Measurement & the Score

Unit-hour categories#

Readiness is counted in unit hours and every unit hour falls into these categories. ODS uses only these terms.

• Activated. The unit hour is scheduled and staffed.
• Serviceable. Activated and fit for assignment: crew, equipment, and supplies all conforming.
• Utilized. Serviceable and committed to a call.
• Excessive task time. Utilized well beyond the interval the task should take. The threshold is set by the system. A common starting point is 150 percent of the allotted interval, not a fixed rule. Examples: waiting with the patient at an ED, extended on-scene waits.
• Out of service. Activated but not serviceable.
• Lost unit hours. The sum of out of service and excessive task time.

A fire company continuing care on scene until a slow transport unit arrives, or a crew waiting with a patient in a hospital hallway, is utilized with excessive task time, contributing lost unit hours (distinct from out of service). The readiness deficit does not belong to the committed unit. It belongs to the party that controls the delay: the ED at the hospital and, on scene, whoever is responsible for the deployment that determines when the transport unit arrives.

"Properly deployed," in the definition of readiness, includes position. Checking it asks two things: did the communications center assign the unit per the deployment plan and did the unit comply with the assignment? The first belongs to communications and the second to the unit's producer.

Reading compliance across tiers#

A response-time standard is a matrix. Moving from dense areas to less dense ones, more time is allotted and fewer ambulances cover the ground. If the tiers were calibrated coherently, compliance would be roughly level across urban, suburban, and wilderness zones. If instead the urban life-threat tier is the hardest to hit while the outlying tiers come easily and hitting the urban tier sweeps up the rest by happenstance, the tiers are miscalibrated, with the urban tier too tight or the outlying tiers too loose. It does not signal that the producer is failing and it requires no accusation of bad faith: the numbers show the standard was not built coherently.

The functional difference to a patient between 90.02%, 90.01%, and 89.99% compliance is nil. An exemption should be granted on its merits and not because compliance is close to the line.

Operational response time belongs to the procuring entity#

Unit-hour procurement (quantity), the deployment plan (location), response-time accountability, and billing travel together.

No individual producer controls that clock; it is driven by caller interrogation, triage, dispatch, the posting plan, call classification, and zone definitions. Most of those are deployment and system-design decisions. For that reason a producer should not be held to operational response time. Operational response time against the standard belongs to the procuring entity, which holds the deployment plan (the three parties are defined in System Structure).

This is also why a "compliant" response can hide a broken one. In a system that allows the transport unit 30 to 40 minutes on distant or low-acuity calls, a fire company can arrive in 8 minutes and then continue care on scene for half an hour until the ambulance arrives. The fire company is utilized with excessive task time, contributing lost unit hours, while the contract still scores the response compliant. The number is met while the system is failing. Reading operational response time without the readiness picture of each subsystem (see System Structure) is how that failure stays invisible.

In-scope metrics live in every subsystem#

Everything a party does control is fair to hold it to and each of the five subsystems along the patient's path has its own in-scope set, on both the readiness and response sides. These are the metrics that flow into the optimal deployment score. These are examples:

• Public. Readiness: citizen recognition, public awareness and education. Response: bystander intervention (CPR/AED), 911 activation.
• Communications. Readiness: communication capacity, deployment plan. Response: EMD and triage quality, the call-processing intervals (public safety answering point, transfer caller, initialize request, interrogate caller, dispatch first response, dispatch transport response), pre-arrival instructions. A request is initialized once the call taker has recorded when and where help is needed and the nature of the incident.
• First response. Readiness: serviceable unit hours, mutual-aid agreements. Response: mobilizing (turnout in the fire service, out-of-chute in private EMS), en route, on-scene assessment, transfer of care, return to service.
• Patient transport. Readiness: serviceable unit hours, mutual-aid agreements. Response: mobilizing, en route, on scene, preparing transport, transport, return to service.
• Definitive care. Readiness: bed capacity. Response: preparing for the patient, transfer of care, providing care.
• Operational response time. Two metrics, first response time and transport response time, each measured against the standard. Each spans communications and the subsystem that responds and belongs to no single subsystem. Its intervals are scored in the subsystems that own them and the whole clock is reported beside the score.

The common thread: a party answers for its own subsystem's readiness and response, not for another subsystem's and not for the operational response times. Wherever a party touches the patient (first response, patient transport, definitive care), clinical quality is measured as QA/QI: did the crew recognize the condition, select the correct protocol, and follow it? Each of these is real, measurable, and inside the party's control.

Split the clock where responsibility transfers#

Where one subsystem depends on another to take the patient, a delay on one side impacts the other (System Structure). The measurement rule: cut an interval at the event where responsibility for the patient changes hands and put each resulting clock on the party that controls its delay. Neither side can then shave its clock without the delay landing on its own scorecard.

Patient transport to definitive care. The time a crew spends at the hospital is commonly reported as one number. It is two clocks, split at the transfer of care; the nurse signature may be used as a surrogate:

1. Transfer of care. At destination until care is transferred. Under federal law (EMTALA), the hospital's obligation begins when the patient arrives and care is requested, whether or not the patient has been moved off the ambulance stretcher (CMS, 2006). In practice the crew stays with the patient until the hospital accepts care, so this clock measures how long definitive care takes to accept the patient. It is definitive care's metric, reported to the hospital.
2. Return to service. Transfer of care to crew available. This is a different job: breaking down and rebuilding the rig. It is patient transport's metric, with its own standard.

The two clocks are reported separately and are not added together.

A practical refinement: set a total time-at-hospital budget split per side, for example 30 minutes as 15 + 15 (20 + 20 is also seen). The rationale is labor. One crew member stays with the patient and owns the transfer-of-care wait; the other breaks down and rebuilds the rig. If the hospital moves fast, the crew still gets its share to ready the unit.

First response to patient transport. The same split applies. There is value in capturing the transfer of patient care from first response to patient transport, which could be by way of a signature. Capturing it establishes a well-defined scope of control for both first response and patient transport.

A second metric sits beside it: first-responder on-scene time before transport arrival. This is the time a first-response unit spends on scene before the transport unit arrives. That time is not in first response's scope of control, so it is not first response's metric. It is part of transport response time: it is set by when the transport unit arrives and it belongs to whoever is responsible for the deployment. It answers a different question than the transfer of care does, so the two are reported separately.

Patient outcomes#

Each subsystem's producer owns that subsystem's readiness and response metrics and, where it delivers care, recognition, protocol selection, and adherence. Medical direction owns protocol quality, the procuring entity owns operational response time, and no single party owns patient outcome.

Patient outcomes matter and ODS measures them. They are feedback to medical direction and the raw material for research into protocol efficacy: the loop by which a system learns whether its clinical protocols work and improves them.

Outcomes are not a metric to grade responders by. Outcome is dominated by the patient's underlying disease and physiology and by the whole system's time to care, none of which a responder controls. Ranking or penalizing crews on outcome drives risk-averse care and defensive documentation and lets a poor protocol pass as a poor crew.

What a responder is held to is the clinical work inside their control:

• Recognition. Did the crew correctly identify the patient's condition?
• Protocol selection. Did they choose the correct protocol for it?
• Adherence. Did they follow that protocol correctly?

If recognition, selection, and adherence are all high but outcomes are still poor, the problem is the protocol rather than the crew and revising it is medical direction's responsibility, informed by the outcome data.