How Emergency Services Respond After a 911 Call

TL;DR. Dialing 911 feels like reaching one person who then personally comes to help. What actually happens is that the call is instantly routed, by cell tower or billing address, to one of roughly 5,700 local call centers, answered by a telecommunicator trained on a scripted medical protocol who is simultaneously deciding what to send and walking the caller through what to do until it arrives. The unit that shows up minutes later was selected by a computer system tracking every ambulance, engine, and patrol car in real time. None of this existed as a system before 1968, and the ambulance part of it barely existed at all before a blunt 1966 federal report said, in print, that American ambulances were unsuitable, poorly equipped, and driven by untrained people. Getting from that report to a paramedic at your door in under ten minutes took sixty years of work that is still not finished.

Key takeaways

  • A 911 call is routed to the correct local call center (a Public Safety Answering Point, or PSAP) using location data attached to the call itself. For landlines this has worked since the 1970s; for cell phones, which don't have a fixed address, it took a federal mandate in 1996 and is still being refined today under Next Generation 911.
  • Dispatchers don't just guess what to send. Structured systems like the Medical Priority Dispatch System walk a call taker through scripted questions and let them deliver real time instructions, including talking a caller through CPR, before any unit arrives.
  • Modern EMS was not a gradual improvement on old ambulance service. It was a direct, documented response to a 1966 government report that found half of U.S. ambulance runs were handled by funeral homes using hearses, because a hearse was one of the few vehicles long enough for a stretcher.
  • An EMT and a paramedic are not the same job. An EMT trains for a few months in basic life support; a paramedic trains for a year or more in advanced life support, including starting IVs, reading heart rhythms, and giving medication.
  • 911 telecommunicators are a distinct, credentialed profession, not clerical staff, and research finds them at PTSD risk levels comparable to police officers, while more than half of U.S. 911 centers report they are short staffed.
  • Response times aren't uniform. Nationally, median EMS response is around 6 minutes in urban and suburban areas versus roughly 13 minutes in rural ones, a gap driven by distance and by a volunteer emergency workforce that has shrunk by roughly a quarter since 1984.

The moment nobody thinks about

Something is badly wrong: chest pain, a collapse, smoke under a door. Three digits get pressed, and within a ring or two, a stranger's voice is asking calm, specific questions. Within minutes, other strangers, wearing specific uniforms, carrying specific equipment, are physically at the door. The person who called never sees the building those strangers work out of, never meets the person who decided which vehicle to send, and rarely thinks about the fact that the whole sequence, from panicked dial to trained hands on a patient, is a coordinated system with its own history, workforce, and failure modes. It only has to work in the worst moment of someone's week for the system to be doing its job. It usually does. That's exactly why nobody notices it exists until the one time it doesn't.

The instant the call connects

The first problem a 911 call has to solve is not medical. It's geographic: which of the thousands of local call centers scattered across the country should answer this specific call, right now. Get that wrong and every minute that follows is wasted.

For a traditional landline, this was solved by the 1970s with what's called Enhanced 911 (E911). A landline's phone number is permanently tied to a physical address in a database, so the call can be automatically routed to the PSAP responsible for that address, and that address pops up on the call taker's screen before they've said a word.

Cell phones broke that assumption completely. A mobile number isn't tied to any address; it's tied to a person who could be calling from anywhere. Early wireless 911 calls could only be routed to whichever PSAP covered the cell tower that happened to pick up the signal, which in a rural area might be tens of miles from the actual caller. The FCC's wireless E911 rules, adopted in 1996 under the Wireless Communications and Public Safety Act of 1999, forced carriers to do better in two stages. Phase I required carriers to hand over the caller's phone number and the location of the serving cell tower. Phase II required something closer to an actual position: latitude and longitude, accurate to somewhere between 50 and 300 meters depending on the technology in use. Even that wasn't the end of it. The 2019 RAY BAUM's Act rules added a further requirement for a "dispatchable location," meaning not just coordinates but, where possible, a street address plus a floor or room number, since a set of GPS coordinates that lands in the middle of a 40 story building doesn't actually tell anyone where to go. Getting this right for every combination of phone, carrier, building, and location technology remains an active engineering project, not a solved problem.

