Closing the referral loop means tracking a referral from the clinician’s request until the specialist’s report reaches the referring clinician and leads to a documented decision. For independent practices, the hard part is rarely sending the referral. It is knowing, at any moment, what has happened since.
A primary care office faxes a referral to a surgical group. The confirmation page prints. Two weeks later, nobody can say whether the patient has an appointment, whether the surgeon needs more records, or whether a consult note is on its way.
The gap is well documented. Researchers at the Regenstrief Institute estimate that between a third and half of referral loops are never closed [1]. CMS now measures the loop by a single outcome: whether the referring clinician actually receives the specialist’s report [2].
This guide sets out a seven-step referral workflow, a status model, the metrics worth tracking, and how ClinicianCore, a secure, HIPAA-compliant unified clinical communication platform built exclusively for physicians, supports the handoffs between practices.
Key takeaways
- Sending a referral confirms transmission, not receipt, scheduling or completion.
- The 2026 CMS closing the referral loop measure counts one outcome: the referring clinician receiving the specialist’s report.
- Every open referral needs a named owner, a current status and a dated next action.
- Track visit completion and consult note receipt as separate milestones, with separate denominators.
- In recent primary care studies, on-time completion of ordered referrals and tests ranged from roughly 40% to 65%.
- Secure communication between organizations supports the workflow; clinical documentation stays in each practice’s EHR.
“A referral is a clinical question handed to a colleague. Until the answer comes back and someone acts on it, that question is still open, whatever the fax confirmation says.”
Dr. Kevin Halow MD MBA FACS CMO & Co-Founder ClinicianCore – Surgeon, Military Veteran
What does closing the referral loop mean?
Closing the referral loop is the process of tracking a referral from the initial request through the specialist’s evaluation, the return of the consult report, and the referring team’s review and follow-up.
For quality reporting, the definition is narrower. The 2026 CMS electronic clinical quality measure CMS50v14 reports the percentage of patients with a referral for whom the referring clinician receives a report from the clinician to whom the patient was referred [2]. Three details in the specification matter for daily operations:
- Only the first referral made between January 1 and October 31 counts, which leaves time for the report to come back within the measurement year [2].
- The report must relate to that first referral and be received after the referral was made [2].
- A procedural report, such as a colonoscopy or diabetic eye exam report, can satisfy the measure without a separate consult letter [2].
A scheduled appointment does not meet the measure. Neither does a completed visit whose note never arrives. For patient care, practices should go one step further than the measure: confirm that the report reached the right clinician, was reviewed, and produced a documented next step.
What does the research show about referral loop gaps?
Recent primary care studies point to the same pattern. Completion falls short at several stages, not only at scheduling. The three studies below came out of the same AHRQ-funded closed-loop diagnostics learning laboratory [4].
| Study | Setting | Finding |
|---|---|---|
| Zhong et al., JAMA Network Open, 2023 [3] | 4,133 test and referral orders at two Boston primary care sites, 2020 to 2021 | 58.4% of orders from in-person visits and 42.6% from telehealth visits were completed within the set time frame |
| Amat et al., Joint Commission Journal on Quality and Patient Safety, 2024 [4] | 12,282 test and referral orders at two academic primary care practices, 2018 to 2021 | 52.9% of resident orders and 58.4% of faculty orders were completed on time |
| Campbell et al., Journal of General Internal Medicine, 2024 [5] | 770 patients with rectal bleeding at two primary care practices | 22.3% did not get timely recommended workup; among those open loops, 36% had no procedure ordered, 26% were ordered but never scheduled and 18% were scheduled but cancelled or missed |
Three cautions apply. These are single-system studies, not national rates for independent practices. Missing documentation does not prove that a patient went without care elsewhere. And the studies measured completion of tests and referrals, a related but different milestone from consult note receipt.
The practical lesson holds across all three. The JAMA Network Open authors found completion between 40% and 65% for every test type, against an acceptable threshold of 80% used in population health metrics [3]. They also noted that patients received no reminder after the visit to schedule the referral [3]. Tracking the order is not the same as tracking the outcome.
Why do referrals break down between independent practices?
Independent practices refer to many specialty groups, imaging centers and labs, usually on different EHRs. Each relationship adds a handoff, and each handoff can fail quietly. Labs see the same problem from the receiving side, as our post on diagnostic lab referral strategy explains.
