HA 645 Unit 4 safety event review example

A Nine-Millimeter Nodule in the Final Report, a Negative Scan in the Discharge Summary: A Safety Event Review of a Follow-Up That Never Happened

Student Name

Health Care Administration, Herzing University

HA 645: Quality Improvement and Patient Safety in Healthcare

Instructor Name

February 5, 2026

A Nine-Millimeter Nodule in the Final Report, a Negative Scan in the Discharge Summary: A Safety Event Review of a Follow-Up That Never Happened

The Event

A 58-year-old patient came to Northfield's emergency department overnight in March 2023 with chest pain and shortness of breath. A CT pulmonary angiogram showed no pulmonary embolism. The final radiology report, issued the next morning after the patient had gone home, also described a 9-millimeter solid nodule in the right upper lobe and recommended further evaluation within three months, consistent with national guidelines for a solid nodule of that size in a patient of that age (MacMahon et al., 2017). No one acted on the recommendation. Fifteen months later, the patient returned with coughing up blood; imaging showed a 2.8-centimeter mass, and the diagnosis was stage IIIA lung adenocarcinoma. Northfield rated the harm severe and filed a serious event report with the state. People are described by role only. This review was written with the emergency department, radiology and nursing staff who were involved, and they reviewed it before it was finalized.

Sequence

March 11, 2023, 11:52 p.m.: patient arrives; chest pain protocol. March 12, 12:40 a.m.: CT pulmonary angiogram performed. 1:25 a.m.: overnight teleradiology preliminary read: "No pulmonary embolism." Preliminary reads at night address the question asked and do not routinely describe incidental findings. 1:58 a.m.: emergency physician reviews preliminary read and troponin results; discharge decision. 2:14 a.m.: patient discharged with instructions to see a primary care physician; the discharge summary, generated automatically, lists the CT result as the preliminary impression. The patient's record lists no primary care physician. 7:40 a.m.: the Northfield radiologist finalizes the report; impression item 1, no pulmonary embolism; item 2, 9-millimeter right upper lobe nodule, recommend evaluation within three months. 7:41 a.m.: a "final report available" item enters the emergency department callback worklist. 9:10 a.m.: the callback nurse closes the item with the note "no change in PE result." March 15: the final report is released to the patient portal; the portal shows the patient never logged in. June 2024: patient returns with hemoptysis; the mass is found.

Defenses, One at a Time

Six defenses stood between the nodule and a missed follow-up. Each is examined, including those that worked.

The final radiology report worked: the radiologist found the nodule and wrote a clear recommendation. It was placed in the impression, where it should be. But it was written as free text, not with the structured incidental-finding statement, so it did not enter any tracking list.

The preliminary read did not fail; it was not designed to report incidental findings. The gap it created, between what the emergency team knew at discharge and what the final report said, is a design feature of overnight coverage.

The callback worklist fired and was overridden. The item reached the callback nurse, who had 64 worklist items that morning. The worklist design shows only whether the "primary finding" changed between preliminary and final reports, and the primary finding, no embolism, had not changed. Closing the item was consistent with how the worklist presents information and how nurses had been taught to use it.

The discharge summary did not carry the final report. It was built at discharge from the preliminary impression and was not updated when the final report arrived. Even if the patient had given it to a physician, it would have said the scan was negative.

Portal release worked as designed and depended on the patient logging in, which the patient did not do.

Primary care follow-up did not exist. The patient had no primary care physician, and no one at Northfield was assigned to find one or to track the recommendation for patients without one.

Contributing Conditions as Properties of the System

The review found five conditions, each a property of how the work is set up. A preliminary-to-final gap overnight, during which patients are discharged on preliminary results. A worklist that compares only the primary finding, so an added incidental finding looks like no change. Worklist volume that makes line-by-line reading of every final report unrealistic in the time available. A discharge summary that is generated once and never updated. And no owner for follow-up recommendations for patients without a primary care physician. None of these conditions depended on one person's carelessness. Harm occurred because gaps in several defenses lined up on the same patient on the same morning, the pattern long described in safety science as holes in successive layers of defense lining up (Reason, 2000).

Corrective Actions, Strongest First

Actions are ordered by how much they rely on memory and vigilance, from those that change the system so the failure is hard to repeat, down to those that ask people to be more careful (National Patient Safety Foundation, 2015).

Stronger. 1. Require the structured incidental-finding statement for any nodule needing follow-up: the reporting system will not allow a chest CT report to be finalized with the word nodule in the impression unless the structured statement is completed. Owner: chair of radiology. Date: June 30, 2026. 2. Route every structured incidental finding to a nurse navigator's tracking list, independent of the emergency department worklist, with the navigator responsible until follow-up is done or documented as refused. Owner: director of imaging services. Date: July 31, 2026.

Intermediate. 3. Redesign the callback worklist to flag any new impression item, not only a change in the primary finding. Owner: emergency department medical director with information services. Date: May 31, 2026. 4. Regenerate the discharge summary and send an amended copy when a final report adds an impression item after discharge. Owner: chief medical information officer. Date: September 30, 2026.

Weaker. 5. Educate emergency and callback nursing staff on incidental findings and the new worklist. Owner: emergency department nurse educator. Date: within 30 days of action 3. This action is listed last because, on its own, it would ask the same people to catch the same problem with the same tools.

The patient and family were informed of the review's findings in writing, as Pennsylvania law requires for serious events, and met with the chief medical officer.

References

MacMahon, H., Naidich, D. P., Goo, J. M., Lee, K. S., Leung, A. N. C., Mayo, J. R., Mehta, A. C., Ohno, Y., Powell, C. A., Prokop, M., Rubin, G. D., Schaefer-Prokop, C. M., Travis, W. D., Van Schil, P. E., & Bankier, A. A. (2017). Guidelines for management of incidental pulmonary nodules detected on CT images: From the Fleischner Society 2017. Radiology, 284(1), 228-243. https://doi.org/10.1148/radiol.2017161659

National Patient Safety Foundation. (2015). RCA2: Improving root cause analyses and actions to prevent harm.

Reason, J. (2000). Human error: Models and management. BMJ, 320(7237), 768-770. https://doi.org/10.1136/bmj.320.7237.768