Once the call reaches the right PSAP, a call taker (sometimes the same person who will go on to dispatch units, sometimes a separate role split across a larger center) doesn't simply ask "what's your emergency?" and improvise. Most call takers work from a structured protocol, the most widely used being the Medical Priority Dispatch System (MPDS), a set of scripted questions organized around dozens of specific chief complaints ("difficulty breathing," "unconscious," "childbirth") with built-in branching logic. The answers assign the call a priority code that tells the dispatcher what to send and how urgently, and, critically, they let the call taker start helping immediately: reading a caller scripted, medically reviewed instructions for CPR, choking, childbirth, or bleeding control before a single unit has arrived. This is not improvisation by a helpful stranger. It's a rehearsed clinical script, refined for decades, delivered by someone certified to deliver it.

Don't be confused: the "call taker" and the "dispatcher" aren't always the same person, and a PSAP isn't always the agency that responds. In a small rural communications center, one telecommunicator commonly answers the phone, runs the medical protocol, and selects and sends units, all in the same call. In a larger metro area, those can be two distinct jobs working side by side: a call taker gathering information and giving instructions while a separate dispatcher, watching a map of every unit's location, decides who actually rolls. And a PSAP that answers a call isn't necessarily the agency that handles it. A primary PSAP takes the initial call, and if the emergency belongs to a different jurisdiction or agency (a call about a boating accident, for instance, that needs a county sheriff rather than a city police department), it transfers the call to a secondary PSAP that actually owns that response.

From a phone to a gurney: the response chain

Deciding what to send is the dispatcher's job, and it depends on more than the MPDS priority code. A structure fire needs a fire engine, likely more than one, and probably an ambulance staged nearby in case anyone is hurt. A violent assault needs police, and possibly EMS staged a block away until the scene is confirmed safe. A cardiac arrest needs the closest unit capable of starting CPR immediately, which is often a fire engine rather than an ambulance, because most communities have more fire stations than ambulance posts and fire crews are frequently cross-trained in emergency medicine specifically so they can be first on scene for medical calls, not just fires.

The tool that makes this possible in real time is computer-aided dispatch (CAD), software that tracks every unit's location and availability status continuously, usually fed by automatic vehicle location (AVL), GPS transponders in each vehicle reporting position back to the system. Instead of a dispatcher trying to remember which ambulance was last seen where, CAD can flag the closest available appropriate unit the instant a call is classified, and it keeps updating that picture as units go en route, arrive, clear, and become available again. The same system usually pulls in mapping data and prior call history for an address, so a dispatcher sending someone to a known trouble spot, or a building with noted hazards, sees that context automatically.

From there the physical chain is straightforward, but every link is timed and, for career fire departments, formally standardized: call answered, call processed and units selected, crew notified and rolling (turnout time), travel to the scene, care beginning at the scene, and, for a medical emergency, transport to a hospital. That last step has its own handoff choreography. Paramedics typically radio ahead to the receiving emergency department with a short verbal report, vital signs, treatment given, estimated arrival, so the hospital team can prepare. Physical handoff at the hospital door is supposed to be fast, but it frequently isn't: emergency departments that are themselves full can leave an ambulance crew and patient waiting on a gurney in a hallway, a delay the industry bluntly calls wall time or ambulance patient offload time. A widely cited national benchmark calls for offloading a patient within 20 minutes, 90 percent of the time; real delays of 20 to 60 minutes are common, and during the worst of the COVID-19 pandemic some crews reported waits of 18 hours or more, an ambulance and its crew held immobile, unable to answer another call, the whole time. Once the handoff is complete, the patient's care becomes the hospital's problem, handed off to a building that has its own round-the-clock hidden machinery, covered in the chapter on how hospitals remain operational around the clock, and this chapter's job is done.

