There's a strange link between how well we can smell and our risk of death - but despite ongoing efforts, and a lot of overblown media coverage, researchers still can't figure it out.

There are few concepts we humans approach less rationally than death and statistics, and this new study is no exception.

Once again, researchers have demonstrated a higher risk of death among those with a poor sense of smell, and even though it's one of the first studies to supposedly examine why this is happening, the results have left us no closer to the truth.

Using data from two US communities, the researchers tracked 2,200 participants between the ages of 71 and 82 across thirteen years. Those who demonstrated a poor sense of smell - tested using 12 common scents - had nearly a 50 percent higher risk of death in 10 years time.

Sex, race and lifestyle showed little contribution to this number, while deaths from Parkinson's disease and dementia showed the strongest link.

"The association was largely limited to participants who reported good-to-excellent health at enrollment, suggesting that poor sense of smell is an early and sensitive sign for deteriorating health before it is clinically recognizable," senior author Honglei Chen of Michigan State University told Reuters.

No matter how many tabloids suggest otherwise, this does not mean that a bad sense of smell is fatal. Nor does it suggest that someday a smell test can be used to predict the onset of neurodegenerative disease.

"The UK's media coverage of the study was broadly accurate," wrote the UK National Health Service (NHS), "but The Guardian's claim that "Routine sense of smell tests could be used to spot signs of dementia" is not supported by the evidence presented in this study."

In fact, the agency argues that a "smell test for dementia" is "not on the horizon at all" and is simply "pure speculation".

While it's possible that Parkinson's or dementia may have some impact on the olfactory nerve, the authors could only explain 22 percent of their results using these neurodegenerative factors.

"We don't have a reason for more than 70 percent of the increased risk. We need to find out what happened to these individuals," says Chen, who aims to keep pursuing the mystery.

The current research, however, was ill equipped to answer that very question. Without examining premature deaths specifically, the NHS argues that a poor sense of smell may have had little impact on the overall risk of death.

What's more, it's entirely unclear whether the participants in this study had always possessed a poor sense of smell, or whether this trait had only come about more recently.

Observational studies like this one are undoubtedly useful, but it's beyond their scope to explain exactly why smell may be correlated with mortality. For now, the answer shall remain a mystery.

This study has been published in the Annals of Internal Medicine.