Physicians describe the barriers plainly. In a study summarized by the Regenstrief Institute in 2024, primary care physicians and specialists named EHRs that cannot share information across health systems, continued reliance on fax machines and patients carrying their own paperwork. They also reported inconsistent information-sharing policies between organizations, and breakdowns where a consultation was completed but the referring clinician was never told the result [1].
National data tell the same story. Fewer than one in five office-based physicians engaged in all four domains of interoperability (finding, sending, receiving and integrating patient information) in 2021, and only 8% of family physicians surveyed said information from clinicians on a different EHR was very easy to use [6]. In 2024, just 13% of family medicine physicians reported ideal interoperability, meaning outside information often arrived automatically in their EHR and was easy to find and use [7].
Authorization adds its own delay. In the AMA’s latest prior authorization survey, the 2025 edition reported in May 2026, 95% of physicians said prior authorization delays access to necessary care, and the work consumed an average of 13 hours of physician and staff time each week [8].
Underneath all of this sit a handful of questions that nobody can answer quickly:
- Did the receiving practice acknowledge the referral?
- Is it ready for clinical review, or is something missing?
- Did the appointment happen?
- Has the consult note been sent, and did it reach the right clinician?
- Who owns the next step?
A fax confirmation answers a transmission question. It answers none of these. Our earlier post on moving referrals off the fax machine covers the cost of that gap in more detail.
How do you close the referral loop, step by step?
1. Start with a clear clinical question
A referral should state why the specialist’s input is needed and what the referring clinician wants answered. “Persistent symptoms after six weeks of treatment; is surgical review indicated?” gives the receiving team direction. A specialty name alone does not.
Include what that question requires: urgency, pertinent history, medication list, recent results and coverage details. Agree with frequent partners on a pre-scheduling checklist so the same records are not requested twice. For time-sensitive concerns, agree on an escalation route and confirm that a named person received the request.
2. Confirm receipt and assign an owner
Give every referral a named owner in the referring practice, with backup coverage when that person is out. Ask the receiving practice to acknowledge receipt and name its next action, for example: “Received. We need the recent imaging report before scheduling.” That one line separates transmission from acceptance and gives the referring office a concrete point for follow-up.
3. Make missing information visible
Replace a vague “pending” with the actual blocker: waiting for clinician clarification, an imaging report, authorization or patient scheduling. Record who owns it and the follow-up date. Keep authorization status separate from referral status. Approval removes a barrier; it does not book a visit or return a note.
4. Track scheduling through to a confirmed appointment
“Patient contacted” and “appointment scheduled” are different states. A voicemail is not a booking. A patient may ask for a later date, decline, or lack transportation, and each outcome needs a different response. When scheduling fails, the referring clinician needs enough detail to decide on another outreach attempt, a different specialist or a change to the plan. In the 2024 rectal bleeding study, about a quarter of open loops were orders that were never scheduled [5].
5. Confirm that the consultation happened
A booked visit can still be cancelled, missed or moved. In the same study, 18% of open loops were scheduled procedures that were cancelled or not kept [5]. Ask the receiving practice to confirm the visit or report the exception, and reconcile expected visits against confirmed ones every week.
6. Request and receive the specialist consult note
Track the report as its own milestone: pending, sent, or received in the right record. Record the visit date and the report-receipt date separately, so a completed visit with missing recommendations stands out. Practices reporting CMS50v14 should follow the full specification, including the rule that the report must relate to the first referral of the year [2].
7. Route the report for review and follow-up
Receipt should trigger a defined internal handoff. Route the report to the referring clinician or a designated covering clinician, document the review, and assign an owner to each resulting action: a medication change, further testing, another appointment or a call to the patient through the practice’s usual channels. A report filed unread is an open loop with a closed status.
Which referral statuses explain the next step?
Adapt these labels to your existing system. They are suggested workflow states for a referral management workflow, not a description of any platform’s automated features.
| Referral status | Meaning | Next action |
|---|---|---|
| Requested | Clinician has ordered the referral | Prepare and send the relevant information |
| Sent | Request has been transmitted | Confirm receipt |
| Received | Partner practice has acknowledged it | Confirm acceptance or identify missing items |
| Information needed | A specific requirement is outstanding | Assign and resolve the missing item |
| Scheduling | Outreach is under way | Confirm an appointment or document the barrier |
| Scheduled | Appointment date is confirmed | Reconcile the visit outcome |
| Visit completed | Specialist evaluation occurred | Obtain the report |
| Report received | Report is available to the referring team | Route for clinician review |
| Reviewed, follow-up assigned | Recommendations have been assessed | Complete or track resulting tasks |
| Exception documented | Referral did not proceed as intended | Record the clinician-directed disposition |
Keep a declined or cancelled referral distinct from a completed one, so reports do not hide unfinished care.
