During review of a sleep study scoring report, the tech notes that the lowest SpO2 is listed as 0%. What should the tech do next?

During review of a sleep study scoring report, the tech notes that the lowest SpO2 is listed as 0%. What should the tech do next?

• A. Check the pulse oximeter before attempting to generate the report again
• B. Review the study and label bad data where the pulse oximeter failed
• C. Do nothing as the report is accurate
• D. Reschedule the patient for a repeat study because of corrupted data

Answer: (B)

The important point is recognizing data integrity when a SpO2 value is implausibly low. A reading of 0% SpO2 almost always signals a sensor problem, data artifact, or corruption rather than a real physiological value. In reviewing a sleep study, the correct action is to examine the SpO2 trace and the accompanying pleth waveform to determine where the signal failed and label those segments as bad data or artifact. This preserves the accuracy of desaturation analysis by distinguishing true physiological events from sensor issues. By labeling the bad data, you can still rely on the rest of the signal for scoring, rather than discarding the entire study or assuming the report is flawless. Checking the device afterward is a good follow-up step, but the immediate, appropriate action in the review process is to annotate the artifact so the report reflects reliable data. Do nothing or reschedule the study would unnecessary react to an artifact when it can be correctly addressed through data labeling.

The important point is recognizing data integrity when a SpO2 value is implausibly low. A reading of 0% SpO2 almost always signals a sensor problem, data artifact, or corruption rather than a real physiological value. In reviewing a sleep study, the correct action is to examine the SpO2 trace and the accompanying pleth waveform to determine where the signal failed and label those segments as bad data or artifact. This preserves the accuracy of desaturation analysis by distinguishing true physiological events from sensor issues.

By labeling the bad data, you can still rely on the rest of the signal for scoring, rather than discarding the entire study or assuming the report is flawless. Checking the device afterward is a good follow-up step, but the immediate, appropriate action in the review process is to annotate the artifact so the report reflects reliable data. Do nothing or reschedule the study would unnecessary react to an artifact when it can be correctly addressed through data labeling.