Feature

Location matters—and not just in real estate. Psychologists have long known that people's surroundings can influence their moods and mental health, and that mental health services are unevenly distributed in communities.

"Mental health is a moving target," says Courtney L. McLaughlin, PhD, director of the school psychology PhD program at Indiana University of Pennsylvania, who studies the intersection of geography and psychology.

All sorts of small interactions can influence a person's mood over the course of a day, she says. And larger events, such as school shootings or natural disasters, can drastically affect the mental health of an entire region, she says.

With so many moving parts, there hasn't been a realistic way to map the space-time-psychology continuum. Now, that's changing as new geolocation technologies—including geographic information systems (GIS) software and global positioning system (GPS) receivers in mobile phones and wearable sensors—allow researchers to track people and populations with precision.

McLaughlin has begun using the ArcGIS GeoEvent Processor, a program developed by the GIS software company Esri, which maps geographical data in real time. The software has been deployed in other industries, such as tracking whether emergency response vehicles take the most direct and efficient routes, for example.

In a pilot project McLaughlin presented at APA's 2016 Annual Convention in August, she coded the GeoEvent Processor to scan geotagged Twitter posts worldwide for the keywords "depressed" and "feeling depressed." Though Twitter allows users to search only 1 percent of tweets, she had 130,000 hits in 36 hours—approximately one every four seconds.

Eventually, she hopes to map mental health trends in particular regions at particular moments in time—data that could reveal how such events as school shootings affect a region's collective mental health, or identify communities with a spike in suicidal thinking.

"The beauty of social media is that it gives us a lens into how groups of people are feeling in real time," she says. "If we could access those patterns, it would give us a pulse on how groups of people are feeling."

Personal sensing

Other psychologists are exploring location data from smartphones with a different goal in mind: understanding how people's movements might signal depression or other mental health concerns. David Mohr, PhD, who directs the Center for Behavioral Intervention Technologies at Northwestern University, and colleagues have analyzed GPS sensor data from mobile phones to identify behavioral markers of depression. In a pilot project, they collected sensor data from undergraduate students for two weeks and found that certain movement patterns were more likely among patients with depression. Those patterns included more irregularity in their circadian rhythms, less variability in the places they went, and a tendency to spend the majority of their time in a smaller number of places (Journal of Medical Internet Research, 2015).

In a follow-up study, they examined data from a different population of students collected over a 10-week period, and found the same relationships between depression and the way students moved through their days (PeerJ, 2016). "This suggests mobility markers, measured by GPS, could be an early warning predictor of depression," Mohr says.

"Personal sensing," as Mohr calls it, involves more than just location. By tracking a person's position in time, researchers can gather information about aspects of movement such as the variability of locations, time spent in locations and how rhythmic the movement is. Going a step further, those data can be combined with other sensor data such as exposure to ambient light, green space, noise, air pollution and many other factors that influence a person's health and well-being.

Such information could directly inform interventions to improve health. In one example, Kenzie Preston, PhD, a senior investigator at the National Institute on Drug Abuse, and colleagues are collecting electronic diary information from heroin and cocaine users in Baltimore, gathering data about their cravings, moods and drug use. They're also collecting information about participants' travel paths, and combining it with environmental data about factors such as physical disorder in the environment, including abandoned buildings or the presence of drug paraphernalia.

The research team is using the data to create an algorithm that predicts when people are at risk for cravings and drug use, Preston says. "The idea is that the data can be used to alert people when they might be vulnerable."

For example, users walking through a neighborhood associated with increased cravings could receive a text message suggesting they reach out to a counselor. Such technology could be adapted to any number of mental health problems in which environment plays a role, including depression and post-traumatic stress disorder, Preston adds.

Protecting privacy

While geolocation holds promise, it also raises serious privacy concerns. "GPS is extremely revealing of a person's identity," says Mohr.

To protect individual privacy, McLaughlin codes her software to intentionally blur geolocation data, pinpointing a random location within an area that's hundreds of square feet in diameter. "We have to be sensitive and thoughtful about the ethics," she says.

Unfortunately, efforts to cloud the data could make them less useful for research and interventions that rely on geographical precision, Mohr says. He agrees that privacy is a hurdle, but he believes it can be overcome by giving users ownership over their data—allowing them to opt out of data collection at any time, delete their data retrospectively, choose what companies or studies they want to share data with and perhaps even sell their data for use in research.

"Your phone company and the apps on your phone are already pulling those data for commercial exploitation. We want to use them for people's benefit as well," he says.

He says researchers shouldn't shy away from this burgeoning field because of privacy concerns. Indeed, he says, geolocation researchers can follow the lead of researchers in genetics—a field that is overcoming similar privacy concerns to change the face of public health.

"Often, new discoveries are made by new ways of measuring," Mohr says. "For the first time, we can begin to understand what people are doing on a moment-to-moment basis, and I think it will give us enormous insights into behavior."

For more on this topic, see "Human Subjects Protection and Technology in Prevention Science: Selected Opportunities and Challenges," by Pisani, A.R., Wyman, P.A., Mohr, D.C., et al., in Prevention Science, 2016.