In order to improve the quality of health care and reduce its costs, researchers need to know what works and what doesn't. One powerful way to do that is through a system of "registries," in which doctors and hospitals compile and share their results. But even in this era of big data, remarkably few medical registries exist.

Dr. Martin Makary is a surgeon at the Johns Hopkins School of Medicine. He and his colleagues published a study online this week in the Journal for Health Care Quality about the major shortcomings in the way patients are tracked after treatment.

I had a chance to speak with Makary about his interesting findings. The following is an excerpt of our conversation, edited for brevity and clarity:

Makary: For 99 percent of people in America, when they go in to have surgery, the outcome [of that operation] is not measured. Nobody's keeping track. So I'm amazed at how one-fifth of the economy, [the share accounted for by the healthcare system], functions with so little measure of its performance.

Harris: What inspired you to do this study?

Makary: I do an operation known as pancreas islet transplantation, and we send our outcomes to a national registry. I tried to get the data from that registry to better understand the disease and the outcomes from different centers, and I found out I couldn't get access to the data. I found it ironic that the data warehouse is funded by the taxpayers, and yet I can't even get access as a participating center. So it really got me curious about how many other diseases are being tracked in a national registry that is taxpayer-funded and that is also not transparent and available to the public.

Harris: What's the point of collecting the data if it's not made available to researchers who want to examine it?

Makary: The truth is, it's held in proprietary housing, and sharing it is not as easy as it sounds in principle. Instead [people who are trying to evaluate the medical system] focus too much on a few things that are very easy to measure, like patient satisfaction. Patient satisfaction may not even correlate with the appropriateness or the quality of the operation. I know people who have had unnecessary operations but are totally satisfied with them!

Harris: Turning to your paper now, what was your biggest surprise?

Makary: The biggest surprise was that [among the 153 registries we identified] there is tremendous variation among the different medical specialties. And the other big surprise was that only 18 percent had any form of auditing to verify their data. Now, on Wall Street we would never accept an 18 percent auditing rate when companies report their earnings. As a matter of fact, if a company misreports their earnings, the CEO goes to jail.

Instead of focusing on outcomes, we have a huge upfront effort to approve a new device or a new medication. We put them through the ringer. And once they're released, nobody's tracking anything. You have no idea how people are doing with a new medication or a new knee joint. New surgery is untracked for the most part.

Harris: There are financial incentives for companies not to know those answers. Hospitals, device makers, drug makers — if they're selling something already, the only news can be bad news, right?

Makary: Absolutely. Life is good right now if you get FDA approval. And hospitals don't want to spend the $60,000 to $80,000 a year to participate in a registry.

Harris: Hospitals could worry about losing patients. If you're looking at who's in the top 10 percent of all hospitals in terms of treatment outcomes, that's only 10 percent of the hospitals. And the other 90 percent might say this doesn't serve their interest in attracting patients.

Makary: You're exactly right. Hospitals are not malicious, but there's no incentive for them, from a business standpoint, to chase down their outcomes and measure their performance for internal quality improvement.

Harris: Can you give me an example of an instance where data gathered in a manner like this changed the way care was provided?

Makary: The best registry in America, in my opinion, is the cystic fibrosis registry. It has been around for 20 years. I believe that because of the cystic fibrosis registry we've been able to prolong the survival of kids with cystic fibrosis by almost two decades.

Also, Dr. Douglas Rex at Indiana University, who was frustrated at the problem that doctors were doing colonoscopies too quickly and inappropriately, decided to secretly record the videos of a whole bunch of doctors doing the procedure. He then told his colleagues, "I'm going to be watching your videos." Instantly, the quality scores improved by 40 percent.

If we're going to get serious about reining in healthcare inflation, we have to get serious about measuring our performance in health care. Everybody agrees about that. The actual work of measuring patient outcomes is what's missing.

Dr. Marty Makary is author of the book Unaccountable: What Hospitals Won't Tell You and How Transparency Can Revolutionize Health Care.

Copyright 2021 NPR. To see more, visit https://www.npr.org.