HA 645 Unit 2 measure specification example

Did the Follow-Up Scan Happen? A Measure Specification for Closing the Loop on Incidental Lung Nodules Found on Emergency Department CT

Student Name

Health Care Administration, Herzing University

HA 645: Quality Improvement and Patient Safety in Healthcare

Instructor Name

January 22, 2026

Did the Follow-Up Scan Happen? A Measure Specification for Closing the Loop on Incidental Lung Nodules Found on Emergency Department CT

Measure Name, Purpose and Type

Name: Incidental Pulmonary Nodule Follow-Up Completion, Emergency Department CT.

Purpose: Northfield Regional Medical Center performs about 3,900 chest CT scans a year in its emergency department. Some show a lung nodule unrelated to the reason for the visit that, under national guidelines, needs a follow-up scan to rule out early cancer (MacMahon et al., 2017). No public rating tracks whether that follow-up happens, and published work shows that it often does not (Blagev et al., 2016). This measure tells Northfield how often it closes the loop.

Type: process. It counts whether a recommended step of care was completed, not whether a patient's outcome changed. In the familiar framework of structure, process and outcome, a process measure is appropriate when the link between the process and the outcome is established, as it is for surveillance of suspicious nodules (Donabedian, 1988). A companion outcome measure, stage at diagnosis for nodule-related cancers, is too rare at Northfield's volume to track usefully.

Denominator

Population: emergency department patients aged 35 years or older who had a CT of the chest, including CT pulmonary angiography, at Northfield, whose finalized radiology report contains the structured incidental-finding statement "Pulmonary nodule, follow-up CT recommended" with a recommended interval.

Window: CT scans finalized during the measurement quarter. Because follow-up intervals run to 12 months, the measure for a given quarter is reported 13 months after the quarter ends. For example, the measure for CTs finalized January through March 2026 is reported in April 2027.

Index date: the date and time the radiology report was finalized.

Numerator

Rule: denominator patients with a completed follow-up chest CT, at Northfield or elsewhere, performed on or before the index date plus the recommended interval plus 30 days.

Date logic: the recommended interval is taken from the structured report field, in months, and converted to days at 30.4 days per month, rounded up. A patient with a 6-to-12-month recommendation is counted as complete if the follow-up CT date is on or before the index date plus 365 days plus 30 days, that is, within 395 days. Where a range is given, the upper bound is used. A follow-up CT performed before the lower bound still counts, since earlier imaging does not fail the patient.

Evidence of completion: a Northfield CT result in the electronic record, or an outside CT report received through the health information exchange or scanned into the record. A patient's statement alone does not count.

Exclusions, Each With Its Reason

1. Known active cancer at the index date, documented in the problem list. Reason: the national nodule guidelines do not apply to patients with a known primary cancer, who are followed by oncology. 2. Immunocompromised status documented at the index date. Reason: the guidelines exclude these patients, whose nodules are often infectious and are managed differently. 3. Death within the follow-up window. Reason: follow-up could not occur. 4. Enrollment in hospice within the window. Reason: surveillance imaging is not consistent with hospice goals of care. 5. A documented informed refusal of follow-up imaging by the patient after notification. Reason: the loop was closed; the patient decided. Refusals are counted and reported separately so that a rising refusal rate is visible.

Not excluded: patients without a primary care physician, patients who are uninsured, and patients who live outside the service area. These are the patients most likely to be lost, and excluding them would make the measure look better by removing the problem it exists to find.

Data Sources and Who Pulls Them

Denominator: radiology information system, structured incidental-finding field and recommended-interval field, pulled monthly by the radiology quality analyst. Exclusions: electronic record problem list (cancer, immunocompromise), registration system (death), hospice enrollment flag, and the nurse navigator's refusal log, pulled by the same analyst. Numerator: electronic record imaging results and the health information exchange query, run by the analyst; outside reports scanned into the record are found through the document index. The analyst's queries are stored in the quality department's shared library so another analyst can reproduce any quarter.

Sampling, Stratification, Baseline and Target

Sampling: none. All eligible cases are included, about 75 to 85 patients a quarter at current volume.

Stratification: results are reported by whether the patient has a primary care physician of record, by insurance status (commercial, Medicare, Medicaid, uninsured), by age band (35 to 64, 65 and older), and by CT type (pulmonary angiography or other chest CT). These strata are chosen because each is a plausible reason for a patient to be lost after discharge.

Baseline period: CTs finalized January through June 2024, so that every follow-up window had closed before measurement. Baseline result: 146 patients in the denominator after exclusions; 45 completed follow-up; 30.8%.

Target: 70% for CTs finalized from July 2026, the first quarter after the planned tracking process begins. Because the main measure takes 13 months to report, a leading measure is reported monthly alongside it: the percentage of denominator patients whose primary care physician, or the patient directly if there is none, received written notice of the nodule and the recommended interval within 14 days of the index date. Target for the leading measure: 95%.

Balancing measure: follow-up CTs ordered for patients whose report did not recommend follow-up, counted quarterly, to detect overuse of imaging and radiation as attention to nodules increases.

Known Weaknesses of the Measure

Three weaknesses are acknowledged so they are not discovered later. First, the denominator depends on radiologists using the structured incidental-finding statement. A nodule described only in free text in the findings section would be missed by the query, and published work suggests such nodules are the least likely to be followed up (Blagev et al., 2016). The radiology department will audit 30 randomly chosen chest CT reports a quarter for nodules described without the structured statement, and report that count beside the measure. Second, follow-up performed at a facility that does not share records through the exchange will be counted as not done unless the navigator obtains the report. The navigator will call a sample of patients recorded as not done each quarter to estimate how often this happens. Third, the 13-month reporting lag means the measure cannot guide weekly work; the leading notification measure exists for that purpose, and the two should always be read together.

References

Blagev, D. P., Lloyd, J. F., Conner, K., Dickerson, J., Adams, D., Stevens, S. M., Woller, S. C., Evans, R. S., & Elliott, C. G. (2016). Follow-up of incidental pulmonary nodules and the radiology report. Journal of the American College of Radiology, 13(2), R18-R24. https://doi.org/10.1016/j.jacr.2015.12.008

Donabedian, A. (1988). The quality of care: How can it be assessed? JAMA, 260(12), 1743-1748. https://doi.org/10.1001/jama.1988.03410120089033

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