The people this all depends on

The person who answers the phone is a telecommunicator, the professional term for what most people call a 911 dispatcher, and it is a real, demanding, and increasingly credentialed profession in its own right rather than a clerical job that happens to involve a headset. The National Emergency Number Association offers an Emergency Number Professional (ENP) certification requiring years of documented experience, points earned through training and service, and a proctored exam, renewed every four years. Research on the job itself has found real cause for that seriousness: a 2012 study by psychologist Michelle Lilly, published in the Journal of Traumatic Stress, found telecommunicators experience duty-related trauma exposure, and resulting PTSD risk, comparable to sworn police officers, despite never physically leaving the console, because hearing a crisis unfold in real time over the phone, sometimes involving a child or someone the dispatcher knows, produces its own lasting distress. That workforce is also, currently, in genuine short supply: an industry-wide survey by the International Academies of Emergency Dispatch and the National Association of State 911 Administrators found more than half of U.S. 911 centers reporting a staffing emergency, with vacancy rates at some centers exceeding 50 percent of authorized positions, in a job that pays a national average around $46,000 a year for work that requires someone in the chair every hour of every day.

The people who arrive at the scene split into distinct trades with distinct training.

Don't be confused: an EMT and a paramedic are not interchangeable titles for the same job. An Emergency Medical Technician (EMT) typically completes 120 to 150 hours of training, often over a few months, and is certified in basic life support: CPR, oxygen delivery, bleeding control, splinting, and safely moving and monitoring a patient. A paramedic holds an EMT certification already and then completes 1,200 to 1,800 additional hours, often through a year-plus, associate-degree level program, qualifying them for advanced life support: starting intravenous lines, reading cardiac rhythms, performing advanced airway procedures, and administering a defined list of medications, all typically under a physician's standing protocols rather than in-person supervision. The National EMS Scope of Practice Model, published by the federal traffic safety agency NHTSA, formalizes this as a four-rung ladder (Emergency Medical Responder, EMT, Advanced EMT, Paramedic), where each level builds on the skills of the one below it. Two people can arrive in the same ambulance, in visually similar uniforms, with legally different authority over what they're allowed to do to a patient.

Many of the people showing up in fire gear are also EMTs or paramedics; fire-based EMS, where a fire department is the community's primary medical first responder, is common precisely because fire stations are spread more densely than dedicated ambulance posts, and because a fire engine with trained crew can often reach a cardiac arrest faster than an ambulance can. Police officers remain the default response for calls involving violence, active danger, or a need for legal authority, and in a small but growing number of communities, they're now joined, or in some calls replaced outright, by mental health crisis specialists, discussed further below.

Behind all of them is a quieter layer almost nobody outside the profession thinks about: the technicians and administrators who keep the CAD software, the radio network (increasingly built on a shared technical standard called Project 25, or P25, specifically so that police, fire, and EMS radios from different manufacturers and different agencies can talk to each other during a shared incident), and the underlying address and location databases all actually working. None of them ride in an ambulance. All of them are the reason the ambulance knows where to go.

Where this system came from

The United States' first 911 call was placed on February 16, 1968, in Haleyville, Alabama, a small town of about 5,000 people. Alabama state legislator Rankin Fite placed the call from Haleyville's mayoral office to a bright red telephone at the local police station, answered by U.S. Representative Tom Bevill. The Alabama Telephone Company had deliberately moved ahead of a national plan the FCC and AT&T had only just announced that January, choosing Haleyville specifically because it had the switching equipment and experienced staff to make the demonstration work. Adoption after that first call was slow and uneven, not instant: 17 percent of the U.S. population had 911 coverage by 1976, roughly a quarter by 1979, half by 1987, and it took until around 2000 to reach approximately 93 percent coverage. Even within a single metro area the rollout could be strikingly patchy; parts of the Chicago suburbs didn't get 911 service until well into the 1980s, and some towns delayed it even longer over cost.

