3 stories to watch in health tech in 2020

Judging by the numbers, digital health companies had another huge year in 2019.

Total investment in the sector was expected to reach a record $8.4 billion, as firms raked in cash to develop novel wearables, personalized digital services, and machine learning tools to improve treatment of debilitating diseases. Several companies also entered the public markets, and tech giants such as Apple (APPL), Amazon (AMZN), Google (GOOGL), and Microsoft (MSFT) ramped up their health investments and recruitment of clinical leaders.

But much of that activity is fueled by hype and hope. Heading into the new year, digital health companies will face more pressure — and potential collapse — if they don’t deliver results for their clients and financial backers. In the political realm, many of the biggest players will also be forced to address privacy issues and concerns about bias in algorithms used to deliver, or restrict, health services for tens of millions of Americans.

Here are three stories to watch in 2020, a year in which the dream of seamless digitized care may slam into some cold clinical realities.

Can health tech make real money?

Despite all the excitement and investment surrounding digital health, there’s a big hurdle the field has yet to surmount: that tricky business of making substantial money.

Consider the big tech companies that are aggressively building out health businesses. They’ve been investing in projects like building an algorithm to detect diabetic retinopathy and sponsoring virtual observational health studies — endeavors that are ambitious and impressive, but effectively guaranteed not to bring in meaningful money in the near or even medium term. So: How long is their runway? Big tech companies didn’t become some of the most valuable businesses in the world because they allowed unprofitable lines of business to burn money. At some point, these companies are going to want to start seeing financial returns. One potential way they might do that? By placing more and more emphasis on their boring but lucrative cloud businesses, pitching data storage and analysis to life sciences companies and hospitals.

Similar challenges face the health tech startups looking to scale up and brave the public markets. Sure, digital health executives will eagerly tell you they see a path to profitability. But they’ll never become successful public companies in the long term if they can’t start turning meaningful profits. And to do that, they’ll have to convince their customers — whether health systems, insurers, employers, or even consumers — that their products can provide durable and lasting benefits.

As health tech entrepreneur Neal Khosla put it in a recent post on Medium: “Let’s face it: digital health has been one of the most disappointing investment areas of the last two decades.” It’s safe to expect that there will be increased pressure for that to start changing in 2020.

Who owns these data anyway?

Big technology companies are bulls in a china shop of sensitive personal health information. The question facing them and their hospital partners in 2020 is whether they will ignore the sound of shattering trust or try harder to restore it.

Their dilemma was highlighted recently by the controversy surrounding a data sharing deal between Google and the hospital chain Ascension in which the tech giant gained access to millions of records without notifying patients or getting explicit consent. The parties defended the arrangement by asserting that it is meant to improve care for patients and complies with the federal privacy law known as HIPAA.

Federal regulators are now investigating whether that is the case. But some privacy advocates argue that the question, while relevant to the law on the books, is beside the point in the broader world in which these companies are operating. After all, they say, HIPAA was passed in 1996 at a time when the rise of massive multinational companies like Google, Amazon, and Facebook (FB) was not yet contemplated.

It’s anyone’s guess whether Congress will act on proposals to update HIPAA or replace it with stricter controls on information sharing. But a safer prediction for 2020 is this: Large companies and hospitals trading on patient data will continue to face harsh criticism unless they create more transparency around information sharing and involve patients in their decision making.

Will bias be battled?

Ignoring bias in health care algorithms will no longer be possible in 2020.

As uptake of artificial intelligence increases, the watchdogs are watching closely. One research group recently published a paper concluding that an algorithm developed by Optum to target care to needy patients was biased against black people. After receiving several more inquiries, the researchers unveiled a project to eliminate bias in health care algorithms at the University of Chicago’s Booth School of Business. And in December, a first-ever workshop on algorithmic fairness was held at the 2019 Conference on Neural Information Processing Systems (NeurIPS), one of the world’s most influential gatherings of machine learning specialists.

The simple truth is this: Any company that fails to audit an algorithm for bias before deploying it in health care is courting disaster. That’s because the same machine-learning tools that power these algorithms can also be used to check their work. “Health care bias is now very quantifiable,” said Irene Chen, a Ph.D. student at the Massachusetts Institute of Technology who helped organize the fairness workshop at NeurIPS. She added that health tech companies are also acutely aware of the consequences of inaction: “The companies I’ve talked to are very careful and they want to know what’s going on,” Chen said. “No one wants to be called in front of a Senate hearing to answer the question, ‘What’s going on with your algorithm?’ ”

3 stories to watch in health tech in 2020