July 6, 2023 – Eric Collard has at all times been an athlete; he played college football, took up triathlons as an adult and now, at 44, commonly bikes, runs, lifts weights, plays golf and more. Collard, the Ottawa-based director of a nonprofit, also takes his food regimen seriously. By almost every measure – blood pressure, cholesterol, blood sugar – Collard is fit and healthy.
The only outlier amongst Collard's health indicators is his body mass index (BMI), which puts him within the obese category. “I'm a big guy,” he says of his 6'4″ height and 255-pound weight. “But I'm also healthy, and BMI shouldn't be the one measure of health.”
The body mass index has been around since the early 19th century, but it wasn't until the mid-1980s that doctors in the United States began using it to define obesity. It has stuck around for decades, even though many members of the medical community believe it is flawed – a patient like Collard shows why.
A new study from Rutgers University sheds new light on the accuracy of BMI as an indicator of increased risk of death. The results? If you're classified as obese based on BMI alone, the measurement is largely wrong.
“The literature on BMI is inconsistent, and so we wanted to handle the restrictions of previous studies,” said study co-author Dr. Aayush Visaria, a resident in internal medicine at Rutgers New Jersey Medical School. “The population composition within the United States has modified, so it was also time to duplicate the research on today's population.”
For their study, Visaria and his team analyzed data from more than 500,000 adults in the United States from the National Health Interview Survey from 1999 to 2018 and the US National Death Index from 2019. In these cases, BMI was calculated based on height and weight information. They included data on demographic characteristics, socio-behavioral factors, comorbidities, and access to health care. They divided the group into nine BMI categories.
Participants were on average 46 years old, 50% female, and 69% non-Hispanic. Of these individuals, 35% had a BMI between 25 and 30, which is considered overweight, and 27.2% had a BMI greater than or equal to 30, which classifies them as obese. Visaria and his team then followed the participants for a median of 9 years and a maximum of 20 years, and showed that 75,807 participants had died. When breaking down the data by ethnic category, the team found that overweight Hispanic participants were at increased risk of death, but not black or white participants. In general, however, the death rate increased once a BMI of 30 or more was reached.
The Rutgers group concluded that further studies are needed to better account for weight trajectories, body composition, and causes of death. They also concluded that BMI alone should not be the basis for clinical decisions.
Away from BMI
Visaria says that while BMI alone is not a good measure of health, he understands why it is a popular method in the medical community. “It's a straightforward calculation,” he said, “and it's replicable and scalable. For those reasons, it's still utilized in public health.”
Matthew Davis, MD, a bariatric surgeon at Methodist Medical Group in Memphis, agreed with that analysis. “Unfortunately, the identical aspects that make the system really easy to make use of – it simply takes under consideration an individual's height and weight, that are easily measurable – also make it open to criticism,” he said. “Because of its simplicity, it doesn't take note of the myriad other aspects that may contribute to obesity or poor health, namely social, environmental or genetic aspects.”
Recently, there has been a shift away from this long-standing health measure. In June, delegates at the American Medical Association's annual meeting adopted a new guideline intended to clarify how BMI should be used in medicine.
In a press release, the association stated: “The AMA recognizes problems with using BMI as a unit of measurement due to its history of harm, its use for racial exclusion, and since BMI is based on data collected from previous generations of non-Hispanic white populations.”
Because of those limitations, the association says, “it proposes using the value in conjunction with other 'valid' risk measures, including so-called visceral fat, i.e. fat hidden deep in the body, estimated body fat, waist circumference, and genetic/metabolic factors.”
Wendy Schofer, MD, founding father of Family in Focus, is among the many doctors who've long believed BMI is flawed. “Obesity, as defined by BMI, says nothing about the health of a person with a particular BMI,” she said. “We make a lot of assumptions based on these numbers.”
Instead, Schofer said, doctors need to higher understand their patients and their lifestyles. “We need to understand what is important to the individual, what health means to them, and how to develop and achieve the goals that are important to them,” she said. “I leave weight and BMI out of the conversation. Instead, we talk about how we want to feel, what we want to do, and how we want to be.”
Collard said his doctor also takes a more holistic approach to his visits, which he appreciates. “My doctor recognizes that my BMI is irrelevant to my overall health,” he said.
Visaria said his latest study is step one in a crucial, ongoing series on BMI. He recently shared the outcomes of a study on bone density measurements and BMI that showed the restrictions of each techniques in visualizing the connection between muscle mass and bone density. Future studies will include measurements of fat tissue, heart problems, hypertension and other metrics. “We want to look at the long-term effects on morbidity,” he said. “Providers need to consider a variety of other measures in addition to BMI.”