Neighbor saving neighbor during an opioid crisis using a smartphone

December 22, 2020
Contact: Laurel Thomas


A University of Michigan researcher says empowering communities to form an opioid crisis response network based around the technology we carry with us every day can save lives when an overdose is suspected.

Gabriela Marcu, assistant professor at the U-M School of Information, and colleagues detail their pilot study in Philadelphia in which they put a cell phone app and the overdose reversal medication naloxone into the hands of volunteers willing to respond in a crisis.

Using the UnityPhilly app she and the team created, volunteer bystanders were alerted to potential overdoses nearby to which they could respond and administer a nasal spray containing the medication.

Marcu said the pilot showed the value of involving communities in the ongoing opioid battle that may not be as much in the headlines today but is still raging.

Talk about the success of UnityPhilly and the volunteer efforts.

Over the course of one year, 112 volunteers used our app to signal 291 suspected opioid overdoses. Naloxone was administered to the overdose victim by these volunteers 74 times—22 times by someone who used our app to respond to the signal, and 52 times by the signaler themselves (sometimes more than one dose is needed to revive the victim, which is why one might signal and also administer).

A lot of research went into creating the program, and even the name, all tied to empowering the community. Can you explain?

We started by working with community-based organizations, and understanding existing efforts such as overdose reversal training. We also spoke with members of the community who had experience responding to an overdose, or even overdosed themselves, to help us understand the challenges involved. Ultimately, what we found was that the community was interested in the concept and open to piloting the app with us. In a neighborhood like Kensington (within Philadelphia), the number of overdoses is so overwhelming that community members saw the app as another way to bolster their efforts to respond and provide livesaving help.

Due to distrust of police and concerns about prejudice toward nonmedical opioid users, community members appreciated the ability to ask for help from one another without necessarily having to interact with professional responders. This motivation for uniting members of one’s community to provide lifesaving intervention inspired the name of the app: UnityPhilly.

With all of the angst in communities with law enforcement and distrust of police, how easy is it to get volunteers to participate in a program that could end up involving emergency personnel?

Some may choose not to call 911 at all because they do not want to risk an interaction with professional responders (especially police, though even others are avoided, too), or they may not even want to speak to them on the phone. They preferred the idea of using the app to signal nearby volunteers and not having to speak to anyone by phone. Although our app does alert professional responders to ensure medical help arrives, the first interaction the signaler has is more likely to be with another volunteer.

Recruiting participants was somewhat of a challenge, as it is with any research study, but some reported they would discuss the app with others and encourage them to get involved with using it too. So word of mouth diffusion is another avenue for us to explore.

A year or so ago opioid misuse was one of the largest public health crises facing our nation but it really has been eclipsed in terms of attention by COVID-19 in the past year. Has much changed in the situation one way or the other?

Studies are showing that overdoses have increased during the pandemic, likely for a variety of reasons. And it’s likely that there is less help when someone does overdose. People are generally encouraged not to use alone, for example, in case they overdose, but that goes against social distancing measures as we are all more isolated. There are also fewer people in public spaces as we all try to stay home, but 58% of the time naloxone was administered in the street during our study. So, there may be fewer people around now to signal an alert or revive an overdose victim with naloxone.

How does a program like this look in a COVID-19 world?

Our system relies on people helping people by being around one another. To safely do that, we’d have to monitor vaccination, potentially ensure that any volunteer on the app has been vaccinated before asking them to respond to an overdose event. We are also exploring how technology could monitor someone for signs of an overdose, without a bystander witness, and autonomously signal for help.


More information: