kellyemrick.com ↗
The care
completion gap.
A test can be complete while the next step in care is still missing. Explore where follow-up stalls, what the evidence shows, and what better follow-through could change.
Hypothetical due cohort. An episode is a recommendation, not a unique patient.
What happened after the recommendation?
Four measures tell different parts of the story. A documented resolution does not always mean the original service was performed.
One cohort. Seven outcomes.
Every episode belongs to one outcome category.
- Verified on time120 · 60%
- Verified late20 · 10%
- Clinical alternative20 · 10%
- Informed decline10 · 5%
- Transfer awaiting service15 · 7.5%
- External care unverified5 · 2.5%
- Other unresolved10 · 5%
Three lenses on follow-through
Same denominator. Different definitions.
How these measures are calculated
On time: verified on-time services ÷ all episodes in the due cohort.
Eventual: verified on-time plus late services ÷ the same cohort.
Disposition: verified services, clinician-reviewed alternatives, and informed declines ÷ the same cohort.
Original due dates and cohort rules must be retained. Missing dates need explicit handling before applying these measures.
Where the next action belongs
Inspect an outcome and the evidence needed for its next step.
| Outcome | Episodes | Share | Status | Evidence or next action |
|---|---|---|---|---|
| Verified on time | 120 | 60% | Verified | Retain the service evidence and applicable due date. |
| Verified late | 20 | 10% | Verified | Retain the actual service date; preserve the late designation. |
| Clinical alternative | 20 | 10% | Disposition | Document clinician review and the alternative plan. |
| Informed decline | 10 | 5% | Disposition | Document the informed choice and appropriate communication. |
| Transfer awaiting service | 15 | 7.5% | Unresolved | Confirm responsibility; verify the service or another valid disposition. |
| External care unverified | 5 | 2.5% | Unresolved | Obtain and review the required external service evidence. |
| Other unresolved | 10 | 5% | Unresolved | Clarify the barrier, accepted owner, and next action. |
The pathway breakdown is illustrative and is not a comparison of actual clinical performance. Table filters narrow inspection; downloads retain the full selected cohort. Research studies are shown separately in Evidence.
Match the response to the obstacle
Select a possible barrier. These are response options, not measured causes in this cohort.
Ask who is missing from the average
No demographic or access-group observations are supplied, so this dashboard does not invent disparity estimates.
What an equity-ready dataset needs
- Group-level eligible episodes and verified outcomes using the same cohort rules.
- Documented missingness for language, access, or other approved characteristics.
- Enough observations to interpret estimates and apply appropriate disclosure protection.
- Information about transportation, communication, and administrative barriers without assuming why a gap exists.
Patient burden and access context: Kyle & Frakt, 2021; Sinnott et al., 2026.
Every handoff needs an owner.
Explore four stages of follow-through. This is a conceptual care pathway, not a numerical patient funnel.
Make the next step explicit
Define the recommended action, its clinical context, the applicable due date, and the patient communication plan.
A clear recommendation and due-date basis, with the original date retained for the reporting cohort.
Watch for: Ambiguous instructions, missing dates, or an assumption that the patient already knows the next step.
Two clocks. Two different questions.
The actual service date determines whether completion was on time or late against the applicable due date.
The verification date determines what was known in a particular reporting snapshot. Late evidence of on-time care can support a logged correction.
Closure requires more than a click
Ownership may change. The service remains unresolved until it is verified or another valid disposition is documented.
A report that care occurred is a useful signal. Verification requirements must be met before it counts as a verified service.
Record the clinical review or informed decision separately. Neither is counted as the originally recommended service.
Process guidance: ASTP/ONC, 2025 · IHI & NPSF, 2017. Clinical deadlines require an appropriate clinical pathway; no universal deadline is supplied here.
What could better outreach change?
Adjust the assumptions to explore additional late completions in the unresolved cohort. This is a deterministic scenario, not a prediction from the trials.
Set the assumptions
Model assumptions and limits
Additional late completions = unresolved episodes × reach probability × completion probability among those reached.
Only currently unresolved episodes are eligible. Expected counts may be fractional. No gain is applied to already completed episodes, alternatives, or declines.
Contact hours = unresolved episodes × average contact attempts per episode × minutes per attempt ÷ 60. This excludes treatment time and other administrative work.
No time horizon, clinical outcome, cash saving, or causal effect is inferred. The original on-time rate stays fixed.