Which referral metrics should independent practices track?
A practice can send every referral on time and still struggle to get specialist recommendations back. Start with a small set of measures that show where work stalls.
| Metric | What it shows | Suggested definition |
|---|---|---|
| Receipt confirmation rate | Whether partners acknowledge requests | Acknowledged referrals ÷ referrals sent |
| Referral-to-appointment time | How long patients wait for a booked visit | Median days from request to confirmed date |
| Visit completion rate | Whether scheduled care happens | Completed visits ÷ scheduled visits |
| Report receipt rate | Whether specialist findings come back | Reports received ÷ completed visits |
| Visit-to-report time | How long the referring team waits | Median days from visit to report receipt |
| Overdue referrals by stage | Which handoff needs attention | Open referrals past target, grouped by status |
| Unresolved exceptions | Where a clinician decision is needed | Exceptions open past the agreed review date |
Define the denominator and time window for each metric. Report receipt among completed visits answers a different question from report receipt among all referrals sent. Judge only cohorts that have had time to progress, and break results down by specialty and partner practice when volume allows. As a reference point, the JAMA Network Open authors cited 80% as an acceptable population health threshold and 95% as an ideal goal for loop closure [3].
When can a physician-to-physician consult answer the question instead?
Not every referral needs a visit. When the question is focused, such as interpreting a result or adjusting a treatment plan, a structured consultation between physicians may answer it sooner and spare the patient a trip. ConsultCore, part of ClinicianCore, a secure, HIPAA-compliant unified clinical communication platform built exclusively for physicians, records the clinical context, the specialist’s reasoning and the decision in one place. When payer requirements are met, that record can support interprofessional consultation coding.
When the specialist decides a visit is needed, the consult becomes the start of a better-prepared referral. For more on why timing matters, see how specialist consultation delays affect patient outcomes.
How does ClinicianCore support referral coordination?
ClinicianCore is a secure, HIPAA-compliant unified clinical communication platform built exclusively for physicians. Three of its modules map to the workflow above:
- Practice HQ is the secure workspace for internal coordination: messaging, voice, video and schedules for physicians and practice teams. It supports the handoff from coordinator to clinician when a report arrives.
- ConsultCore handles physician-to-physician consultation, with a structured record of context, reasoning and decision.
- Trusted Entity connects verified outside organizations, such as specialist practices, labs, imaging centers and pharmacies, through explicit relationships. Each referral can have its own topic thread with the right people, so records, questions, scheduling updates and the consult note stay together in an audit-ready record.
A referring office can ask about missing records in the same thread where the receiving office posts its update, and a colleague covering the work can see the full history. That is what unified clinical communication looks like across practice boundaries. The status model and metrics in this guide are operating practices; teams still need to assign owners, post updates and document clinical decisions in the right record.
ClinicianCore is built for physicians and practice teams. Patient scheduling and outreach stay with each practice’s established processes, and clinical documentation stays in the EHR.
What does a closed referral loop look like in practice?
This is an illustrative workflow, not a customer case study.
Dr. Mitchell, a primary care physician, refers a patient to Dr. Chen’s surgical group with a specific question and an urgency level. The coordinator sends the relevant records and opens a referral thread with the connected surgical practice.
The surgical team confirms receipt and asks for a missing imaging report. The coordinator supplies it and sets a follow-up date. Once the appointment is booked, the surgical practice posts the date in the same thread.
After the visit, the primary care team confirms that the evaluation took place and marks the report as pending. When Dr. Chen’s consult note arrives, it is routed to Dr. Mitchell, who reviews the recommendations and assigns the follow-up tasks. At every step, both offices can see the last confirmed event and who acts next.
How should an independent practice get started?
Start with one frequent referral partner. Agree on required information, status language, owners, backup coverage and escalation. After a set period, review a manageable batch of referrals together, find where requests stall and whether consult notes come back, and adjust before expanding. Small, consistent changes like these are where real medical practice efficiency gains come from.
Closing the referral loop takes a visible chain of responsibility, from the first request to the specialist’s recommendation and the referring clinician’s response. ClinicianCore, a secure, HIPAA-compliant unified clinical communication platform built exclusively for physicians, gives that chain a shared place to live between practices through the Trusted Entity network.
