Millions of Americans who have been labelled overweight or obese based on their body mass index (BMI) are in actual fact perfectly healthy, according to a new study.

Scientists in California say that 34.4 million Americans considered technically overweight due to their BMI are actually healthy based on a range of cardiometabolic health markers, as are some 19.8 million 'obese' people. The massive misclassification isn't just about which words we use, either, say the researchers, since the flawed BMI's usage in the health insurance industry unfairly penalises some, while rewarding others.

"In the overweight BMI category, 47 percent are perfectly healthy," said researcher Jeffrey Hunger from the University of California, Santa Barbara. "So to be using BMI as a health proxy – particularly for everyone within that category – is simply incorrect. Our study should be the final nail in the coffin for BMI."

The researchers looked at data from the most recent US National Health and Nutrition Examination Survey to analyse the link between BMI – a measure calculated by dividing a person's weight in kilograms by the square of their height in metres – with a range of specific health markers. These cardiometabolic assessments included blood pressure, blood sugar, and cholesterol, among others.

What they found was that BMI incorrectly pegs people's health at both ends of the weight scale.

"Not only does BMI mislabel 54 million heavier individuals as unhealthy, it actually overlooks a large group of individuals considered to have a 'healthy' BMI who are actually unhealthy when you look at underlying clinical indicators," said Hunger. "We used a fairly strict definition of health. You had to be at clinically healthy levels on four out of the five health indicators assessed."

A good example of how BMI can offer skewed results is when it is used to calculate the health of athletic individuals. While these people can be incredibly fit, under BMI their heavier muscle tone can mistakenly lead to them being classified as overweight or obese.

The discrepancy is at its most apparent when looking at the BMI of some famous (and clearly not unhealthy) athletes. For example, according to an analysis early last year, NFL player Tom Brady is obese, and Olympic athletes Usain Bolt and Michael Phelps just narrowly miss the 'overweight' camp.

The findings, reported in the International Journal of Obesity, are important, because while BMI has long been noted for its limitations and flaws, the measure is nonetheless in wide use by US companies to determine their employee's health insurance costs.

Also, a new rule proposed by the Equal Employment Opportunity Commission (EEOC) could penalise people with BMIs higher than 25 – the upper limit of the 'healthy' range – by making them pay higher premiums.

"We need to move away from trying to find a single metric on which to penalise or incentivise people and instead focus on finding effective ways to improve behaviours known to have positive outcomes over time," said Hunger.

But if BMI is such an inaccurate way of gauging people's health, why does it remain in such common usage? The answer, according to the researchers, lies in its simplicity.

"The reason I think people rely on BMI is because it's easy; if you know someone's weight and you know someone's height, then out pops this magical number," A. Janet Tomiyama, one of the researchers, told Amina Khan at the Los Angeles Times.

"But getting blood pressure is pretty easy too. It takes maybe 20 seconds if you have the machine. And so I really think focusing on better health markers like blood pressure is a better way to go about it – particularly when we're talking about financial penalties."