Which Three Methods Can Be Used to Gather Data to Evaluate CPR Quality?

The three methods are direct or remote human observation, physiologic monitoring and feedback devices, and defibrillator or manikin data review with post-event debriefing. In practice these are often layered: a trained observer scores compressions in real time, a sensor puck or monitor streams depth and rate, and software later reconstructs the entire resuscitation. CPR quality is not a single number. It is a set of measurable variables, including compression rate, depth, recoil, fraction, ventilation volume, and peri-shock pause, and each method captures a different slice of that set. Choosing among them depends on whether you need immediacy, objectivity, or system-level learning.

How each method captures CPR quality

Human observation works because a trained person can apply a checklist to what they see and hear. The observer times compressions against a clock or metronome, counts depth against a visual estimate, and notes interruptions. The strength is judgment: a reviewer can catch context a sensor misses, such as a rescuer leaning on the chest or a poorly timed pulse check. The weakness is also judgment. Visual depth estimation is unreliable, and a single observer cannot track rate, depth, recoil, and ventilation simultaneously without help. Remote review of video, either live from a telehealth link or recorded in a simulation lab, extends observation beyond the bedside but adds a delay and depends on camera angle.

Physiologic and feedback monitoring works by transducing motion or pressure into a signal. An accelerometer in a compression puck or defibrillator pad measures vertical displacement and derives rate and depth. Some systems add force sensors, and capnography estimates ventilation adequacy from exhaled carbon dioxide. The device compares the signal to a target range and displays a prompt, creating a closed feedback loop. The mechanism matters: accelerometers infer depth from acceleration, so chest compliance, mattress softness, and rescuer hand position can bias the reading. That is why device measurements are precise in relative terms but not automatically accurate in absolute terms.

Data review and debriefing works by aggregating the other two into a timeline. Defibrillators and monitors log every compression, shock, and pause with timestamps. Manikins log the same variables in training. Software turns the log into a report, and a team reviews it against guidelines. The why is learning theory: feedback improves performance most when it is specific, timely, and tied to a review conversation rather than a raw number. A report nobody discusses changes little.

What the numbers show

Guideline targets give the methods a common yardstick. The American Heart Association sets adult compression rate at 100 to 120 per minute and depth at at least 2 inches (5 cm), with full recoil between compressions. Chest compression fraction, the proportion of resuscitation time spent compressing, is targeted above 60 percent and ideally above 80 percent. Peri-shock pause is kept under 10 seconds, and ventilation is titrated to visible chest rise with capnography guidance where available (American Heart Association, 2020 Guidelines for CPR and ECC).

These thresholds explain why one method rarely suffices. A rate of 118 per minute satisfies the target but may signal fatigue if depth is falling. A depth of 2.1 inches satisfies the target but may hide incomplete recoil. Fraction above 80 percent can still coexist with long individual pauses that matter more than the average. Each number is a threshold, and each method reconciles thresholds differently: manual counting is fast but coarse, device feedback is precise but sensor-dependent, and debriefing is comprehensive but retrospective.

Where the methods diverge

Real resuscitations expose gaps that clean training data hide. Motion artifacts from vehicle transport or patient movement corrupt accelerometer signals. Capnography reflects circulation as much as ventilation, so a low value may indicate poor compressions rather than poor breathing. Observer scoring in a crowded room is subject to drift and anchoring, where the first impression colors later counts.

Device readings can be biased by mattress compressibility, so a soft bed inflates apparent depth. Manual counts degrade after roughly two minutes of continuous compressions as rescuer fatigue sets in. Debriefing data is only as good as the clock sync between devices, and unsynced clocks scramble the timeline. Training manikins calibrated differently from clinical devices can make a good rescuer look poor, or the reverse. Video review raises privacy and consent questions that can block access to footage.

The counterintuitive detail is that more data can degrade learning. Teams shown long, dense reports often fixate on a single red metric and miss the interaction between rate, depth, and pauses. A short, focused debrief on two or three variables outperforms a comprehensive dashboard for most clinical teams.

Common misconceptions

A frequent error is treating device feedback as ground truth. Feedback devices improve performance, but their output is an estimate shaped by physics and calibration. Another is assuming observation is obsolete. Observation remains the only method that captures team communication, role clarity, and leadership, which no sensor measures.

A third misconception is that debriefing must be exhaustive. Structured, brief debriefs that name one strength and one change are more sustainable than hour-long reviews. Finally, people conflate training evaluation with clinical evaluation. Manikin data tells you about skill acquisition; defibrillator data tells you about real-world delivery. They answer different questions and should not be substituted for each other.

Comparison of the three methods

Timeline of a quality evaluation

Practical actions

Start with a defined variable set rather than every available metric. Pick rate, depth, fraction, and peri-shock pause, then add ventilation only where capnography is available. Brief observers on a one-page checklist and rotate them every few minutes to limit fatigue.

For live feedback, verify device placement and confirm the mattress or surface is consistent across cases so readings are comparable. Set prompts to the guideline ranges, and treat a prompt as a question to investigate rather than an automatic correction. After each event, run a hot debrief within minutes and a cold debrief within 72 hours. Use aggregate reports quarterly to spot patterns, and retrain when a team misses a threshold on two consecutive reviews. These steps keep the three methods reinforcing each other rather than competing.

Limitations and what the evidence cannot settle

No method captures CPR quality perfectly. Observation is subjective, devices are sensor-dependent, and debriefing is retrospective. Cross-device comparisons are difficult because manufacturers use different algorithms and calibration standards, and published accuracy figures vary by setting. The evidence supports feedback and debriefing as performance-improvement tools, but it does not establish a single best method for every environment. Teams should document which method they used and interpret results within that method's known bias. Where local data are absent, prefer qualitative review over invented benchmarks.

This article is general educational information about quality measurement, not a diagnosis, treatment plan, or clinical protocol. It does not promise improved outcomes from any method. For clinical decisions or device selection, consult a qualified local clinician or resuscitation educator, and for an active emergency call local emergency services immediately.