Additional late completions
Modeledexpected additional completions from 30 unresolved episodes
Explore the reach–completion relationship
Expected additional late completions as reach changes. Completion among reached stays at the selected assumption.
At 80% reach and 50% completion among reached, the model adds 12 late completions.
View scenario curve data
| Reach | Additional late completions |
|---|---|
| 0% | 0 |
| 25% | 3.8 |
| 50% | 7.5 |
| 75% | 11.3 |
| 100% | 15 |
What to test in practice
A successful contact does not establish completion. Check both steps when evaluating an outreach change.
Contact workload is an input to planning. It is not evidence of cash savings or staff capacity available elsewhere.
Late recovery improves eventual completion. Preserve the original on-time result when reporting the intervention.
What the evidence actually shows.
Study findings retain their own populations and observation windows. They do not change when demonstration filters or sliders change.
Active outreach improved follow-up
Adjusted completion within 120 days after enrollment. The observation window is not a recommended clinical deadline.
Study data and interpretation
| Arm | n | Adjusted completion |
|---|---|---|
| Usual care | 2,702 | 22.9% |
| EHR reminders | 3,254 | 22.7% |
| Reminders + outreach | 2,569 | 31.0% |
| Reminders + outreach + navigation | 3,455 | 31.4% |
Navigation versus outreach: +0.4 percentage points, 95% CI −3.6 to 4.2. This does not establish equivalence.
Navigation helped close part of the gap
One-year colonoscopy completion. Navigation n = 479; usual care n = 488. Results reflect one health-center organization and a pandemic-affected study period.
Effort does not guarantee improvement
Zubkoff et al. (2024) found no overall improvement in primary follow-up outcomes in a multisite VA implementation trial. Track outcomes alongside implementation activity.
Tracking systems need local validation
Aunger et al. (2025) found promising pulmonary-nodule follow-up interventions, while noting substantial bias in many nonrandomized studies and heterogeneous methods.
More follow-up is not always better care
Ganguli et al. (2019) describe burdens associated with cascades after incidental findings. Include clinical review, appropriate alternatives, and balancing measures.
Evidence map: design, contribution, and limits
| Source | Design | Why it matters | Key limitation |
|---|---|---|---|
| Callen et al., 2012 | Systematic review | Historical follow-up failures and variable definitions | Older, heterogeneous studies |
| Atlas et al., 2023 | Cluster randomized trial | Outreach comparison | Already-overdue screening cohort |
| Coronado et al., 2025 | Randomized trial | Navigation after abnormal stool testing | One health-center organization |
| Zubkoff et al., 2024 | Stepped-wedge randomized trial | No overall improvement in primary outcomes | VA setting and variable local changes |
| Varon et al., 2024 | Systematic review and meta-analysis | Cervical screening follow-up interventions | Mixed designs and outcomes |
| Aunger et al., 2025 | Systematic review | Pulmonary-nodule follow-up interventions | High bias in many observational studies |
| Shah et al., 2026 | Retrospective implementation report | Sustained incidental-finding program | Abstract-level endpoint verification |
| Tyler et al., 2023 | Network meta-analysis | Transitions and outcome diversity | Indirect comparisons and heterogeneity |
| Ganguli et al., 2019 | Physician survey | Burdens of care cascades | Self-report and recall |
| Kyle and Frakt, 2021 | Cross-sectional survey | Patient administrative burden | Historical self-reported data |
| Sinnott et al., 2026 | Qualitative interviews | Access and digital workload | Context-specific interpretation |
| IHI and NPSF, 2017 | Implementation guide | Referral process design | Guidance, not effect estimation |
| ASTP/ONC, 2025 | SAFER implementation guide | Responsibility and follow-up safety | Guidance, not dashboard validation |
All 13 references and source links
- Assistant Secretary for Technology Policy/Office of the National Coordinator for Health Information Technology. (2025). 2025 SAFER guide: Test results reporting and follow-up. Source ↗Official 2025 release page. The linked guide retains August 2024 footers; source version checked October 11, 2026.