The ambulance riding to meet that call has an even more startling origin story. As late as 1966, roughly half of the ambulance runs in the United States were provided by funeral homes, using hearses, for the simple practical reason that a hearse was one of the few vehicles long enough to carry a person lying flat. Funeral attendants driving those hearses typically had little or no medical training; the same vehicle might carry a badly injured patient to the hospital one hour and a body to the funeral home the next. In 1966, the National Academy of Sciences published a report that named the problem in blunt terms: Accidental Death and Disability: The Neglected Disease of Modern Society, commonly called simply "the white paper." It documented that accidental injury killed roughly 107,000 Americans in 1965 alone, temporarily disabled more than 10 million more, and left about 400,000 permanently disabled, at an estimated annual economic cost near $18 billion, while finding that ambulance standards nationwide were inconsistent and "often low" and that "most ambulances used in this country are unsuitable, have incomplete equipment, carry inadequate supplies, and are manned by untrained attendants." The report, paired with the Highway Safety Act of 1966, which created the agency that became NHTSA and gave it authority over EMS training standards, is widely credited as the direct trigger for building organized, medically trained ambulance services in the modern sense.

A year before that report was published, one specific answer to its diagnosis was already running in Pittsburgh. In 1967, Freedom House Ambulance Service began serving the city's Hill District, staffed entirely by young Black men trained by Austrian anesthesiologist Peter Safar, a pioneer of CPR technique, in a 32 week, 300 hour course covering anatomy, advanced first aid, and emergency driving, a level of training essentially unheard of for ambulance crews at the time. In its first year, the service ran nearly 5,900 calls with a dead-on-arrival rate of under 2 percent, a sharp contrast to the hearse-and-funeral-home standard it was replacing. Freedom House's training model and its medical director's later textbook went on to shape paramedic training nationally, an early proof, running in parallel with the federal white paper, that trained pre-hospital care actually saved lives rather than just moving bodies faster.

Standards that hold the pieces together

Fire response has a specific, numeric national benchmark in NFPA 1710, a standard from the National Fire Protection Association covering career (paid) fire departments. It specifies, among other targets, that a first engine should arrive on scene within 4 minutes for 90 percent of calls, with a second unit within 6 minutes, after a call-processing and turnout sequence measured in seconds, not minutes. A companion standard, NFPA 1720, sets deliberately more lenient targets for volunteer and combination departments, acknowledging outright that volunteers responding from home or work cannot realistically match a career department's speed, with targets that stretch to around 14 minutes in the most sparsely populated areas.

On the medical side, the National EMS Scope of Practice Model, covered above, is the closest thing to a single national standard for what an EMT or paramedic is legally allowed to do, though it's implemented and enforced state by state, so the exact skills permitted at each level can vary at a state line.

The current standards frontier is Next Generation 911 (NG911), a still-unfinished nationwide upgrade from the old analog, voice-only 911 network to an internet-protocol-based system capable of carrying text messages, photos, video, and richer location data directly into a call center. Text-to-911 is the most visible piece, useful for callers who can't safely speak (a domestic violence situation, for instance, or someone who is deaf or hard of hearing), but the deeper goal is a system that can finally treat location, video, and building data as normal parts of a call rather than as separate, harder-won add-ons.

Keeping it all running

None of this holds together without continuous, unglamorous upkeep. CAD software and the P25 radio networks it coordinates with need patching, testing, and periodic hardware refreshes, generally handled by dedicated public-safety IT staff rather than the dispatchers who use them daily. Location databases need constant correction: a new cell tower changes which PSAP a nearby call should route to, a new apartment building needs its address added to the Master Street Address Guide (MSAG) that matches street addresses to the correct emergency response zone, and a VoIP phone subscriber who moves apartments but never updates their service address can send a real emergency call to the wrong county entirely. None of that maintenance is a one-time project; it is a permanent, recurring task that grows every time a neighborhood changes.

People need recurring upkeep too. Telecommunicators, EMTs, and paramedics all hold time-limited certifications that require periodic continuing education and retesting to keep, not a credential earned once and kept for life, precisely because clinical guidelines and dispatch protocols themselves keep being revised as new evidence comes in.

