Frugal Innovation Saving Lives

Transcript

Shane Hastie: Good day, folks. This is Shane Hastie for the InfoQ Engineering Culture Podcast. Today, I'm sitting down with Jason Friesen, who is the founder of Trek Medics. That is correct, isn't it, Jason?

Jason Friesen: Yes, sir, it is. And thank you very much for having me on today.

Shane Hastie: Nice to meet you. My usual starting point is, who's Jason?

Introductions [00:25]

Jason Friesen: Well, I am a paramedic by training based here in San Diego, California in the United States, and I've been working as a paramedic or an emergency medical systems for about 15 years now. But past 10 years or so, I've been the executive director of Trek Medics, which is a nonprofit technology organization dedicated to improving access to emergency systems through innovative mobile phone technologies. And we do that primarily through our Beacon Dispatch Platform, which I'm happy to talk more about today with you.

Shane Hastie: Great. So one of the things that intrigued me when we were put in touch was that working in emergency response in low and middle income countries, the frugal innovation, the very poor infrastructure in a life and safety critical environment. How do you overcome that?

Jason Friesen: The advent of all this or the start of the Beacon Platform really came about when I was living in Haiti after the earthquake in 2010. For your listeners who may be familiar, not only was that an extraordinarily catastrophic earthquake, but there was a follow on second disaster, so to speak, which was the cholera outbreak. And so I remember very clearly I was working for another non-governmental organization at the time, and I was in a meeting with the Ministry of Health and all of the NGOs, and they were talking about fatalities from cholera.

In disaster situations connectivity saves lives [01:50]

And they made the point that 95% of cholera fatalities were happening before the patients reached the hospital. And so the question was, what do we need to do to reduce those fatalities? And the consensus answer was, well, we need to build more cholera treatment centers. And myself as a paramedic, and actually at the time I was with another German from another organization who was a paramedic, our thinking was, well, actually, the problem isn't so much the number of cholera treatment centers you have, it's how quickly or how easily people can get to those.

Our thinking was there are people in the community who have transportation, maybe not ambulances, definitely not ambulances, more like pickup trucks or just personal vehicles who were already transporting people to the hospital and they all had flip phones in their pockets, cell phones. And so maybe if there was a way to alert those people when they were needed and coordinate them to get to the victims more quickly, then you could reduce response times and increase the rate at which they were getting to the hospital before dying.

And so that was, like I said, the advent of Beacon where it was everybody's got these flip phones, smartphones were out then, but they were by no means prolific in Haiti. And so myself and some colleagues set out designing a simple messaging system that would use SMS to send alerts to people who could help so that they could get to victims, put them in their private vehicles and transport them to the nearest cholera treatment center. Now, admittedly, we didn't get it built while I was there in Haiti.

It took me a couple years to realize that the NGOs were focused on other things. And so finally I left after about two years, but went back to New York City and paid out of pocket to get the prototype built. And that's how it all started was we got this simple messaging system, as I was calling it back then, that was using SMS only to be able to alert, coordinate, and track responders using whatever mobile phone they had in their pocket. And the prototype for that was about 2013 we built that.

Shane Hastie: So in Haiti, the example of getting people hospital quicker, tell us some of the stories of how this technology at the simple end, the frugal end, but also as it's evolved, how has that made a difference in the world?

Motorcycle taxi as ambulances [04:05]

Jason Friesen: I think a really great example of that is in Tanzania, after getting the prototype built and tested and launched in a pilot program in Mwanza, Tanzania, which is in Northern Tanzania along Lake Victoria, met with a local doctor there who was very determined to get some kind of pre-hospital emergency care system set up, but same problems. The infrastructure was in place. They had people willing to help. They all had mobile phones, but they didn't have ambulances per se.

They did not have a 911 or an emergency call center, but they had motivation and they had the raw materials. And so this gentleman, his name is Dr. Marko Hingi, we worked together with him to train local motorcycle taxi drivers. They were oftentimes being blamed for a lot of the motorcycle traffic accidents, but also the first ones to see them or be on scene.

