You Can Already Tell Who Didn't Come Back. The Question Worth Asking Is Why Not.

We're all working to get better at spotting the patients who slip through, and the field has come a long way. What almost none of us captures is the one thing that tells us what to do about each of them: the reason she didn't return. Knowing who slipped through is the part we've gotten good at. Knowing why is the part that brings her back.

We're all working on the same thing in breast imaging follow-up: getting better at the what and the who. What's supposed to happen next for a patient. Who's overdue. With good tracking, even who's likely to miss before she does. The field has come a long way on that, and the progress is real.

But look at what all of it has in common. It's facts. Statuses. The recall happened, the visit didn't, the clock ran out. And underneath every one of those facts sits a question none of us has really cracked. When she doesn't come back: why not?

That's the question worth answering, and it's the one we've built almost nothing to capture.

---

The Reason Is Invisible by Design

Think about why the gap was so hard to see in the first place. A record made of events that happened can't show an event that didn't. The better tracking systems now fix that much, and surface the woman who's gone overdue. But go one layer deeper than that. Even once you can see she didn't come back, the record still can't tell you why — because a reason isn't an event either. There's no billing code for "the copay scared her." No field for "she's still rattled from last year's biopsy." No checkbox for "the letter went to the address she left in 2023." The reason lives in her head, and sometimes in the memory of the navigator who reached her, and nowhere anyone can actually use. We can show the gap. We have no way to show the cause of the gap.

One "Overdue" Hides a Dozen Reasons

This is the same trap as the single compliance number, moved down a level. A blended rate blurs five clinical tiers into one figure that describes none of them. An "overdue" flag does the same thing to causes. It blurs a dozen different reasons into one status that tells us nothing about what to do.

A woman who didn't return because she couldn't afford the copay, a woman avoiding it because the last workup terrified her, a woman whose reminder went to a dead phone number, a woman who heard "probably benign" as "you're fine," and a woman buried under a family crisis are not five versions of one problem. They need five different things. Cost needs a financial conversation. Fear needs a warm, human reassurance call. A dead number needs the record fixed. A misunderstanding needs a clearer word from someone she trusts. Sending all five the same automated reminder is the why-not version of the single number: it treats clinically and humanly distinct situations as if they were one, and it quietly fails most of them.

The Reason Is the Only Thing We Can Actually Act On

Here's why this matters more than the status does. We can't un-miss the visit. It already didn't happen. The only thing we can change is what we do next, and what we should do next is set entirely by the why. The status tells us to act. Only the reason tells us how. That makes the reason the highest-leverage fact in the whole pipeline, and it's the one we don't collect.

The Part That Stings: We Already Learn the Reasons, and Then We Throw Them Away

The reasons aren't unknowable. That's what's maddening. The navigator learns them every single day. Every time she finally gets an overdue woman on the phone, she hears the why. "I couldn't take the time off." "I was scared." "I never got the letter." "I didn't think I had to come back." And then it evaporates, because there's nowhere for it to go. No structured field, no tally, no place it accumulates. So a program rediscovers the same handful of reasons one phone call at a time, and never once learns from them at scale.

We even have proof of how powerful a single captured reason can be, because researchers bothered to measure one of them. A clean mammogram is followed by a return to screening about 77% of the time. After a false positive that sent a woman to short-interval follow-up, that drops to 61%, and the effect still shows years later (Miglioretti et al., Annals of Internal Medicine, 2024). That's a why-not, named and quantified: a prior scare keeps women away. It took a national study to surface one reason. Imagine knowing the rest — for our own patients, by name, the week they go overdue instead of a decade later in a journal.

Capture the Why and Two Things Change at Once

For the individual patient, we stop guessing. We route by reason instead of sending one generic nudge to everyone: the cost case to financial counseling, the frightened case to a real person, the bad-contact case to a record fix. The call lands on what's actually in the way.

For the program, something bigger happens. For the first time, we can see the reason distribution, and that's the only thing that tells us whether to fix the patient or fix the system. If a large share of surveillance no-shows trace back to cost, that is not a navigation problem, and no volume of reminder calls will ever touch it. It's a financial-counseling problem wearing a follow-up costume. We'd never know, because today the reasons die one call at a time. Counted, they'd point straight at the fix.

We've Been Answering the Question the Data Gives Us, Not the One That Matters

We've poured our effort into the question the data hands us — which patients didn't come back — and skipped the one it won't give up so easily, which is why she didn't, even though the why is what every next move depends on. So the single most useful fact about a lost patient, the reason we lost her, walks out the door with her, and we go find it again, by hand, on the next one.

We know what we know. What we'd give to know why is the thing we've never bothered to write down. The day a program starts writing it down, it stops guessing why it's losing her and starts knowing — while there's still time to do something about it.

---

Sources

Miglioretti DL, Abraham L, Sprague BL, et al. Association between false-positive results and return to screening mammography in the Breast Cancer Surveillance Consortium cohort. Ann Intern Med. 2024;177(10):1297–1307. doi:10.7326/M24-0123. (Return to screening ≈77% after a true-negative result; ≈61% after a false positive recommending short-interval follow-up; effect persists for years.)

Freeman HP, Rodriguez RL. History and principles of patient navigation. Cancer. 2011;117(15 Suppl):3539–3542. (Foundational framework for the principle that navigation must be tailored to the specific barrier, not applied generically.)

Ferrante JM, Chen PH, Kim S. The effect of patient navigation on time to diagnosis, anxiety, and satisfaction in urban minority women with abnormal mammograms. J Urban Health. 2008;85(1):114–124. (Navigation interventions that identify and address individual-level barriers produce measurably better follow-up outcomes than standardized reminder protocols.)

---

About the Author

Richard D. Lippert Jr. is the President & Founder of Mammologix and a registered radiologic technologist with more than 30 years of experience in breast imaging operations. His work spans mammography patient tracking, follow-up navigation, lay communication, FDA/MQSA-related support, medical outcome audit, and the operational systems that help facilities ensure no patient slips through the cracks. He founded Mammologix in 1995.

View full bio →