When it breaks

The clearest documented case of the backend itself failing is the April 2014 multistate 911 outage. A software defect in a call-routing platform run by Intrado, a 911 technology vendor, involved an internal counter that was supposed to assign each incoming call a unique identifier; when the counter hit its coded limit, the system began silently dropping new calls instead of rolling over correctly. The result, spread across CenturyLink's service territory in seven states, left an estimated 11 million people unable to reach 911 for more than six hours, with roughly 6,600 calls never getting through at all, including calls involving a heart attack, an overdose, a home intrusion, and reported domestic violence. It wasn't a storm or an attack. The FCC's investigation explicitly found it was a preventable "sunny day" software failure, and it fined CenturyLink $16 million, at the time the largest 911-related penalty the agency had ever issued, along with smaller fines against Intrado and Verizon for related notification failures.

The other well-documented failure mode isn't a single event; it's a persistent gap. Response times in rural areas are measurably, substantially longer than in cities. One large study of nearly 1.8 million EMS encounters found a median response time around 6 minutes in urban and suburban areas versus roughly 13 minutes in rural ones; broader national data puts rural response times at 20 minutes or more beyond the national average, and for the highest-acuity calls, rural response and transport together can average well over 90 minutes against roughly 69 minutes nationally. Distance explains part of that gap, but so does workforce: the number of volunteer firefighters nationwide, who staff most rural fire and EMS response, fell from roughly 898,000 in 1984 to about 677,000 by 2020, a drop of more than a quarter, while the U.S. population grew by around 40 percent over the same span. In communities where volunteers still provide the overwhelming majority of EMS coverage, that decline shows up directly as slower ambulances, not an abstract statistic.

The scale of it

Americans place an estimated 240 million calls to 911 every year, according to the National Emergency Number Association; in one recent year with comprehensive reporting, 45 states alone logged over 213 million calls delivered to primary call centers. Those calls are answered across roughly 5,700 primary and secondary PSAPs nationwide, staffed by telecommunicators working in shifts around the clock. On the response side, more than 18,000 local EMS agencies, ranging from big-city fire departments to small volunteer squads, field a national EMS workforce estimated at over 800,000 credentialed EMTs and paramedics combined, the large majority certified at the basic EMT level.

Trade-offs and what's next

The unfinished NG911 transition is both the clearest opportunity and the clearest financial headache in this system. Some states have made real progress; others are years behind schedule and far over original cost estimates. California, for one widely reported example, has spent more than $450 million on its NG911 build-out and pushed full rollout to 2030, well past earlier projections, while some smaller localities say outright they cannot afford the transition without dedicated federal money. Congress has sent hundreds of millions of dollars toward PSAP upgrades, but public safety groups argue that funding still falls well short of what a full nationwide transition requires, and every year the older analog system runs in parallel is a year that system also has to be kept alive and staffed.

A genuinely live policy debate is over who should be dispatched for certain categories of call in the first place. Programs like CAHOOTS, running in Eugene and Springfield, Oregon, since 1989, send an unarmed, two-person team, a crisis worker paired with a medic or nurse, for calls involving mental health crises, intoxication, or homelessness that might otherwise default to a police response. In 2019, CAHOOTS handled roughly 24,000 calls routed through the city's 911 and non-emergency lines, about a fifth of the city's total call volume, with police backup needed in only about 150 of those calls. The model has been studied and copied by other cities since, even as CAHOOTS itself has faced local funding disputes, and the underlying question it raises, whether every call a dispatcher fields is best served by an officer with a badge and a weapon, is now a standard part of how cities design their emergency response systems rather than a fringe idea.

Layered under both of those is the staffing problem already mentioned: telecommunicator vacancies running above 50 percent at some centers, and a shrinking volunteer base propping up rural fire and EMS response. A system this dependent on real-time human judgment, at the phone and in the field, doesn't have an easy technological substitute for simply not having enough trained people answering the phone or driving the truck.