So we came together and set up a training program to train the motorcycle taxi drivers and then to equip them with our Beacon Platform so that they would get alerts. And now we just presented a couple weeks ago on this, they're averaging seven minute response times for motorcycle traffic injuries and other emergencies as well, including women in labor, sick children, acute illness, and injury.

And that is a program that now we have a mobile app, we've got a much more sophisticated platform in general, but because some of these motorcycle taxi drivers don't have smartphones, many of them are still using it by SMS, or those who do have smartphones, if they run out of credits or they run out of internet connectivity, they can switch over to SMS and still continue to receive these alerts.

So the redundancy is there as well. But that's, in my mind, one of the clearest examples of going from nothing to something and working with communities so that they can implement effective emergency response systems.

Shane Hastie: When the infrastructure stops working, we're so dependent upon the richness of our technology, and yet it degrades. And in these disaster situations or in some countries, it degrades fairly rapidly, how do you build that resilience in?

Building resilience into the system [06:15]

Jason Friesen: Well, there's certain levels, right? A lot of people ask us the question, what happens when the mobile phone towers go down? And the reality is, when the mobile phone towers go down, we're all in trouble. Unless you have satellite phones or something like that, or you still have redundancy with VHF radios, when the mobile towers go down, we're all in trouble. But if you look at the resiliency that's now happening on the infrastructure side, I'll just give you one example.

I think it was during Hurricane Katrina in 2005, all of the mobile towers went down during the hurricane, and it took us several days before they got them up and running again. But then fast-forward 10 years or so to Hurricane Harvey, which was in Houston, Texas, and Hurricane Sandy, which was in New York City, mobile towers were up and running at least at 2G, 3G levels within hours. So the telecom infrastructure has gotten much more resilient. And as long as we have 2G or 3G, we can still send messages back and forth to alert people and to keep track of where they are.

And so that's really been a big benefit. But again, if the entire mobile phone infrastructure collapses, we're all in trouble. But as long as they've got 2G and 3G, we can work with that and work off the SMS. Now, once they get 3G and 4G back up and they've got good bandwidth, then of course, they can get back into the mobile app and the web application and really use our platform. So it's firing on all cylinders, so to speak, but we can do it in very restrained settings.

Shane Hastie: So how has the platform evolved? How has the product evolved? What's it contained that enables this? And I love that graceful degradation perhaps.

Graceful degradation of service rather than collapse [07:53]

Jason Friesen: Yeah, that's a great way to put it. So Beacon is two components. There is a web application for dispatchers, so that's where they can monitor everything that's going on, who the responders are, where the active incidents are, who's going where, doing what, what their state is. And that's just available through any web browser. We like to say, all you need is a link and a login and you can set up an emergency communication system.

But then there's also the other component, which is for the responders, which is the mobile app, and that's how they receive alerts and are told where to go and can keep dispatchers updated. There are also a bunch of other features within it. There's chat and push to talk and, of course, maps and reports and all of that. But those are the two main components is the web app for dispatchers and the mobile app for responders. But like we've been talking about, if the responders don't have a smartphone or they lose internet connectivity, they can switch over to SMS.

But between those two, you can coordinate pretty sophisticated and high volume emergency response systems. We're working with a range of partners from ambulance services in Somalia to fire departments in Belize to disaster search and rescue in Florida. And then we also work more and more with a lot of what we call alternative response groups. So here in the United States, we're working with a lot of mental health crisis response teams and domestic violence, substance use.

Working with alternate emergency service providers [09:15]

We've got a huge problem in the United States with opioid overdoses. So we're working with community groups. And that's an interesting part too, because the way we've built the software is that a lot of the groups that we work with don't have a formal call center. They don't have the Houston NASA call center with all of these screens in front of them to manage their complicated response systems.

