What Is Patient Leakage?

August 26, 2026

Patient leakage is every patient who needed your care and ended up somewhere else. It happens in three places: website visitors who research you and leave without booking, existing patients who go dormant in your EHR, and referred patients who never complete the journey to your schedule. Most practices never measure it, which is exactly why it persists.

Running a multi-provider orthopedic practice specifically? The subspecialty-routing and post-op recall version of this problem is in Orthopedic Patient Leakage: Where It Happens and How to Stop It.

What counts as patient leakage

A patient counts as leaked when three things are true: they needed care your practice provides, they had already reached you in some way, and they did not receive that care from you. Reaching you can mean visiting your website, calling your front desk, arriving with a referral, or sitting in your EHR as a past patient who is due for care. Whether they booked with a competitor or put the care off, the result for your practice is the same.

Two situations do not count. A patient who needs a service you do not offer was never yours to lose, and a clinically appropriate referral out to another specialist is good care, not leakage. For how we stop the website leak specifically, see our page on stopping patient leakage on your practice website.

Where leakage actually happens

Leakage is not one hole. It is three.

The website leak. Most practice websites convert less than 1% of their traffic into booked patients. The other 99% includes people actively comparing you against two competitors tonight. 82% of patients try to book outside office hours, and when nobody engages them, they book with whoever answers first.

The EHR leak. Years of past patients sit dormant in your system: the opposite knee, the annual exam never scheduled, the PT plan abandoned at week three. They already chose you once. Nobody has time to call them.

The referral leak. Patients referred to your practice who stall between the referring provider and your front desk. Every step a patient has to take alone is a step where some of them disappear. See referral management software for what closes this leak specifically.

Patient leakage vs referral leakage vs care leakage

These three terms often get used interchangeably. They describe different problems, and each one needs a different fix.

Patient leakage is the broadest term. It covers every patient who needed your care and did not get it from you, through any channel: the website, the EHR, or a referral. Health systems often use the same phrase for patients who leave their network for outside providers.

Referral leakage is the subset that arrives with a physician's recommendation already in hand and never becomes a completed visit. Losing one costs more than losing an anonymous website visitor, because someone with clinical authority already sent that patient to you. It is also common. In a study of 103,737 referral scheduling attempts in a large health system, only 34.8% resulted in a documented completed appointment (Patel et al., Journal of General Internal Medicine, 2018). A 2011 review reported that 25 to 50 percent of referring physicians did not know whether their patients had actually seen the specialist (Mehrotra, Forrest, and Lin, Milbank Quarterly, 2011). Our guide to what referral leakage is and how to fix it goes deeper, and our referral leakage statistics roundup lists the published numbers with their sources.

Care leakage is care your practice already ordered, recommended, or scheduled that did not happen on time: a diagnostic that was never scheduled, a surveillance recall nobody made, a post-op follow-up that lapsed. Because the appointment was never created, it never shows up as a no-show. Read what care leakage is and how specialty practices stop it for the full breakdown.

Why practices underestimate it

Leakage is invisible in the metrics practices already track. Your schedule looks full, so nothing seems wrong. But a full schedule and a leaking funnel coexist easily: the schedule fills with whoever gets through, not with the highest-need patients who quietly left. The only way to see leakage is to measure the gap between the demand that reached you and the patients you actually saw.

What causes patient leakage

Leakage rarely comes from one bad decision. It comes from small points of friction, each at a moment when a patient is ready to act.

No response when patients reach out. Patient demand does not keep office hours. A visitor who researches you at night and finds only a phone number and a contact form has no way to act until morning, and by then a competitor may have answered.

Long waits for a first appointment. Delays give patients time to look elsewhere. AMN Healthcare's 2025 wait time survey found it takes an average of 31 days to schedule a physician appointment across 15 of the largest US metropolitan areas (AMN Healthcare, 2025). In the 2018 referral study above, wait times for completed appointments were significantly lower than wait times for incomplete ones.

