Every emergency response in this country starts with a person in a chair answering a telephone, and almost everything that happens afterwards is shaped by what that person gets in the first thirty seconds. This piece covers the working mechanics of that job as I have seen it from the radio and quality assurance side: how a call is controlled when the caller cannot control themselves, what protocol-driven questioning actually does, what pre-arrival instructions accomplish before the engine arrives, what the telecommunicator owns during a mayday, and why the staffing problem in 911 centers is a field problem too.

What happens before anything rolls

The phrase first first responder gets used at banquets, and it is literally accurate about sequence. No apparatus moves, no crew gets alerted, no hospital gets notified and no mutual aid request leaves the county until a telecommunicator has answered a line, established a location, decided what kind of problem this is, selected a response from a matrix somebody wrote in advance, and put it into a computer aided dispatch system. That is a triage decision made on incomplete information from a witness who is frequently wrong about what they are looking at, and it determines how many units go, how fast they go, and which of the calls holding in queue goes first.

The federal occupational classification does not describe the work that way. The Standard Occupational Classification system maintained through the Office of Management and Budget places public safety telecommunicators inside the office and administrative support major group rather than with protective service occupations, and that placement survived the last major revision of the classification. The bill commonly known as the 911 SAVES Act, which would direct that they be reclassified as a protective service occupation, has been introduced in more than one Congress, and as of my last check on it the reclassification had not been enacted, so verify the current status on Congress.gov rather than taking my word for it. Some states have separately moved telecommunicators into public safety or hazardous duty retirement categories by statute, which is a state-by-state question to put to your own human resources director.

The clock on the work is written down in standards, and the numbers are worth knowing before you argue about performance with anybody. NFPA consolidated its emergency services communications material into NFPA 1225, Standard for Emergency Services Communications, which absorbed the content previously carried in NFPA 1221 and in the telecommunicator professional qualifications standard, and it expresses answering and alarm processing expectations as percentile thresholds rather than averages, such as a large majority of alarms answered within fifteen seconds and processed within about a minute, with defined exceptions for call types that legitimately take longer. Take the exact figures, the exception list and the measurement method from the edition your jurisdiction has actually adopted, because the exceptions are where most of the local argument lives.

The call as it sounds from the other chair

A caller in crisis loses the ability to organize information in the order anybody needs it. People give you their phone number when you ask for an address, they give you a landmark that means nothing outside their own family, they describe the room they are standing in rather than the building, and they answer a question you asked two questions ago. The call taker has to impose a structure on that without sounding like a bureaucrat, which is done with short imperative sentences, with the caller’s name once you have it, and with repetition persistence, meaning you ask the same question in the same words as many times as it takes because rephrasing gives a panicked caller a new thing to process.

Location and a callback number come before everything else, and the reason is unsentimental. Wireless calls drop, batteries die, callers walk out of coverage, people hang up because they think they have told you enough, and a caller who is about to be assaulted may have seconds. If the line dies after the address and the words unconscious and not breathing, a response is already moving, whereas a beautifully complete narrative with no verified location is a recording of an emergency rather than a dispatch. The separate article on this site about how caller location reaches dispatch covers the technical side of that, including where the automatic location still fails, and none of it removes the need to get the address out of the caller’s mouth.

The job is also described as multitasking in every hiring advertisement, and the honest description is rapid task switching with a cost at each switch. A telecommunicator working a combined position may be holding a phone call, watching two or three talkgroups, typing into CAD, answering a ringing administrative line and keeping an eye on a unit that has not answered a status check, and each interruption costs a second or two of reorientation plus a chance of dropping a thread. How much of that happens is a design decision made by the center rather than a personal quality of the employee, because splitting call taking from radio, adding a position, or moving alarm monitoring off the busiest console changes the switching load directly.

Guidecards, determinant codes, and what the questions are for

Structured call interrogation in emergency medical dispatch grew out of work done in Salt Lake City in the late 1970s by Jeff Clawson, whose protocol became the medical priority dispatch system now maintained by the International Academies of Emergency Dispatch and published through Priority Dispatch Corporation. Two other families are in wide use in the United States, being the emergency medical dispatch guidecards published by APCO and the protocols published by PowerPhone, and there are fire and police versions in the same lineage. Find out which one your center licenses and what version it is running, because the vocabulary your call takers use, the codes on your CAD ticket and the way your responses are tiered all come out of that product rather than out of local invention.