Back to the doorway

The strangers who show up at the door, minutes after that first frightened call, arrive at the end of a chain that started with a location signal resolving to the right call center, a scripted set of questions determining what was actually happening, a computer picking the closest available unit out of every vehicle in the county, and, if it's a medical call, a hospital already being warned they're coming. None of that was inevitable. Sixty years ago, the same call might not have connected to anyone in particular, and the vehicle that eventually showed up might have been a hearse driven by someone with no medical training at all. The version that exists now, imperfect, unevenly funded, and running short-staffed in a lot of places, is still the product of a specific, documented decision, made in the 1960s, that this could not keep being left to chance.

The leap: what it replaced, and the work behind it

Before 1968, there was no single number to dial in the worst moment of your life. Every police department, fire station, and ambulance service had its own phone number, and those numbers changed from town to town. Los Angeles alone had around fifty separate police departments with fifty separate numbers. If you did not know the right one, or were away from home, your best option was to dial "0" and hope a telephone operator could figure out which agency to connect you to, and then hope again that whoever answered was not already on another line. That fragility had a public face. When Kitty Genovese was attacked outside her Queens apartment in March 1964, part of what the case exposed, underneath the disputed story about silent neighbors, was that a person who wanted to summon help had no simple way to do it. The President's Commission on Law Enforcement recommended a single nationwide number in 1967, and 911 was designated the next year.

The size of the leap shows up most sharply in who lives now who would have died then. When someone's heart stops today, a telecommunicator can talk a bystander through chest compressions before any vehicle arrives, and starting CPR within a couple of minutes roughly doubles the odds of surviving to hospital discharge; a defibrillator shock within three to five minutes of collapse can push survival to 50 or even 70 percent. None of that scripted, coached, timed response existed in a world of hearse-ambulances and scattered phone numbers. The contrast was stark enough to embarrass the country: by the late 1960s a soldier wounded in Vietnam, reached by helicopter and in surgery within an hour, had a better chance of surviving his injuries than a driver in a car crash on an American highway. Closing that gap took the whole chain this chapter describes, and it runs on people sitting in a chair every hour of every day.

You feel the leap in scenarios so routine they barely register. A parent watches a toddler go limp and dials three digits, and a calm voice is already telling them how to clear the airway while an engine rolls. A neighbor sees smoke and does not have to remember which of a dozen numbers reaches the fire station across the district line. Someone alone with crushing chest pain presses 911 and the nearest unit, chosen by a computer out of every vehicle in the county, is moving before they finish describing it. The morning it fails looks like the September 2025 fiber cut that left multiple Louisiana and Mississippi cities unable to reach 911 at all: the number you have trusted your whole life rings into nothing, and you are back to the 1960s, guessing at who to call. That the failure is rare and the success invisible is the measure of how much steady human work sits behind the dial tone.

Real-world examples and recent developments

The pieces described above, location data, CAD software, and alternatives to a police response, are each built and run by specific named organizations.

  • RapidSOS (2018): a private company that built a data clearinghouse letting smartphone location data reach 911 centers without waiting for a new radio standard. Apple began routing iPhone location data through RapidSOS in 2018, and Google added Android's Emergency Location Service the same year, giving many PSAPs a more precise fix on a cell caller's position than carrier data alone provided. Apple's iOS 12 securely and automatically shares emergency location with 911
  • FirstNet, built with AT&T (nationwide buildout launched 2018, completed March 2023): a dedicated, physically separate broadband network reserved for police, fire, and EMS, created through a public-private partnership between AT&T and the federal First Responder Network Authority after all 50 states and territories opted in. It exists precisely because P25 voice radio, described above, was never built to carry the data, video, and mapping traffic modern responders now depend on. History, First Responder Network Authority
  • CentralSquare Technologies (formed September 5, 2018, from the merger of TriTech, Superion, Zuercher, and Aptean's public-sector business): one of the largest CAD software vendors behind the generic term used earlier in this chapter, with its systems reportedly handling roughly a quarter of all 911 calls received nationwide. Superion, TriTech, Zuercher and Aptean's public sector business merge to form CentralSquare
  • 988 Suicide and Crisis Lifeline (launched July 16, 2022): a three-digit number, modeled deliberately on 911, that routes callers in a mental health crisis to a trained crisis counselor instead of a police dispatcher. It runs as a separate system from 911, though the two are meant to coordinate: a 988 counselor can still request a 911 response if a caller's life is in immediate danger, the same underlying tension CAHOOTS was built to address from the other direction. 988 Suicide & Crisis Lifeline, Federal Communications Commission