Frequently asked questions
What does closing the referral loop mean?
Closing the referral loop means following a referral until the specialist’s report reaches the referring clinician and that clinician acts on it. The CMS measure CMS50v14 counts report receipt, not scheduling (CMS, 2026). ClinicianCore’s Trusted Entity network keeps each referral conversation with connected organizations in one secure thread.
Is a referral closed once the appointment is scheduled?
No, a scheduled appointment is only one milestone. In a Boston primary care study, just 42.6% of tests and referrals ordered at telehealth visits were completed on time (JAMA Network Open, 2023). ClinicianCore’s Trusted Entity threads keep visit updates and the consult note attached to the original request.
Who should track referrals in an independent practice?
A named referral coordinator with backup coverage should own tracking, while clinicians keep responsibility for clinical decisions. Researchers estimate that between a third and half of referral loops are never closed (Regenstrief Institute, 2024). ClinicianCore’s Practice HQ supports the internal handoffs between coordinators and physicians when a report arrives.
Why do referrals break down between independent practices?
Referrals break down because each handoff crosses a different system, team and process. Only 8% of family physicians said information from clinicians on a different EHR was very easy to use (ONC, 2024). ClinicianCore’s Trusted Entity network gives verified partner organizations a shared, audit-ready thread for each referral.
Does referral tracking require both practices to use the same EHR?
No, practices can coordinate referrals securely across different EHRs, as long as each documents clinical care in its own record. Fewer than one in five office-based physicians used all four interoperability functions in 2021 (ONC, 2024). ClinicianCore works alongside the EHR, keeping referral communication with connected partners organized.
How should a practice measure referral loop performance?
Measure each stage separately: receipt confirmation, scheduling, visit completion, report receipt and clinician review. In one primary care study, 26% of open loops were procedures ordered but never scheduled (Journal of General Internal Medicine, 2024). ClinicianCore’s referral threads show coordinators which stage each referral has reached.
References
[1] Between a third and half of referral loops are not closed; physicians cite EHRs that cannot share information across systems, fax reliance and patients carrying paperwork as key barriers. Source: Regenstrief Institute, Closing the referral loop: perspectives and experiences of primary care and specialist physicians (Savoy et al., International Journal of Medical Informatics, December 2023), February 2024
[2] CMS50v14 measures the percentage of patients with a referral for whom the referring clinician receives the specialist’s report; first referral January 1 to October 31; procedural reports can satisfy the numerator. Source: Centers for Medicare & Medicaid Services, eCQM CMS50v14: Closing the Referral Loop: Receipt of Specialist Report, 2026 measurement period
[3] Of 4,133 orders, 42.6% from telehealth visits and 58.4% from in-person visits were completed on time; completion was 40% to 65% for all test types against an 80% acceptable threshold and 95% ideal goal; patients received no post-visit reminder. Source: JAMA Network Open, Zhong et al., Completion of Recommended Tests and Referrals in Telehealth vs In-Person Visits, 2023
[4] Across 12,282 test and referral orders (2018 to 2021), resident orders were completed on time 52.9% of the time versus 58.4% for faculty (Amat et al., Joint Commission Journal on Quality and Patient Safety, 2024). Source: Agency for Healthcare Research and Quality, R18 Closed Loop Diagnostics: AHRQ R18 Patient Safety Learning Laboratories (grant period 2022 to 2024)
[5] 22.3% of 770 patients with rectal bleeding did not receive timely workup; 36% of open loops had no procedure ordered, 26% were ordered but never scheduled, 18% were scheduled but cancelled or not kept. Source: Journal of General Internal Medicine, Campbell et al., Completion Rates and Timeliness of Diagnostic Colonoscopies for Rectal Bleeding in Primary Care, 2024
[6] Fewer than one in five physicians engaged in all four interoperability domains in 2021; only 8% of family physicians found information from clinicians on a different EHR very easy to use. Source: ONC (HHS), Raising the Bar on Interoperability: A Decade of Data Show that Sometimes Isn’t Good Enough, 2024
[7] In 2024, just 13% of family medicine physicians reported ideal interoperability. Source: ASTP/ONC (HHS), A Decade of Data Examined: The Evolution of Health IT in the United States, December 2024
[8] 95% of physicians said prior authorization delays access to necessary care; it consumes an average of 13 hours of physician and staff time per week. Source: American Medical Association, Only 1 in 3 doctors trusts insurers’ prior authorization promises (2025 AMA prior authorization physician survey), May 2026