And so for those groups, what we've done is we've actually built it into the mobile app so that they can, and by SMS too, quite honestly, so that they can actually create incidents and send out alerts from the mobile app or SMS if they don't have a proper call center. And one example I'll give of that is in Hartford, Connecticut, we work with a group, the Harm Reduction Group, that they work with folks who have substance use disorder and they're on the streets every day. And they listen to the police scanner.

And when they hear a suspected opioid overdose sent out over the police scanner, they pull out the mobile app, plot that on the map in the mobile app, and they send it out to their street outreach workers who are all carrying Narcan in their pockets. And so Narcan is the opioid overdose antidote, as they call it. And so they can respond very quickly to a person who's experiencing an overdose, administer Narcan, and revive them. About half the time they get there before police, fire, EMS and half the time they get there after.

But even when they get there, after they're able to offer access to social services and continuing care that police, fire, EMS can't do. So for these very much community-based down and dirty, so to speak, really working out of their cars, these response systems are possible to set up these community response networks.

Shane Hastie: Taking it a little bit into where's the technology going, what are you working on at the moment?

Observability helps with scaling [11:04]

Jason Friesen: Biggest challenge has been or area of growth has been in scaling. We've started with very local smaller groups. But as our software has gotten more and more sophisticated, we've been getting involved with higher volume programs. And so what we've been doing is really focusing on making sure that our software can be scaled. Learned very early on that there are two ways to scale. You can scale with hardware or you can scale with software. And clearly scaling with software is the more efficient way to go.

And that's where we've been able to utilize a range of tools to make sure that our software is working as efficiently as possible. And prime to that growth has been New Relic. We've been using their observability to really improve our API, reduce response times on the endpoints, but also to pinpoint areas where we've had a lot of bottlenecks. So I'll just give you one example. Happened very recently. We've been updating and upgrading our mapping application and making it possible for emergency response agencies to upload their own maps.

Well, we got one group that uploaded maps with 10,000 map markers, and they had basically mapped out all of the highways and freeways around their area with all of these mile markers and whatnot. Well, we never really expected anyone to upload a map with 10,000 map markers onto it. But thanks to New Relic, we were able to identify very quickly where the choke points were and work through them so that the next time someone comes around with a huge map like that, it won't be a problem.

So New Relic's observability has been incredibly beneficial to us, but we've also worked with other groups. PagerDuty is another example that really takes what New Relic is observing and helps us set up an internal response system for our engineers. I like to say that PagerDuty is to engineers what Beacon is to emergency responders, right? It tells which engineers need to respond and what they need to do when they get there, and it makes sure they're working on a schedule.

We're a nonprofit organization, so we get a lot of funding from technology companies. And another one has been Cisco. And Cisco has been very good to us, and they've been very helpful in terms of improving our cybersecurity. Because again, as I mentioned before, as we get more and more involved with governments, the threshold for cybersecurity goes up and the requirements increase. And so Cisco has been a huge help to us to make sure that we're getting into compliance with those needs.

But really at this point, our big focus is making the software more scalable for higher volume systems and also kind of polishing and making custom features and functionality available to our users because different response groups have different needs. We've got the core functionality in there and the core features and services that they need, but everybody does things a little bit different. So now we've been able to really make Beacon much more customizable and adapt to a wider range of response agencies.

Shane Hastie: The challenge of working globally like you are and particularly in that emergency response space dealing with, I'm assuming lots of different government legislation, dealing with telecoms providers, how do you navigate those waters?

Challenges with telecoms providers in different parts of the world [14:15]

Jason Friesen: Case by case, country by country. Telecoms have been an adventure. I don't think any of your listeners who work in this space will be surprised to hear that telecoms are very independent actors. Even if you're working with a multinational, they tend to do things differently in each country. And we've set up systems where you need to be able to call for help on any phone, no matter what carrier they're using. And then the responders, because we're using the phones that they have in their pockets, they need to be able to receive messages no matter what carrier they're using.