Recent developments

  • The September 2025 Louisiana and Mississippi 911 outage: damage to AT&T fiber optic cables knocked out 911 service across multiple parishes, including Louisiana's three largest cities, and several Mississippi counties including Jackson, echoing the 2014 Intrado failure above but rooted in physical infrastructure rather than a software counter, and part of a pattern of 911 outages reported in at least eight states during 2025. Service restored after widespread 911 outages impacted Mississippi and Louisiana
  • AI call triage at 911 centers (2025 to 2026): short-staffed communications centers in Snohomish and Kitsap counties, Washington, began routing non-emergency calls to an AI system built on the Aurelian platform starting in late 2025, and Salem, Oregon, launched a six-month pilot of a similar system, called Ava, in March 2026, aimed at freeing human call takers to focus on genuine emergencies rather than replacing them on emergency lines. Tri-Cities in Wash. Turn to AI for Understaffed 911 Dispatch

Glossary

PSAP (Public Safety Answering Point). The local call center that answers a 911 call, determined automatically by the caller's location.

Telecommunicator. The formal professional title for a 911 dispatcher or call taker.

E911 (Enhanced 911). The system, developed from the 1970s onward, that automatically delivers a caller's location and callback number along with a 911 call.

NG911 (Next Generation 911). The ongoing upgrade of 911 infrastructure from an analog, voice-only network to one built on internet protocols, supporting text, images, and richer location data.

Medical Priority Dispatch System (MPDS). A structured protocol of scripted questions and pre-arrival instructions that guides a call taker through a medical emergency call.

CAD (Computer-Aided Dispatch). Software that tracks the location and availability of every emergency unit in real time and helps a dispatcher select the appropriate one to send.

AVL (Automatic Vehicle Location). The GPS-based system that reports an emergency vehicle's real-time position into a CAD system.

EMT (Emergency Medical Technician). An EMS provider trained in basic life support, including CPR, oxygen delivery, and bleeding control.

Paramedic. An EMS provider trained beyond the EMT level in advanced life support, including IV therapy, cardiac monitoring, and medication administration.

Wall time / ambulance patient offload time. The delay between an ambulance's arrival at a hospital and the formal transfer of patient care to emergency department staff.

P25 (Project 25). A shared technical radio standard that lets police, fire, and EMS radios from different manufacturers and agencies communicate during a joint response.

MSAG (Master Street Address Guide). A database matching street addresses to the correct emergency response zone, used to route 911 calls correctly.

NFPA 1710 / 1720. National Fire Protection Association standards setting response-time benchmarks for career (1710) and volunteer or combination (1720) fire departments.

White paper (EMS context). Shorthand for the 1966 National Academy of Sciences report Accidental Death and Disability: The Neglected Disease of Modern Society, widely credited with launching modern U.S. EMS standards.

Sources and notes

Open questions

  • Exact PSAP counts and annual call volumes vary somewhat by year and by which states report complete data to the National 911 Program; treat the figures here (around 5,700 PSAPs, roughly 240 million calls a year) as representative rather than exact for the current year.
  • The pace, cost, and eventual completion date of the nationwide NG911 transition, and the long-term future of specific alternative-response programs like CAHOOTS, were still actively unsettled policy questions at the time of writing.

This closes the "Health and emergency" part of the book. From the ambulance and the emergency department, the last part turns to something quieter: what keeps the machines themselves running after they're sold. How replacement parts and repair technicians keep products functioning 👉