In 2018, Dallas set an ambitious goal to speed up medical emergency response times: get to the scene within five minutes at least 90% of the time.
The new goal — based on guidelines set by the National Fire Protection Association — was a reaction to growing dissatisfaction with how long it took ambulances to respond to 911.
A year later, having made no progress toward meeting the goal, the city shelved it. It was replaced with a new one: nine minutes, nearly double the previous year's target.
The city is all but guaranteed to meet it. When the city published its goal to "maintain percentage of EMS responses within nine minutes above 90%" in November, it was hitting that mark 90.4% of the time.
In a statement sent to the Observer, Dallas Fire-Rescue claimed that the new goal is "consistent" with national standards. A city spokesperson agreed, writing, "This specific goal is set in line with standards set forth by the National Fire Protection Association (NFPA)."
The NFPA's 2020 standard allows for five minutes of "turnout" and travel time for the arrival of a first responder. It allows for an additional four minutes of travel time for a vehicle equipped with "advanced life support," or ALS, but that's in addition to — not instead of — the first responder.
"While we are always striving to meet them both, the only one we have consistently met is the standard which governs the percentage of times an ALS unit responds to an incident within 9 minutes," wrote DFR in a statement.
City Hall made public safety one of its six "strategic priorities" in 2017 and promised to keep track of its progress along a variety of measures, from homicide clearance rate to the "average time it takes the first paramedic to arrive on scene after dispatch and provide medical care."
Its goal for the later measure was an average of five minutes. This was not the national standard — which is 90%, not the average — but at least it was something to work toward. In October 2017, response times were 15 seconds too slow, and they remained around that level for the next year.
Meanwhile, Dallas residents were growing increasingly unhappy with the city's emergency medical services. A 2016 survey commissioned by the city found 88% of people were "satisfied" with the "ambulance/emergency medical services." A similar survey conducted two years later found that satisfaction had dropped to 78%.
So the city got more aggressive. Officials changed the goal to "Percentage of EMS responses within five minutes" and set a target of 90%, mirroring the national standard. But the city's performance on this measure was dismal — about half of calls met the target — prompting criticism from The Dallas Morning News.
"It may be time for the city to study Fire-Rescue resources and deployment in a comprehensive way, just as it is doing with police," wrote the News' editorial board. Houston commissioned such a study in 2016, which recommended hiring hundreds of new firefighters — ideally with EMT certifications.
Instead, officials from DFR and the city manager's office got together to set a new goal. They replaced the five with a nine, and all of a sudden a "key performance measure" was in the green.
But experts question the wisdom of such creative accounting.
Nine minutes is not the standard, according to Lori Moore-Merrell. She's the president of the International Public Safety Data Institute, a Washington, D.C., nonprofit that analyzes data for fire departments across the country.
"Don't set a metric that you know you already meet," she said. "If you can't meet the industry standard, say so. Notify your decision-makers, notify your community."
In its statement, DFR noted that it "is currently in the process of analyzing response data from our tiered EMS dispatch system in an effort to more efficiently dispatch resources to 911 calls" while adding more ambulances and fire stations. As the city grows, its job only gets more difficult.
Advanced life support, which requires paramedics with specialized equipment, has been the primary method of responding to cardiac arrest and other medical emergencies in the United States since the 1970s. Most of Dallas' emergency response teams are equipped with ALS.
"ALS care provides the highest level of emergency responder medical care to our citizens, and the new measure more accurately reflects our performance in that regard," wrote a DFR spokesperson in defense of the new measurement.
But researchers have questioned the wisdom of this approach. "Basic life support" provided by emergency medical technicians is cheaper, they have found, and often more effective. "ALS has no established benefit over BLS for patients with cardiac arrest," wrote Harvard researchers in 2014.
Because basic life support requires only basic techniques, like chest compressions and automated external defibrillation, it can be done by whomever happens to get to the scene first — whether it's an ambulance, fire truck or the chief's SUV. The emphasis is on getting the patient to the hospital as quickly as possible.
NFPA has included five minutes to BLS in its standard since at least 2004. The nonprofit's guidelines are periodically updated and represent an international consensus on proper risk-mitigation techniques.
"Time matters," said Rade Vukmir, an emergency room physician and researcher who has published a series of academic papers on the subject.
For many cases, he said, there's a rule of thumb: for every extra minute it takes for someone to get to the hospital, the likelihood they walk out OK decreases by 10%.
Vukmir, who has a Texas medical license and used to practice in Amarillo, said Dallas isn't alone in struggling with what he calls "big-city problems."
"If something happens at 1 o'clock in the afternoon and you're trying to get across the city, it takes time," Vukmir said. "It's not their fault. There's traffic everywhere."
In fact, Dallas' average response time to medical emergencies compares favorably to other big Texas cities. Houston's fire department gets to the scene in around six minutes on average. San Antonio's average is a few seconds worse.
Earlier this week, DFR trumpeted its own average response times — 5 minutes, 18 seconds — on its homepage.
But after the Observer asked the department to explain why their goal differed from City Hall's, the statistic was removed and replaced with the city-approved measure.