I was 7 years old when my doctor told my parents that watching and waiting was no longer an option.

I’d been diagnosed in the first year of my life with an atrial-septal defect, a hole in the heart that sends blood flowing the wrong way, forcing the right side of the heart to work harder than it should. In some cases the hole closes on its own during early childhood, but mine hadn’t shown any change, and now my heart was beginning to grow unevenly. Without surgery, I would face an adulthood characterized by fatigue, shortness of breath, and possibly heart failure. To prevent these things, a surgeon at the Children’s Hospital of Philadelphia would have to slice open my chest, crack my sternum, and stitch the hole in my heart together.

But there was another option, the doctor explained: He’d heard of a clinical trial that was recruiting pediatric patients with my condition. If I were placed in the experimental group, a cardiologist would insert a catheter into my upper thigh and direct it toward the hole in my heart. The catheter would deliver a tiny, metal mesh umbrella, which would cover the hole in my heart until my cells grew over it, making the umbrella a permanent part of my body. I would be in the hospital for just a weekend, with no broken ribs, no cardio-bypass machine, and no huge scar on my chest.

After my mother, who designs preclinical studies for a living, did the research and found that preclinical research in animals looked promising, she took my doctor up on his offer to refer us. I was assigned to the group of patients that would receive the device (others were randomized to the standard surgery group, although my cardiologist noted that many of those assigned to surgery might choose to drop out of the trial).

As promised, the catheterization and device placement went smoothly, and I was out of the hospital in just a couple of days. In the months that followed, regular checkups and ultrasounds showed that the device was still working fine—it hadn’t moved out of place, and no pieces of metal had broken off and entered my bloodstream, both things that had been concerns in the animal trials.

After four years, my mother scheduled another checkup—but this time, doctors told her that the trial was over, and that I wouldn’t be having any more follow-up appointments. I haven’t had an ultrasound since, and to this day, I’ve never seen the results of the trial. The pediatrician who referred me retired shortly after I had my procedure, and my attempts to reach the doctor who ran the study have been unsuccessful, as have my requests to talk to someone on the hospital's media team. It’s more than a little unsettling to think that I’ll live for another 70 years or so with this device in my heart, but may never know what the research said about how it works.

* * *

It often takes years for a drug or device to move from clinical trials to Food and Drug Administration approval. Even after the last patient has completed her last follow-up appointment, it still may take a couple more years to analyze the data and publish the results. If the data aren’t impressive, the study may never be published at all, meaning the public—and, more importantly, the trial participants—may never know the results.

In 1997, the same year that I had my procedure, Congress passed the Food and Drug Administration Modernization Act, which authorized the U.S. Department of Health and Human Services to develop a registry of clinical trials. The act, part of a growing movement to improve clinical-research transparency, required investigators and trial sponsors to register their trials before they began recruiting patients. Three years later, the National Institutes of Health made the registry, which lived at clinicaltrials.gov, available to the public—but while anyone could go online and look up the purpose of an investigational drug, or the locations where the trial was taking place, the site didn’t require the people registering the trials to include any information about the results.

In 2007, Congress passed the Food and Drug Administration Amendments Act (FDAAA). Among other things, the new law required investigators to publish summaries of their trial results on clinicaltrials.gov; the results database became public the following September. The law only applied to trials completed before December 2007, though, meaning my 1997 study was exempt.

In the years since the FDAAA was enacted, several respected medical journals have stated that they would not publish studies unless they had been registered. Thanks to these regulations, reporting of trial results on the registry has been climbing steadily.

Even so, the world of clinical trials is far from perfectly transparent. “It’s a changed environment,” said John Ioannidis, a professor of medicine and health policy at the Stanford University School of Medicine, “but some of the key problems are still there.” An analysis published last March in the New England Journal of Medicine found that the results of about half the registered trials still go unreported.

Even when a study is registered or published, there may be information that didn’t make it into the database or the final version of the paper. The only way to know every measurement that a researcher collected during a study is to have a copy of the full protocol, which can be difficult to come by: A recent paper in PLOS Biology examining 441 studies from the past 14 years found that only one had published its full protocol.

“You’re not guaranteed that these are the complete results,” said Ioannidis, who coauthored the PLOS study. There may be information from a trial that “has not been reported at all, but also there is a lot of information that has been misreported or selectively reported.”

And that’s just for successful clinical trials—if a drug or device fails during the research stage before it’s approved by the FDA, the results don’t need to be reported. It’s unlikely that these failed trials will be published in a medical journal, or even that the results will appear anywhere at all, meaning the reason for the failure may never become publicly available. “Unfortunately, even if you are a participant in a trial, the trial may still disappear and you may never find out about the results,” Ioannidis said.

Recently, the NIH and the Department of Health and Human Services have taken steps to prevent these studies from disappearing altogether. In 2014, the two agencies released a Notice of Proposed Rulemaking suggesting that results from all trials, even those that didn’t progress to FDA approval, should be reported on clinicaltrials.gov. The proposal hasn’t yet been enacted, but even if it becomes a rule, there are still loopholes: If a drug or medical-device company goes out of business, for example, what happens to their still-unpublished studies? Who, in those cases, will be responsible for making the results publicly accessible?

According to a 2013 review article in the Journal of Invasive Cardiology, at least nine different devices have been developed thus far to treat atrial-septal defects via catheterization. As of now, only two of these have been approved by the FDA. In the years after we learned that my trial had ended, my mother scoured medical journals searching for the results of my trial, but her search failed to turn up any concrete answers. We do know that two devices—one that was later approved, and one that was abandoned—may have been in trials at the time that I had my catheterization. Perhaps I’m living with one of these two. Or perhaps the data from my trial was never published.