- Atlas, S. J., Tosteson, A. N. A., Wright, A., Orav, E. J., Burdick, T. E., Zhao, W., Hort, S. J., Wint, A. J., Smith, R. E., Chang, F. Y., Aman, D. G., Thillaiyapillai, M., Diamond, C. J., Zhou, L., & Haas, J. S. (2023). A multilevel primary care intervention to improve follow-up of overdue abnormal cancer screening test results: A cluster randomized clinical trial. JAMA, 330(14), 1348–1358. Source ↗
- Aunger, J., Yip, K. P., Dosanjh, K., Scandrett, K., Ungureanu, B., Newnham, M., & Turner, A. M. (2025). Interventions to improve adherence to clinical guidelines for the management and follow-up of pulmonary nodules: A systematic review. CHEST, 168(1), 248–268. Source ↗
- Callen, J. L., Westbrook, J. I., Georgiou, A., & Li, J. (2012). Failure to follow-up test results for ambulatory patients: A systematic review. Journal of General Internal Medicine, 27, 1334–1348. Source ↗
- Coronado, G. D., Petrik, A. F., Thompson, J. H., Leo, M. C., Slaughter, M., Gautom, P., Hussain, S. A., Mosso, L., Gibbs, J., Yadav, N., Mummadi, R. R., Johnson, E. S., & Jimenez, R. (2025). Patient navigation to improve colonoscopy completion following an abnormal stool-test result: A randomized controlled trial. Annals of Internal Medicine, 178(5), 645–654. Source ↗
- Ganguli, I., Simpkin, A. L., Lupo, C., Weissman, A., Mainor, A. J., Orav, E. J., Rosenthal, M. B., Colla, C. H., & Sequist, T. D. (2019). Cascades of care after incidental findings in a US national survey of physicians. JAMA Network Open, 2(10), e1913325. Source ↗
- Institute for Healthcare Improvement & National Patient Safety Foundation. (2017). Closing the loop: A guide to safer ambulatory referrals in the EHR era. Institute for Healthcare Improvement. Source ↗
- Kyle, M. A., & Frakt, A. B. (2021). Patient administrative burden in the US health care system. Health Services Research, 56(5), 755–765. Source ↗
- Shah, R. M., Dehadrai, A., Keshwani, A., Schmitz, B., & Carr, J. C. (2026). Closing the loop: 9-year outcomes of an electronic medical record-based protocol for reporting incidental imaging findings. Journal of the American College of Radiology, 23(1), 45–51. Source ↗Online publication September 3, 2025; January 2026 issue. Abstract reviewed.
- Sinnott, C., Ansari, A., Price, E., Horder, K., Alboksmaty, A., Willars, J., Beech, J., Alderwick, H., & Dixon-Woods, M. (2026). Experiences of access to general practice in England: Qualitative study and implications for the NHS 10 year plan. BMJ, 392, e087367. Source ↗
- Tyler, N., Hodkinson, A., Planner, C., Angelakis, I., Keyworth, C., Hall, A., Jones, P. P., Wright, O. G., Keers, R., Blakeman, T., & Panagioti, M. (2023). Transitional care interventions from hospital to community to reduce health care use and improve patient outcomes: A systematic review and network meta-analysis. JAMA Network Open, 6(11), e2344825. Source ↗
- Varon, M. L., Geng, Y., Fellman, B. M., Troisi, C., Fernandez, M. E., Li, R., Reininger, B., Schmeler, K. M., & Allanson, E. (2024). Interventions to increase follow-up of abnormal cervical cancer screening results: A systematic literature review and meta-analysis. PLOS ONE, 19(2), e0291931. Source ↗
- Zubkoff, L., Zimolzak, A. J., Meyer, A. N. D., Sloane, J., Shahid, U., Giardina, T., Memon, S. A., Scott, T. M., Murphy, D. R., & Singh, H. (2024). A virtual breakthrough series collaborative for missed test results: A stepped-wedge cluster-randomized clinical trial. JAMA Network Open, 7(10), e2440269. Source ↗
Research method: targeted narrative synthesis, not a systematic review of all care-completion interventions. Shah et al. (2026) was reviewed at abstract level. Guidance informs process design; it does not validate this dashboard or its scenario model.
Build a follow-through plan.
Use the current cohort and scenario as a starting point. Check the preparation steps, then download a concise plan for discussion.
Prepare a focused pilot
0 of 6 readyThese checks reflect your preparation choices. They are not an external readiness assessment.
All three pathways
200 recommendation episodes
30 unresolved episodes
Current eventual completion: 70%
Scenario eventual completion: 76%
Expected additional late completions: 12
Expected contact workload: 8 hours
Historical on-time rate: 60%, unchanged
Keep the first version focused
One nonurgent pathway, explicit eligibility, a retained original due date, and enough observation time.
Review flagged and unflagged examples. Check outside care, ownership, duplicate episodes, and missing dates.
Set responsibility for review, define the contact process, and document balancing measures before judging results.