What the protocol produces is a chief complaint and a determinant code, and neither is a diagnosis. In the medical priority system the codes run from omega at the low end through alpha, bravo, charlie and delta up to echo for the most immediately life-threatening findings, and your agency’s response matrix maps those codes to what rolls and how fast. The value of that is not precision about the patient, since a caller cannot give you precision, but consistency about the response, so that the same set of answers produces the same assignment on a Tuesday afternoon as it does at three in the morning with a different call taker. Deviation is sometimes correct and it has to be documented when it happens, which is a policy question for the center director and, for medical calls, for the physician who provides medical oversight of the emergency medical dispatch program.

The complaint I hear most from field crews and from callers is that the questions delay the response, and in a properly built center that is a myth, so label it as a myth whenever it comes up. In most configurations the call is entered and sent on the location and the problem, and the remainder of the interrogation happens after the assignment has gone out, which means the questioning runs in parallel with the turnout rather than in front of it. Where that is not true, and there are centers where a single person on a single position cannot send until the interrogation ends, the fix is staffing and console design rather than skipping the questions. Training and certification requirements for emergency medical dispatch vary by state, so verify what your state requires with your state office of emergency medical services or your state 911 authority rather than assuming the protocol vendor’s certificate satisfies it.

Pre-arrival instructions and the minutes before the engine

Pre-arrival instructions are the part of the job that saves lives directly, and cardiac arrest is the clearest case. The American Heart Association has published a policy statement specifically on telecommunicator CPR and carries telecommunicator instruction in its resuscitation guidelines, directing that telecommunicators be trained to recognize likely arrest, to treat agonal gasping as a sign of arrest rather than as breathing, and to coach compression-only CPR for adults with suspected out-of-hospital arrest. Take the current wording from the guidelines in force rather than from a laminated card of unknown vintage, because the recognition language in particular has been sharpened over successive revisions.

The agonal breathing problem is the one I would drill hardest. A bystander standing over a patient who is gasping every few seconds will tell you he is breathing, in good faith, and if the call taker accepts that answer the call becomes a lower priority and no compressions happen. The protocols address it by asking about breathing in a way that forces a description rather than a yes, and the quality assurance file is where you find out whether your own call takers are asking it that way under load. Everything after recognition is logistics that sounds trivial and is not, including getting the patient flat on the floor, getting the caller’s phone onto speaker, getting the dog shut in a bedroom, getting the front door unlocked, and sending a second person for an AED when the call comes from a place likely to have one.

The same structure applies to childbirth, choking, severe external bleeding with direct pressure, and the overdose call where naloxone is in the house, and in each case the instruction is scripted because a scripted instruction given calmly beats an improvised one given by a person who is also typing. The limits are real and should be stated plainly to field crews who think instructions are magic, since some callers physically cannot do compressions, some refuse, some are not with the patient, and some hang up. A call taker who follows the protocol on a patient who dies anyway did the job correctly, and the quality assurance review has to be written to say so, otherwise your best people learn to fear the review of the worst calls.

What the telecommunicator owns during a mayday

This site has a separate article on building a mayday SOP that crews can execute, and my concern here is the half of that SOP written for the person in the chair, which is the half most commonly missing. When the word mayday comes across a talkgroup the telecommunicator has a defined set of actions to perform, and the useful ones are concrete: acknowledge it so the transmitting member knows a human being heard, tone or announce emergency traffic to clear the channel if the incident commander has not, confirm the incident commander received it, record the identifying information the member gives, mark the time, and start a written running log separate from the CAD entry. Those actions need to be on a card at the console, in the order they are done, because the console operator is the only participant who can be counted on to have a keyboard and a clock in front of them.

Radio emergency activation is part of the same sequence and depends on housekeeping nobody enjoys. When a member presses the emergency button on a P25 radio, what the dispatch console displays is an alias drawn from the system’s subscriber database, and if that database still says a radio ID belongs to a unit that was reassigned two years ago, the dispatcher receives a meaningless string at the worst possible moment. Ask your radio system administrator when the alias table was last reconciled against the actual radio issue records, and ask what the console operator is supposed to do with an emergency activation that arrives with no voice behind it, because that is a common event and it needs a defined answer rather than an improvised one.