And it's not easy. I'm just going to be completely blunt. It's a real challenge. I think the telecoms oftentimes, depending on the country you're in... If you're in New Zealand, I'm sure it's the case, Australia, I know it is in the United States, Canada, much of Western Europe, you mentioned the legislation, it is in the legislation that the telecoms need to support the emergency communications infrastructure, right? That's already been passed and there's no questions about it. And as far as the telecoms see it, it's the price of doing business.

But when we get into low and middle income countries where they don't have this set up, there is a lot of hesitance, reluctance, I would even say pushback because they see this as possibly a cost that they can't recover. And we have worked in other countries to try and help them draft legislation for what in the United States we refer to as universal service fees. So on your monthly phone bill, there's an extra 75 cents or a dollar, and that goes directly to the 911 call center to make sure it's funded.

And we've tried to work with countries and governments to pass that legislation. But as you can imagine, the telecoms aren't always keen on that, and the politicians too aren't always keen on another tax, right? I think everybody wants to be taxing through the mobile phone contracts or agreements. So in their mind, it's kind of like get in line. But we have had some really successful relationships and we've had less successful relationships. And I think that a lot of it comes down to public pressure.

Here in the US, North America and the Caribbean, we have a regularly scheduled catastrophic disaster season, that's hurricane season, and the islands in the Caribbean are getting battered every year. And I think that the more the citizens see that relief is possible, the more pressure they put on their governments to make this happen. So it's slow-going, but it's promising. We're in it for the long haul. So we're definitely eager to grow with them and see this come to pass.

Shane Hastie: We're talking to a technology audience, how can they help?

How technologists can help [16:53]

Jason Friesen: Well, I appreciate that. Again, we're a nonprofit organization and we're focused on groups that have been overlooked or neglected by the incumbents, so to speak. I often like to say that our software is for people who can't afford or don't need the legacy dispatching technologies used in advanced response systems. So getting support from technology companies who identify with our mission and who want to see us grow and help us reach more people, that's always a huge thing.

Like I mentioned, New Relic, PagerDuty, Cisco, Twilio, they've been huge supporters of ours for years. And if there are more technology companies out there who, again, identify with our mission and want to support us, we're happy to hear it. But we are also always looking for subject matter experts, people who can advise us on an individual level. I mean, we're going through a process right now with a certain type of authorization in the United States, and it's like, wow, there's a lot of stuff here that is new to us.

If there are people, subject matter experts in this area who want to reach out and at least be technical advisors, we're always glad for that kind of expertise. Because what we started trying to do and where we are now, I never would've believed it if you would've told me this 10 years ago. So the more support and expertise we can recruit, the better things will go for everyone.

Shane Hastie: On that note, how do people get hold of you?

An example of responding to an emergency in Florida [18:12]

Jason Friesen: Well, they can check out our website, trekmedics.org, but also our Beacon Platform. And this is another thing to get involved. We're always happy to have more users, and we've actually built our platform so it can be completely do it yourself. I'll just share real quickly. One story that happened recently was a year ago or so, hurricane Ian was barreling down on Florida.

We sent out some ads on Facebook, pointed out Southwestern Florida to say, if you need a search and rescue software, because hurricanes hit, they flood fire departments, they flood ambulance stations, everybody gets flooded. So if you need a dispatch software, sign up. Long story short, within three days of landfall, there was a group that was using Beacon to dispatch search and rescue and mental wellness experts all across Southwestern Florida. And about four or five days, they did hundreds and hundreds of incidents they responded to.

And so that really helps us because it gives us validation and helps us to show to the authorities. So if there are people out there who need a simple to use dispatching software, I would say go to our website. Trekmedics.org is the public website, but the platform you can find at beacon.trekmedics.org. And feel free to create an account and start playing around with it, and we'd love to hear what you think. We're always looking for more users, especially with established groups, to see how we can support them.

Shane Hastie: Jason, thank you so very much.

Jason Friesen: Thank you, Shane. It's been a real pleasure. Really appreciate the opportunity.

Mentioned

• Trek Medics