Comfort Zone

Chapter 40

Miriam felt virtuous when she made it to the weekly medical conference the next morning. She knew doctors who went every single week and longed to be one of the good ones. "It's the only way I can catch up on the latest advances," one said, and Miriam agreed. So why did catching up on sleep usually win out?

It was Ursula Taylor who normally introduced the speaker, and her stand-in reassured the audience that she was doing well. Miriam thought of the biopsy, the result of which should be ready later that day. Lung cancer or not, she did have a broken wrist from falling out of bed, joining the nearly one million patients who fall in US hospitals yearly.

And she had tried to press Taylor about the narcotic diversion. "The investigation is ongoing," Taylor had said, a tad too forcefully for comfort, furthermore asking Miriam about the filgrastim she'd taken when the tech was murdered.

I won't turn you in, Josh had kidded her. Miriam thought of how fast the rumors of JK's possible involvement had made their way to—everyone—and wondered if she'd have the same fate. At least I'm here to defend myself.

She looked around the room furtively, checking to see if anyone was shooting her curious looks. Why did she feel guilty?

She couldn't help reliving the encounter and thinking about what she should have said to Taylor. I was rattled, I'd just seen a murder! My patient needed help; I would never have taken medicine under normal circumstances. And anyway, if I was taking medicine and altering it, why would I want to draw attention to myself?

Her heart fell when she thought of a likely answer: to draw attention from the real thefts.

"Now that the exit doors are barred," the speaker began, "I can tell you that this talk is the first of a series on medical statistics. I'm calling this one 'The joy of the positive predictive value.' Positive predictive value, or PPV, for those of us on first name basis.

"You may not believe it's a joy. You're trying to think of an excuse to run out, wondering if the doors are really barred shut. Hang in there. Most non-medical people are terrified and confused by numbers and statistical terms, and studies show the same holds true for doctors. But you all know it's increasingly important in medicine these days.

"Everyone's buzzing about evidence-based medicine, and most of you know about the medical initiative Choosing Wisely. Its goal is to educate all of us to be smarter about ordering tests and treatments, taking benefits, risks and even costs into account. Contrary to popular belief in our techie world, more technology doesn't always lead to better care.

"We've committed ourselves to basing our workup and treatment on proof, and not simply blindly accepting what's conventionally been done. Get comfortable with statistical terms and you'll be more comfortable sifting through studies, understanding recommendations, and educating your patient.

"In deciding whether or not to order the mammogram, the prostate blood test, or the electrocardiogram, the question that should go through your mind is:

"If this test is positive, is it more likely to lead to early diagnosis, successful treatment, and a patient who tells everyone you saved his life, OR—is it more likely to be a false positive, leading the patient into more testing and possible harm, not to mention unnecessary worry and a demand to transfer his care to your bitterest rival?

"In order to answer this, you need to know the probability that the patient with a positive screening test truly has the disease. This is the positive predictive value, the PPV. If it's high, then a positive result is more likely to be a true positive. If low, that positive test is probably wrong.

"So, come out from hiding under your chair. My goal today is to make the phrase 'positive predictive value' your new best friend.

"The PPV merges the error that's built into every test with the factors that are specific to the person in front of you. You already know that error is unavoidable, no test is perfect. We call this error the false positives and false negatives of the exam. False positives are false alarms; the test is positive but the patient doesn't have the condition. False negatives, in which the test is negative but the patient does have the condition, for example a mammogram that appears normal but actually contains a cancer, are sometimes called...lawsuits." The speaker went on after the laughter died down.