As the biggest search engine, Google has its fingers in a lot of healthcare pies.

But among the biggest priorities for Google in healthcare are measuring search results to better understand chronic disease and, from an advertiser standpoint, how to respond more immediately to changing consumer healthcare behavior.

Along the way, Google also learned the search engine can’t take on every single healthcare problem entirely on its own. “We were probably too arrogant in the beginning,” says Aviva Shwaid, Google U.S. healthcare lead. “We had big, audacious goals.”

Shwaid, who spoke Friday in suburban Chicago at an event sponsored by KPMG—Seismic Shifts: The Innovation of Life Sciences—told attendees that of Google’s earliest attempts in healthcare was understanding the top 10 trends consumers search the most. “We took a lens to understand the top trends,” she says. “The most searched trend was depression and anxiety, followed by opioids.”

The analysis reveals that geography and the timing of natural disasters was a major contributing factor to searches for depression and anxiety. Searched terms on the opioid crisis peaked in October following an exposé on the topic by the 60 Minutes news program. “This helped us learn to pivot how we message around pain killers,” she says.

Google currently has several projects underway in healthcare. For example, its life sciences and healthcare company, Verily, is working with Duke University School of Medicine and Stanford Medicine on Project Baseline study, which is collecting health data from approximately 10,000 participants, who will each be followed over the course of at least four years. The study will try to better understand the transition from health to disease and identify additional risk factors for disease. Beyond this initial study, Project Baseline will test and develop new tools and technologies to access, organize and activate health information.

The Project Baseline data repository is built on Google’s computing infrastructure and hosted on its cloud platform. “We are working with partners to dig into the real-time data,” Shwaid says. Since 2016, Verily also has been working with Nikon Corp to apply data and machine learning for better screening of the conditions leading to blindness tied to diabetes and provide eye care specialists tools for easier diagnosis of disease. “We are trying to understand images by analyzing scans of eyes at risk because of diabetes,” Shwaid says. “The idea is to predict risk.”

For advertisers, especially for big drug makers, Google is working with pharmaceutical companies to create proactive versus reactive digital marketing programs by better understanding search patterns. For example, Google worked with one pharmaceutical company that makes acne medicine to develop an app with features that enabled teenagers with acne to take complexion “selfies,”  post the images every couple of weeks on a timeline and compare how their appearance improves over time using the medication.

“We learned how to tap into behavior by analyzing real-time behavior,” Shwaid says. “The app is a good example of a pharma company that helped solve a problem with adherence since lots of teenagers try acne medications, don’t see results in a few weeks and quit.”

Keep up with latest coverage on digital healthcare by signing up for Internet Health Management News today.