Dear FIFA: There Is No Scientific Test to Prevent Age Fraud

You could be forgiven for missing that the qualifying matches for a big youth soccer event—the Africa Cup of Nations—are underway right now, what with the Olympics taking center stage all week. But then you might miss the latest “doping” scandal playing out in soccer headlines: Almost half of Nigeria’s U-17 (under age 17) team has been sidelined based on lab scans.

Yet the dismissal is not over a traditional malfeasance such as performance-enhancing chemicals or hormonal supplements. It is about age fraud. Clinical testing suggested that these players were actually older than 17.

As a result, 24 young male athletes from Nigeria were deemed ineligible to participate, the Confederation of African Football (CAF) told Scientific American. (Originally, 26 players were excluded but two players were reinstated after appeal.) The decision to bar these Nigerian players from the tournament was based on a procedure that FIFA, soccer’s international governing body, put in place in 2009. It urged youth players to submit to magnetic resonance imaging (MRI) scans of their wrists in an attempt to determine their age eligibility. (CAF, which is under the auspices of FIFA, requires all players to submit to the scans as part of the regulations for the U-17 Africa Cup of Nations.)

Admirably, the protocol is designed to prevent older, stronger players from entering youth tournaments in settings where birth certificates or other ways of verifying age may be absent. According to FIFA, wrist MRIs and the bone growth they show can indicate if a player is older than 17. Yet a deeper dive into the study behind this decision reveals that its foundation is shaky at best.

FIFA rolled out this protocol based on a study conducted by its Medical Assessment and Research Center and published in the British Journal of Sports Medicine. For that work, researchers performed wrist MRIs (deemed more ethical than exposing the subjects to x-ray radiation) on about 500 young men between the ages of 14 and 19. The teens all hailed from Switzerland, Malaysia, Algeria or Argentina. The researchers then categorized each participant’s wrist bone development into one of six stages based on the fusion of his distal radius, a bone area near the wrist joint. Stage 6 represented complete fusion and stage 1 represented no fusion. Citing this work, FIFA says that if the MRI shows complete fusion of a player’s wrist, it is 99 percent certain that the player is over 17 (spoiler alert: that’s not really true).

Wrist bone development certainly does change during adolescence. Before puberty the human body cranks out cells that help produce new cartilage along the ends of bones. These areas of the bone, known as growth plates, are eventually converted into bone material, which helps the bones lengthen. Yet as teenagers, both males and females start producing more estrogen, which slows that cellular production process—and their growth plate cartilage subsequently gets thinner and thinner until some of the bones in that area fuse together, and the growth plates close and are replaced by solid bone. (That’s why wrist scans are sometimes used alongside those of the fingers, in tandem with other biological information, to estimate a growing child’s eventual height.) The exact timing of this fusion, however, varies a lot between individuals. For example, FIFA’s own researchers said in an analysis published earlier this year that using this same age cutoff for female athletes is extremely problematic because many girls already had complete fusion before age 17.

In FIFA’s citing of the foundational MRI study for boys, however, its major mistake lies in applying these population-level statistics to individual athletes. A closer examination of the full data shows that the actual wrist bone growth stages can occur at a wide range of ages. Among 16-year-olds, for example, roughly the same number of kids were at stages 1, 3, 4 and 5, representing a spectrum of growth at that age. “There is so much overlap that you may have a person who is 18 who could have only minimal fusion to complete fusion, and there are people who are already fused going from ages 20 to 16,” says Vicente Gilsanz, a professor of radiology and pediatrics at the University of Southern California. Although only one player in the 16-year age group was graded as completely fused (stage 6), the standard deviation in that grouping is also pretty telling.

The reason standard deviation is so illuminating comes down to basic statistics. The average age for complete fusion, according to the analysis, is 18.3 years. Yet there might be some variation in that fusion timing, and that possibility is represented by standard deviation calculated in the analysis—0.9 years. Anyone setting such important guidelines should typically include two standard deviations in either direction from the average age to capture most of that variation, says Frank Rauch, a professor of pediatrics who specializes in bone health at McGill University. (In a perfect bell curve one standard deviation will only capture about 68 percent of the values whereas two will include about 95 percent, and so on.) Here, two standard deviations from that average age of complete fusion would include ages 16 to 20. “Maybe people are not concerned about unfairly excluding kids from competition,” Rauch says, “but that’s the inevitable problem.” FIFA had not responded to a request for comment by publication time.

Meanwhile, subsequent work also throws more cold water on the science of these wrist scans: One study on 86 young male Ghanaian players concluded, “There was no significant correlation between the chronological age and the degree of fusion.” Another analysis that similarly employed wrist scans among young players found that three supposed 14-year-old male soccer players had stage 5 or stage 6 fusion. In that analysis, too, the authors similarly noted, “no correlation was observed between age category and grade of fusion.” But in both studies the authors dismissed the results—suggesting perhaps players were simply not aware of their real ages.

There are also genetic and environmental factors to consider before applying this study’s findings to the real world. The analysis underpinning FIFA’s age regulation does include some individuals from Algeria(which is in north Africa), yet it does not include anyone from sub-Saharan Africa, where this ruling has been applied before—and is again now. Very little puberty research is focused on the sub-Saharan region, so it is difficult to pinpoint information about the onset and tempo of puberty in Nigeria. But scientists have uncovered variation elsewhere. Multiple studies have shown that kids embark on puberty at different times based on factors including nutrition, environment and ethnic background. Puberty in African-American girls who grow up in the U.S. begins roughly a year earlier, on average, than among white girls, for example.

Ultimately, there is no known foolproof, scientific test that will allow doctors—or sports regulators—to determine an individual’s age. The science suggests that applying a lone wrist MRI test to make such determinations is inappropriate at best and potentially harmful at worst. Political leaders and regulators searching for age tests in other settings—such as classifying immigrants seeking asylum (because different rules apply for minors and adults)—could potentially look to this type of age cutoff test, too, with troubling results. Right now “people aren’t trying to use magnetic resonance imaging [to classify the age of refugees] but they are using x-ray imaging to look at bone age and make a determination as to whether they are children or not,” says Babette Zemel, a pediatrics professor focused on child growth and development at The Children’s Hospital of Philadelphia. “Looking at the bones can give you a good idea if one child is skeletally more mature, but it is insufficient for determining whether or not a child should be permitted to be on a sports team or claim asylum status because it doesn’t tell you about their chronological age.”



comments powered by Disqus