Author: Alan Dash, Director, Project Management
Traditional medical transportation services are rife with challenges, including missed appointments, inability to schedule same-day service and schedule delays. Luckily, ride-share programs1 for ambulatory patient transport are on the rise. These ride-share programs are easy to use, document pick-up and delivery, accept digital payment methods and provide lower cost options. Although there are many benefits to ride-share programs, pitfalls exist. These include: lesser experienced drivers, inconsistent ability to accommodate wheelchair lifts, potential security issues and HIPAA compliance risks.
While many medical transportation options exist today, there is a desire in the industry to find unique alternative means for patient transport. This article discusses one such method—outpatient delivery via autonomous vehicles—and offers some thoughts on design impacts and benefits to the healthcare sector.
Today, there are many ways to navigate to a healthcare facility including public transportation, ride- shares, third party leasing and city sponsored buses. The easiest way to locate a healthcare facility is to enter the hospital name into a vehicle GPS (Global Positioning System), which searches for the name and location using information that has been geocoded into public databases. The GPS does its job to get us to the campus, maybe even to the building. Then we humans need to take over because GPS is not necessarily going to place us at the right door. So we read the signage to find exactly where we need to be, where to park, etc. It takes this bit of human intervention to get to the actual appointment location.
Figure 1: 3-6
What the Future Holds
Research is underway for the development of what’s believed to be the future of patient delivery – autonomous vehicles (AVs). The tremendous value that ride-share programs have shown in both reduction of missed appointments and costs suggests that the benefits of AVs will likely outweigh the risks—and that AVs may soon be delivering patients to their healthcare facilities. Lest that seem too futuristic, seven major cities around the world (four in the U.S.) are participating in the Bloomberg Aspen Initiative (a collaboration between Bloomberg Philanthropies and The Aspen Institute) to research and develop means and methods, from technology to policy, for promoting AV patient delivery.
Locating a Hospital Without Human Intervention
According to a recent CBRE Research2 article, by 2029 we could potentially adapt to a fully-autonomous means where passenger control over a vehicle is considered limited or not all. In just 10 years’ time, an ambulatory patient could be picked-up by an autonomous car and brought to an appointment.
Based on our experience, it can take three years to design and plan a new hospital facility and 10 or more years for a complex campus. Construction will take place months or even years after initial designs are approved. This time gap can be quite costly for owners because of technology changes. And because we are not delivering patients autonomously today, consideration for this isn’t taking place today.
What Hospitals Can Do Today
It’s going to happen – the benefits of AV patient delivery will be great, and we should prepare now. We can do this by understanding how an AV will use GPS data. There are two main considerations with the scenario of transporting a patient from one location to another using AV.
- Data Entry
There are three main keepers of location data and several smaller players; some are free (US Census) and some private (NavTech & TeleAtlas). To get a new facility in a database of locations, a Geocoder (a private data vendor) needs to get the new location and address geocoded into the subscriber data base. This takes time and generally relies on a geocoder to input the location information.
- Exact Location
Although location data is important to get a patient to the right building, it’s going to be more important to ensure the patient is sent to the right door. This is where we will see a change in location data, because we need the data to be more specific now. In Figure 1, the GPS was asked to take a vehicle to a medical center facility, first by name, then by address. This works to get the vehicle to that facility. However, in Figure 2 we see a large campus of multiple buildings that share the same address. In this scenario, the AV would stop at the address pinned to drop the patient. The AV cannot read signs to locate the correct building and door. Because this is not exact, a patient would need to walk hundreds of yards to their appointment. Unfortunately, this scenario will happen unless some consideration is brought into the design process early.
In Figure 2, there are three colored flags indicating potential public entry doors; however, the building will likely have one address. To deliver a patient to the correct door, the design team would need to label their drawings and have exact location data entered in the geosphere with every door of the new facility. This includes their specific latitude and longitude points blended with each line of service to be provided at the designated door. Even if the current door is not a public entry, tomorrow it might become one. Although the labeling solution sounds simple, for an AV to deliver a patient to the correct spot there will be much work involved. Early planning is important.
While it is crucial to discuss patient drop-off locations as the facility is being designed, now we will need to visualize the capability for a vehicle to stop and drop a patient autonomously. A normal street will not be a problem; however, if the point of entry is a courtyard or in some non-accessible spot for vehicle traffic, then the nearest appropriate spot on the street should be geocoded as a virtual door for that service line. Because programming location data is not quick or easy, it’s important to ensure geocoders are brought on early in the planning process and that enough time is allotted in the project schedule for them. Hiring too late could mean that the facility might open without the location information being shared in the geosphere, causing missed or late appointments. Ideally, the project team will include technology advisors who can guide you through this process, reducing errors and mitigating spend. Remember, you are building in and for the future. As far as we know, cars won't be able to intuit the best drop off points for their patients, therefore we will need accurate and thoughtful geocoding strategies to heighten their usefulness in AV patient delivery.
For more information contact Alan.Dash@cbre.com or call +1615-948-0905
- Esposito, L. (2018, October 10). Lyft and Uber Help Patients Make It to Medical Appointments. Retrieved from https://health.usnews.com/health-care/patient-advice/articles/2018-10-10/lyft-and-uber-help-patients-make-it-to-medical-appointments
- Greenwood, R. (2016, April). On the Road: A Futuristic Look at Self-Driving Vehicles and CRE. Retrieved from https://www.cbre.com/research-and-reports/Americas-ViewPoint---On-the-Road-A-Futuristic-Look-at-Self-Driving-Vehicles-and-CRE-April-2016
- Zion Market Research. (2018, October 4). Healthcare Transportation Services Market in U.S. Will Reach USD 31.31 billion by 2026. Retrieved from https://globenewswire.com/news-release/2018/10/24/1626157/0/en/Healthcare-Transportation-Services-Market-in-U-S-Will-Reach-USD-31-51-Billion-By-2026-Zion-Market-Research.html
- Belk, D. (2018, December). True Cost of Healthcare. Retrieved from Truecostofhealthcare.org
- Lovelace, B. (2018, June 9). Doctors Need Patients to Keep Their Appointments. Uber and Lyft Want to Help Make That Happen. Retrieved from https://www.cnbc.com/2018/06/08/uber-and-lyft-see-an-opportunity-shuttling-patients-to-the-doctor.html
- Cronk, I. (2016, September 2). Transportation Shouldn’t be a Barrier to Health Care. Retrieved from https://www.statnews.com/2016/09/02/transportation-barrier-health-care/