One measure of the impact of a scientific idea is how often it is cited by other scientists. The most cited newspaper of all time, according to a 2014 analysis by Naturehas been cited by 344,007 other scientific articles since its publication in 1951. (The topic? You’ll never guess, for reasons we’ll discuss below.) Researchers’ job prospects are influenced by their h indexa measure that rewards having a high number of highly cited articles (and perhaps, although no one would actually admit it, by their Kardashian Indexwhich compares their cumulative quotes to the number of Twitter followers they have).
You can also use similar techniques to parse entire fields, which a new study led by Omeet Khatra of the University of British Columbia, is attempting to develop sports and exercise medicine. In the Orthopedic Journal of Sports Medicine, Khatra and colleagues compiled a list of the 100 most cited papers in the field, offering insight into the influence of individual papers and broader trends. There are many interesting findings, but perhaps the most telling is this: Only one of the 100 papers is a randomized controlled trial, which is the gold standard for experimental evidence.
Field
An important caveat to this analysis is that the boundaries of sports and exercise medicine are quite blurry. Khatra’s definition includes managing sports injuries, improving sports performance, and using exercise to improve health. This is very broad, but the method used to identify the best articles was a little more idiosyncratic. They began by identifying a list of 46 journals specializing in sports and exercise medicine, then identified the 100 most cited articles in these journals.
This means that important articles published in non-specialist journals do not appear on the list. Original study by AV Hill from 1923 on VO2 max was published in the Quarterly Journal of Medicine; Karlman Wasserman’s 1964 article on anaerobic threshold was published in the American Journal of Cardiology. In fact, one might expect that the most groundbreaking discoveries are the most likely to be published in general journals like Nature And Science (where, for example, a classic paper from 1937 on the aerobic power of runners who set world records).
This is therefore not an exhaustive list, but it nevertheless covers a large part of the field. It is dominated by Medicine and Science in Sport and Exercise, the flagship journal of the American College of Sports Medicine, which contributes no fewer than 49 of the articles. Next on the list are American Journal of Sports Medicinewith 18, and Sports medicinewith 7. The oldest article on the list is from 1973, reflecting the relatively recent emergence of the field as a distinct discipline: MSSEfor example, was not launched until 1969.
The champion
Topping the list with 7,228 citations was Gunnar Borg’s 1982 article, “Psychophysical Basis of Perceived Exertion.” Borg was the one who put forward the concept of a subjective scale of perceived exertion, which originally ranged from six to 20, although there is a more logical modified version that ranges from zero to ten. He began developing this idea in the 1960s, but the 1982 English-language journal is the one that gets cited whenever people talk about perceived effort. (Another Borg article on the subject, from 1973, appears at number 48 on the list.)
You might not think that asking people to put a number on their level of work is a major scientific breakthrough. But Borg’s work was enormously influential. He argued that his scale is “the best indicator of the degree of physical exertion”, integrating signals from the muscles, lungs, heart and brain. Over the past two decades, more and more researchers have taken this argument seriously in attempting to explain the brain’s role in determining our physical limits, as well as as a practical tool for guiding training. In summary: I would say that Borg’s article is a worthy advocate.
The themes
The largest group of articles on the list focuses on methodological tools: how to perform a VO2 max test, how to calculate body composition, how to calibrate your pedometers and accelerometers, which validated questionnaires to use to ask your subjects about their exercise habits exercise, and soon. This is also what is observed in other fields: the most cited article of all time that I mentioned at the top is a methodological article on “measuring proteins with the folin phenol reagent”.
Methods articles may not seem very exciting, but they can certainly be controversial. Several of the articles on the list focus on statistics, including article number nine, from 2008, by Will Hopkins and colleagues: “Progressive Statistics for Studies in Sports Medicine and Exercise Science.” This approach to statistics is designed to reveal subtle effects on performance in studies with small samples. But it was the subject of strong criticism, particularly following a 2018 article in FiveThirtyEight by Christie Aschwanden, claiming that it is more likely to produce false positive results than traditional statistical methods.
Another big bucket is official guidelines, primarily those published by the American College of Sports Medicine on topics such as resistance training, exercise for cancer, hydration, weight loss, blood pressure and workouts for the elderly. These are useful insights to cite in the introduction to an article when you want to support general claims like “exercise is good for you” or the like, but they are not particularly groundbreaking.
After that, it’s more of a mixed bag. The most popular part of the anatomy is the knee, which is the subject of 15 articles, mostly related to ACL injuries. Next comes the brain, which features in three articles on concussions in sports. Two other themes which multiply the mentions: the persistent mystery of late onset muscle painand the emerging health scourge of excessive sitting.
There are three articles on football physiology, one on the biomechanics of baseball throwing, and one on Hakan Alfredson’s Famous Heel Drop Protocol for Achilles Tendinosis, which squeaks into 98th place. (Fun story: Alfredson is an orthopedic surgeon who had Achilles problems in the 1990s. When his boss refused to give him time off for surgery because his condition wasn’t serious enough, he decided to aggravated his Achilles with painful heel drops, but accidentally healed himself.)
The proof
I mentioned at the top that only one of the studies on the list is a randomized controlled trial, meaning that subjects were randomly assigned to receive either an intervention or a placebo. Instead, most experimental papers use lower levels of evidence, such as cohort studies And case series, none of which use randomization or control groups. The largest category, with 38 articles, is narrative reviews, which examine the results of multiple studies on a topic but do not combine them into a single large meta-analysis.
I think most sports scientists would agree that this area needs more randomized trials, as well as other methodological improvements such as larger subject groups and more sophisticated statistical analyses. But the flaws in the top 100 are probably not specific to sports science. Watson and Crick’s discovery of the structure of DNA and Einstein’s theory of general relativity are also not on their respective lists: the greatest advances become textbooks that don’t even require citation. “If you want quotes,” Peter Moore, chemist at Yale University said Nature“Designing a method that allows people to do the experiments they want, or more easily, will take you much further than, say, discovering the secret of the Universe.”
To learn more about Sweat Science, join me on Twitter And Facebookregister at E-mail, and look at my book Enduring: Mind, Body, and the Curiously Elastic Limits of Human Performance.
