Preventing Diagnostic Errors in Ambulatory Care: Deploying a Fail-Safe Notification Tool for Incomplete Radiology Tests
Project Lead: Nina Dadlez, MD, MSc
Project AIM(s):
- To deploy an electronic missed radiology test notification tool across 10 hospital-based ambulatory clinics within 4 months.
- To improve the 90-day radiology test completion rate by 10% within 6 months in the intervention group compared to usual-care controls. We will also survey 100 intervention 100 usual-care clinicians to assess the potential impact of missed-test events and to identify improvement opportunities.
- To conduct 3 focus groups of 8-12 patients selected from among patients who missed a scheduled test in order to understand patients’ reasons for missing the test and workable solutions.
Narrative Description
Patient and Test Characteristics
A total of 3,374 radiology tests were included in our analysis. Intervention group physicians ordered 1769 tests and usual care physicians ordered 1906. Patients in the intervention and usual care groups were well matched in terms of race and need for interpreter services, but intervention patients were slightly younger (median of 57 vs 59, p=0.0002), had a lower percentage of women (68.5% vs 71.6%, p=0.04) and had a slightly higher Medicaid population (25.4% vs 22.4% p<0.001). Intervention and usual care groups had a similar distribution of imaging test utilization by modality with ultrasounds being the most common study followed by mammograms and CTs. 722 patients had multiple tests performed.
Radiology Test Completion
Test completion improved with 18.8% of missed imaging test completed within 90 days among intervention group patients and 16.1% among usual care patients (p=0.03), a 16.8% relative improvement. The overall completion rate in the intervention group was driven primarily by CT (15.3% vs 20.7% p=0.06) and general radiology studies (12% vs 19.6% p=0.02). Statistically significant differences were not observed in mammography (18% vs 15.8% p=0.42), nuclear medicine (14.2% vs 15.8% p=0.68) or ultrasound (16.2% vs 18.6% p=0.30).
In the adjusted analysis, accounting for inter-group differences in sex, age, and insurance types and c clustering with random effects at the level of the physician, the intervention group had a 36% greater odds of test completion than the usual care group (OR 1.36 [1.097-1.682] p=0.005).
Time to Test Completion
Intervention group patients’ tests were completed more quickly than those of the usual care group. Unadjusted Cox regression modeling showed a 19% higher rate of test completion amongst the intervention group (HR 1.19, [1.02-1.39] p=0.03). When the Cox model was adjusted for sex, age and insurance type and accounted for clustering at the level of the physician, there was a 32% higher rate of test completion amongst the intervention group (HR 1.32 [1.10-1.58] p=0.003).
Physician Surveys
Response rates to physician surveys were 59% for the usual care group and 43% for the intervention group. There was a similar awareness of missed radiology tests amongst the usual care and intervention groups (32% vs. 36%).
Patient Focus Groups
Unfortunately, our qualitative research experts concluded that the results could not easily be grouped into actionable interventions given the small sample and inability to achieve thematic saturation.
In summary we learned that an Outpatient Results Notification Tool can significantly improve the rate of 90-day radiology test completion. This was driven primarily by CT scan and general radiology tests. Mammograms and nuclear medicine did not show significant improvements, likely due to existing robust order-follow up processes in place for mammography and patient navigator programs for clinics serving high-risk patients .
We discovered that patients who miss radiology tests may also miss their scheduled focus group times as we had a poor showing. From the patient surveys that were done, we saw that the reasons for missing radiology tests were diverse and did not point to a single actionable intervention.
Peer review publications and presentations
Wasima S, Harvey W, Hon S, LeClair A, Lominac E, Mayer N, Mazzulo J, Roberts K, Weingart S, Dadlez N. “Preventing Diagnostic Errors in Ambulatory Care: Deploying a Fail-Safe Notification Tool for Incomplete Radiology Tests”. Poster Presentation. Institute For Healthcare Improvement Forum. December 2021.
Dadlez NM, Le Clair AM, Wasima S, Mayer N, Harvey WF, Roberts K, Mazzullo J, Lominac E, Koethe BC, Weingart SN. Preventing lost-to-follow up diagnostic imaging in ambulatory care: evaluation of an electronic notification tool. BMJ Open Qual. 2023 Jul;12(3):e002334. doi: 10.1136/bmjoq-2023-002334. PMID: 37463784
Diagnostic quality problem type, failure, or category (symptoms, observed problems, gaps in performance) addressed by the intervention
- Information gathering
- Information integration
- Information interpretation
- Establishing an explanation (diagnosis)
Root causes/causative factors addressed by the intervention
- Workflow (includes testing, follow-up, and referrals)
- Physical environmental or work system factors
- Health information sharing and accessibility via health IT
Setting of the diagnostic quality improvement intervention
- Ambulatory medical care setting (e.g., clinic, office, urgent care)