Scheduling that depends on a phone call. When the only path to an appointment is calling during business hours or waiting for a callback, every missed connection is a chance for the patient to give up. Forms and generic chat widgets collect contact details, but someone still has to call the patient back.

Handoffs nobody owns. Referrals stall between the referring office and your front desk. In that same study, 38.9% of referral scheduling attempts lacked an appointment date. Dormant patients leak the same way inside the practice: everyone agrees they should come back, and nobody has time to call them.

No measurement. Leakage that nobody measures has no owner, so it persists. Health systems report the same blind spot: in About Healthcare's Patient Leakage and Keepage Report, 90% of health systems said they were not highly confident in their visibility into patient leakage (About Healthcare, 2021/22).

How to measure your leakage

Start with three numbers: monthly website visitors, booked appointments that came from the website, and dormant patients in your EHR who are overdue for care. The first two give your conversion rate. If it is near 1%, you are typical, and that is the problem. We built a patient leakage calculator that runs the math for your practice in about two minutes. For the fix-it side, read our playbook on how to reduce patient leakage.

Leakage metrics worth tracking

Once you have a baseline, give each leak its own number. A single blended figure hides which leak is largest, and each one has a different fix.

  • Website conversion rate: appointments booked from the website divided by monthly website visitors.
  • After-hours response: how many patient inquiries arrive outside office hours, and how many of those get a response before the next business day.
  • Referral completion rate: referrals that become completed visits divided by referrals received, tracked alongside the days from referral to appointment.
  • Recall rate: dormant or overdue patients who book again divided by all patients overdue for care.
  • Source attribution: the share of new patients you can trace back to where they first found you.

Review these monthly. A monthly view shows whether a change on the website, at the front desk, or in referral intake actually moved the number. To put a value on each leak, read how much patient leakage costs a practice.

How patient activation stops the leak

Patient activation treats all three leaks as one job. The Inbound Agent engages every website visitor 24/7, screens their needs, and books them before they bounce to a competitor. The Recall Agent works the EHR and brings dormant patients back into open slots. The Outbound Agent generates net-new demand and traces every patient from first click to completed care, so leakage stops being invisible.

Practices typically move from under 1% website conversion to 8 to 12% engagement. Baldwin Bone and Joint turned the traffic it already had into 263 qualified surgical leads in a single quarter with a 60% booking rate. Same website, same marketing spend. The leak was the difference.

Frequently asked questions

What is patient leakage?

Patient leakage is every patient who needed your care and ended up somewhere else. It happens in three places: website visitors who research you and leave without booking, existing patients who go dormant in your EHR, and referred patients who never complete the journey to your schedule.

What is the difference between patient leakage, referral leakage, and care leakage?

Patient leakage is the broadest term and covers every patient who needed your care and did not get it from you, through any channel. Referral leakage is the subset that arrives with a physician's recommendation and never becomes a completed visit. Care leakage is care your practice already ordered, recommended, or scheduled that did not happen on time.

What causes patient leakage?

Leakage usually comes from small points of friction at the moment a patient is ready to act: no response outside office hours, long waits for a first appointment, scheduling that depends on a phone call, and referral handoffs nobody owns. Dormant patients leak because nobody has time to call them. It persists because most practices never measure it.

How do you measure patient leakage?

Start with three numbers: monthly website visitors, booked appointments that came from the website, and dormant patients in your EHR who are overdue for care. The first two give your conversion rate, and if it is near 1%, you are typical. From there, track each leak separately, including referral completion rate and recall rate, and review the numbers monthly.

How common is referral leakage?

In a 2018 study of 103,737 referral scheduling attempts in a large health system, only 34.8% resulted in a documented completed appointment. A 2011 review reported that 25 to 50 percent of referring physicians did not know whether their patients had actually seen the specialist.

Where to start

Run your numbers in the leakage calculator, read the questions practices ask before fixing it, or book a 15-minute demo to see the agents work on your own website.

Ready to increase your patient volume?