Local governments collect reams of data, but much of it goes unused. New Orleans shows how cities can tap into this data to reduce EMS response times.
Local governments collect reams of data for program administration, regulatory, or law enforcement purposes. However, much of it is underappreciated, underdeveloped and underused. This treasure trove of data can address some of our most vexing problems.
Like many cities, New Orleans was facing an increasing number of Emergency Medical Services (EMS) calls. From 2014 to 2016, the demand for services had increased by 12 percent, and emergency responders were struggling to keep up. Over those three years, EMS ambulance response times for urgent calls fell from 80 percent calls being addressed in under 12 minutes to just 72 percent.
In addition, the city was facing a geographic disparity: Response times in the outer reaches of the city were faring worse as a result of being located farther from the urban core.
Fortunately, New Orleans city government has a history of using data to solve problems. The city has used data analytics to fight blight, address gang violence and prevent fire fatalities. So the city’s EMS staff agreed to partner with the New Orleans’ Office of Performance and Accountability and students from Louisiana State University’s Masters of Science and Analytics Program to find a cost-effective solution.
To achieve swifter and more equitable ambulance response times, the team focused on two questions: how EMS ambulances were picked to address a 911 call and where they were stationed after responding to a call to wait for the next one.
To conduct their analysis, they accessed the detailed data the city had collected on every EMS call citywide. Using ESRI’s ArcGIS—a mapping and analytics platform—they plotted every 911 call for the past five years and identified over 100 potential waiting locations that would create eight-minute response times for high- and low-traffic conditions. Next, they ran multiple simulations to identify where they could place ambulances to cover the greatest number of calls most effectively.
Analysts had to work with real-world constraints, such as not parking ambulances in front of residential homes. There was also the issue of limited ambulance availability if several were on call at the same time. The stakes were high: determining the right waiting locations could mean the difference between life and death for residents.
After several rounds of analysis and consultations with emergency medical technicians and paramedics, the end result was printable maps indicating optimal locations for EMS ambulances to wait for their next call: one for the day shift and another for the night shift. These maps were scalable based on the number of EMS ambulance teams operating during any given shift, and the model can also be updated for future city demographic changes.
The new approach produced results. The largest impact was that equity was improved: Algiers (4th district) and New Orleans East (7th district), which were the historically poorest served districts, benefited the most under the new placement location maps. Response time compliance improved by 20 percent in the Algiers district and by 9 percent in New Orleans East under the new ambulance placement protocol.
Analysis of the initiative also showed a modest—but statistically significant—improvement in overall response time compliance during the night shift. During the day shift, when there is higher demand and higher traffic levels, response time compliance remained unchanged. Importantly, though, the new placement location maps demonstrated that the city could enhance equity in service across locations without sacrificing efficiency during the busy daytime shift.
This success story underscores the potential of the broader use of administrative data to solve policy challenges.
The problem for local governments is not a lack of data—in fact, cities and counties have more data, more accurately collected, than virtually anyone else. But the vast knowledge contained in these data sets is often trapped in repositories that are highly protected and bureaucratically controlled.
We believe that public data sources—de-identified and privacy protected—should be used to benefit the public and advance social progress. As advocates for evidence-based policymaking, we encourage local leaders to take steps to break through the bureaucratic, legal, and cultural hurdles that prevent us from unlocking the full potential of administrative data.
New Orleans showcased the extraordinary power of using data to improve the application of a city-wide program. There are countless more policy wins waiting for cities and counties willing to take up the challenge and better use their data to improve the lives of their residents.
Linda Gibbs, who served as New York City Deputy Mayor of Health and Human Services from 2005-2013, is a principal with Bloomberg Associates and a Senior Fellow at Results for America. Maia Jachimowicz, who served as Director of Policy in the Office of the Mayor of Philadelphia from 2014-2015, is Vice President for Evidence-Based Policy Implementation at Results for America.