinical Associate Professor at Stanford Medicine and an Associate Professor with Tenure in Computer Science at the University of Michigan.