The path that surprises people is the telephone. The published investigations of the June 2007 Sofa Super Store fire in Charleston, South Carolina, including the NIOSH firefighter fatality investigation and the post-incident assessment the city commissioned, record that a trapped firefighter reached 911 by cellular telephone during the fire, and I would ask every dispatch supervisor to sit with that fact rather than skim it. A call like that lands on a 911 or administrative line at the center, possibly at a law enforcement position, from a caller whose radio may be working poorly or not at all, and whether the information reaches the incident commander in useful time depends entirely on whether somebody thought about that relay in advance and wrote it down. Read the reports themselves for what each one establishes about the sequence and the timing, since my summary here is only the reason to go look.

The recording, the clock, and the QA file

The 911 center is the only participant in an incident that produces a complete, time-stamped, independently recorded account of what was said, and that record is the foundation of both quality assurance and any honest after action review. The CAD timestamps and the logging recorder together give you when the call was answered, when the assignment went out, when each unit acknowledged and marked en route and arrived, when command was established, and when each significant radio transmission occurred, which is material no crew’s memory can match. Retention periods and public records treatment for those recordings are governed by state law and are covered in the separate article on recorded lines and records retention, so establish what yours requires before anyone assumes a recording from eight months ago still exists.

A quality assurance program that works has a shape I would recognize anywhere, and it starts with sampling that is both random and targeted. Random review of a defined number of calls per telecommunicator per month tells you about the ordinary day, and one hundred percent review of a short list of high-risk call types, such as cardiac arrest, structure fire, officer or firefighter emergency and any call that generated a complaint, tells you about the days that matter. Scoring is against the protocol the center licenses and against local policy, with a written form, and the output goes back to the individual in a conversation rather than into a file they never see.

The failure mode I have seen most often in this work is a program that has been quietly converted into a disciplinary instrument, at which point the feedback loop closes. People stop volunteering their own bad calls, supervisors stop hearing about near misses, and the review of the worst calls becomes something everyone dreads and nobody learns from. Keeping quality assurance separate from discipline is a policy decision that has to be made in writing by whoever runs the center, with a stated threshold for when a performance issue leaves the quality assurance track, and it is worth checking whether your center has that written down or is relying on the current supervisor’s temperament.

Authorized, filled, and released to work alone

I am not going to print a national vacancy percentage for 911 centers, because the surveys that produce those numbers come from organizations including NENA and the International Academies of Emergency Dispatch with different methods and different response rates, and the figures they publish differ enough that averaging them would mislead you. What you need is your own center’s three numbers, which are positions authorized in the budget, positions filled, and personnel actually released to work a console alone. The third number is the one that matters for the schedule, it is always the smallest, and in many centers nobody outside the communications division has ever been shown it.

The gap between filled and released is a training pipeline measured in months rather than weeks. A new hire in a typical center works through classroom instruction, protocol certification, telecommunicator certification where the state requires it, and then a long period of supervised work on call taking before moving to radio, and through all of that they occupy a trainer who would otherwise be staffing a position. That is why a successful hiring event does not improve next month’s schedule, and why losing a two-year employee costs more than the recruitment line item suggests. When the staffing shortfall gets covered with mandatory overtime, extended shifts and cancelled leave, the people most likely to leave are the ones who are already trained, which is the loop that makes these shortages self-sustaining.

Some of the drivers are outside any single agency’s control, including pay scales set by a county commission and a federal classification that does not describe the work. Several are inside it, and those are the ones I would pursue first: whether the schedule is predictable far enough out for people to plan a life, whether approved leave actually gets taken, whether a supervisor is physically on the floor during the hardest hours, whether the quality assurance program is punitive, and whether part-time or floater positions exist to absorb vacancies. Telecommunicator wellness has its own article on this site and belongs in the same conversation, because the cumulative exposure of the job is a retention issue and not only a health issue.

