Your Phone Shouldn’t Tell You That You Have Cancer

Today, I told a young patient that we had found a tumour.

He was at least ten years younger than me. He had come to the emergency department with chest pain. Somewhere during the workup, his imaging revealed something neither of us had expected.

Before walking into his room, I looked at the report and felt that familiar dread.

Not because I didn’t know how to tell him.

I was worried that his phone already had.

Our hospital releases results to patients electronically, often before I’ve had the opportunity to explain them. So I walked into his room and asked him directly whether he had seen the report.

He had.

There are conversations in medicine that deserve to happen between two human beings.

This was one of them.

I sat down and explained what I knew. More importantly, I explained what I didn’t know. We had found a tumour, but there were still questions that couldn’t be answered that day. We didn’t yet know exactly what it was or what it would mean for him.

I told him that although this was bad news, knowing was better than not knowing. He had done exactly what he should have done by seeking care. Whatever this turned out to be, finding it meant we could begin figuring out what came next.

He held his emotions together remarkably well.

I could only imagine what had gone through his mind when he first read that report, alone with his phone.

This isn’t the first time I’ve found myself in this situation.

Patients now receive laboratory and imaging results almost instantaneously. There are obvious benefits to giving patients access to their medical information. I believe they should have that access.

But access to information and understanding information are not the same thing.

A laboratory value appears in red because it falls slightly outside a reference range, and suddenly I’m explaining to a frightened patient why an “abnormal” result is clinically meaningless. The computer has told them something is wrong before anyone has had the opportunity to explain that, medically, nothing is wrong at all.

More concerning are imaging reports.

Radiology reports need context. Preliminary interpretations can change. I’ve had to tell patients that something initially reported as normal was subsequently found to be abnormal. I’ve also had to walk back an alarming preliminary finding after the final interpretation determined that it wasn’t pathological.

Receiving devastating information is hard enough when it comes from a physician who is trained to break bad news.

Imagine receiving it as a notification on your phone.

There is another side to this problem.

Some patients grow tired of waiting in the emergency department, see their bloodwork appear on their phone, plug the results into ChatGPT, decide everything looks “normal” and leave before being assessed.

Sometimes they’re right.

But the bloodwork isn’t always the important part.

A set of laboratory results cannot identify a peritoneal abdomen. It cannot hear the hesitation in your voice when you answer a question. Normal bloodwork does not exclude unstable angina, aortic dissection or stroke.

The test result is one piece of the encounter. Increasingly, patients are treating it as though it is the encounter.

The same tension now exists with the medical record itself.

I’ve had patients ask me to change things I’ve written in their chart because they disagreed with my assessment or didn’t like how something was documented.

Sometimes they are absolutely right. Medical records contain errors, and patients should be able to identify and correct them.

But disagreement is not necessarily an error.

A medical record has traditionally served an important purpose: it is where physicians document what they saw, what they thought and why they made the decisions they made.

Now I write every note knowing that the patient may read it before the next physician does.

That changes things.

If I believe alcohol use is contributing to someone’s illness, should I hesitate to document it because they might find the characterization offensive?

And what about psychiatry?

Some of the most important parts of a psychiatric assessment are also the things a patient may find most upsetting to read: poor insight, paranoia, concerns about substance use, suicidal intent or questions about whether someone can safely care for themselves.

These observations can be uncomfortable. They can also be clinically important.

If physicians begin editing them not according to their clinical relevance, but according to how a patient might react when reading them, the medical record becomes something different.

It stops being an independent clinical document and starts becoming a document written partly for an audience.

That should concern us.

None of this is an argument for returning to an era when medical records were hidden from patients.

Patients deserve access to their health information. They deserve transparency. They deserve to know what physicians are thinking, and they deserve a mechanism to correct genuine errors.

But transparency should not require physicians to sanitize their clinical judgment.

Technology has made medical information available faster than at any point in history.

I’m not convinced we’ve spent enough time asking whether faster is always better.

Some results can wait twenty minutes.

Long enough for a physician to pull up a chair.