A child, perfectly healthy, suddenly dies. It’s a heart condition, the doctor says, possibly genetic. The autopsy was inconclusive. It’s hard to know for sure. It just happens sometimes. It happens 11,000 times a year in the U.S. to people under the age of 45. Doctors call it “sudden unexpected death,” but that doesn’t give a grieving family any answers.

But that might be changing, thanks to the power of genetic tests. Mayo Clinic doctor Michael Ackerman pioneered the so-called molecular autopsy in 1999, using a DNA test to explain the sudden death of a 19-year-old woman with a previously-undiagnosed inherited heart condition. Since then, sequencing DNA has become orders of magnitude cheaper and more sophisticated. With medical examiners considering DNA tests as part of autopsy reports, the molecular autopsy has raised new ethical concerns.

Family members, in looking for an explanation for their grief, might end up finding unsettling things about themselves. What medical advice do you give someone carrying the same ticking genetic time bomb that killed their loved one? And who is responsible for telling the family?

“Traditionally with medical examiners, there is no patient-physician relationship. They have very little interaction with the family” says Quianta Moore, a health-policy scholar at Rice University. Molecular autopsies completely flip that paradigm. Moore studied ethical issues around molecular autopsies at the behest of Harris County’s medical examiners, who wanted to DNA test dozens of old cases where the case of death was unknown.

Complicating it all is the fact that molecular autopsies—and DNA tests in general—too often come up inconclusive. Molecular autopsies in 25 cases in the San Diego medical examiner's office provided a likely cause in just 16 percent cases and a plausible cause in 24 percent more according to a study published today in the medical journal JAMA. Other studies in Texas and Minnesota found similar rates of solving sudden deaths. “It’s certainly not where we would like it to be,” says Ali Torkamani, a geneticist at the Scripps Research Institute.

The uncertainty has, in a way, gotten worse as DNA sequencing technology has gotten more comprehensive. Early on, Ackerman and his colleagues tested just four genes, each closely linked to inherited heart arrhythmia. Now, scientists have the ability to sequence every single human gene in one go, all 20,000 of them. Unfortunately, they don’t know enough about all 20,000 genes to say if a specific mutation actually causes a deadly heart problem. A mutation in one gene might kill someone; a different mutation in the same gene might have no effect. Sometimes, even the same mutation in two people don’t behave the same way.

“There’s one thing worse than telling these families we don’t have an answer,” says Ackerman, “and that one thing is telling them we have found the genetic cause when we in fact we have not.” Ackerman has seen numerous surviving relatives who after learning they carried a potentially deadly mutation, treated it aggressively—even implanting defibrillators—only to find later that the genetic basis was weak. “That’s a mess,” he says.

At the same time, families are often eager for any answers, even if it might be inconclusive. One mother, who lost her son, told the San Diego Union Tribune that getting into Scripps’s molecular autopsy study was “like winning the lottery.” The National Association of Medical Examiners has recognized the potential value of molecular autopsies too. It recommends keeping tissue samples after an autopsy and keeping them without formalin, a traditional preservative that can damage DNA.

The uncertainty of DNA tests is not unique to molecular autopsies, of course. Genetic tests for breast cancer, for example, can also results where a mutation is simply of “unknown significance.” As genetic tests become cheaper and more popular, interpreting the medical information they contain will get even more vital. Molecular autopsies just underscores the challenge, which does not end even with death.