What the field owes the floor

The cheapest improvement available to a fire, EMS or police agency is putting field officers in the chair for a few hours and putting telecommunicators on an apparatus for a shift, because most of the friction between the two comes from each side having no picture of the other’s constraints. An officer who has watched a single console handle two working incidents and a phone queue stops asking why the status check came when it did, and a telecommunicator who has ridden a structure fire understands why a crew on air in a basement is not going to answer a radio check. Both directions of that exchange need a named person to schedule them, otherwise they get discussed warmly for years and never happen.

The second thing the field owes the floor is outcome information and inclusion in review. A telecommunicator who coached twelve minutes of compressions on a call at four in the morning usually never finds out what happened, and a policy that routes that information back, within whatever privacy constraints apply in your state, is worth more than a plaque during telecommunicator week. When your agency holds an after action review of a significant incident, invite by name the telecommunicators who worked it, give them the recording and the timeline in advance, and let them speak to the radio traffic, since they are the only participants with an authoritative record of what was said and when.

The third thing is drilling together, which is where the mayday material above becomes real. A mayday drill in which the dispatcher’s half is played by a training officer standing in a classroom proves nothing about what the on-duty console operator will do at two in the morning, so run at least one drill a year where the actual working position handles the actual traffic on an actual talkgroup, with prior coordination so nobody thinks it is real, and pull the recording afterwards for both sides to review. During large incidents, assigning a telecommunicator to the incident talkgroup and nothing else is the same principle applied at scale, and it is a decision the incident commander can request as readily as a second engine.

What to do at your agency

• Have your communications manager produce three numbers in writing this month, being positions authorized, positions filled and personnel released to work a console alone, with the date counted, and read those numbers into the minutes of the chiefs meeting or 911 board meeting that already meets.
• Ask your center director for the written console card that tells a telecommunicator what to do when the word mayday comes across a talkgroup, and if there is no card, write one paragraph into your existing mayday SOP naming who acknowledges, who clears the channel, who confirms the incident commander heard it and who keeps the written time log.
• Ask your radio system administrator when the subscriber alias table was last reconciled against actual radio issue records, and have any mismatched aliases corrected so that an emergency activation displays a unit designation a dispatcher can act on.
• Have the training officer and the communications supervisor jointly schedule one mayday drill this quarter in which the on-duty telecommunicator works the traffic live on a real talkgroup with prior coordination, and pull the logging recorder audio afterwards for a joint review with the crews involved.
• Ask which call interrogation protocol your center licenses, which version it is running, and who provides physician oversight for the emergency medical dispatch program, then request the last quarterly quality assurance summary and read the compliance findings for cardiac arrest calls.
• Name the telecommunicators who worked your next significant incident in the after action review invitation, send them the incident recording and timeline beforehand, and put their radio traffic observations in the written report.
• Assign one officer per shift to spend four hours at a dispatch console this quarter, with the communications supervisor holding the sign-up sheet, and reciprocate by scheduling telecommunicator ride-alongs on apparatus in the same period.

Takeaways

• Nothing in the field moves until a telecommunicator has established a location, classified a problem and selected a response, which makes the first triage decision of every incident a dispatch decision.
• The federal Standard Occupational Classification places public safety telecommunicators in the office and administrative support group rather than with protective service occupations, and the reclassification bill known as the 911 SAVES Act has been introduced more than once, so check Congress.gov for its current status before repeating anything about it.
• NFPA 1225, Standard for Emergency Services Communications, now carries the answering and alarm processing expectations previously found in NFPA 1221 along with telecommunicator qualifications, and the exact percentile figures and exceptions should be read from the edition your jurisdiction adopted.
• A verified location and callback number come before the interrogation because calls end without warning, and everything else in the narrative can be reconstructed while the location cannot.
• The belief that protocol questions delay the response is a myth in centers where the assignment goes out on the location and problem and the interrogation continues afterwards, and where it is true the cause is console design and staffing rather than the protocol.
• Agonal gasping is routinely reported by callers as breathing, so the recognition questions and the quality assurance review of cardiac arrest calls are the difference between compressions starting in the first two minutes and not starting at all.
• The published investigations of the 2007 Sofa Super Store fire in Charleston record that a trapped firefighter reached 911 by cellular telephone, which is the reason every mayday SOP needs a written path from a 911 call taker to the incident commander.
• Vacancy figures published by different national organizations use different methods and disagree, so run the argument at your own agency on your own authorized, filled and released-to-solo counts with the date they were taken.