In a surveillance study published in 2007 in Medicine & Science in Sports & Exercise, Spence and colleagues followed 32 elite and 31 recreational triathletes and cyclists alongside 20 sedentary controls for five months. Anyone who reported two or more respiratory symptoms was swabbed, and the samples went through microscopy, culture and PCR testing. The group logged 37 episodes of upper respiratory illness, and an infectious agent turned up in only 11 of them: six in elite athletes, three in recreational athletes and two in controls. Rhinovirus, the common-cold virus, was the most frequent find. The elite athletes still fell ill at 4.5 times the rate of the recreational group (incidence rate ratio 4.50, 95% confidence interval 1.91 to 10.59), so the higher rate of symptoms was real, whatever was causing it. As cold season gathers pace, that gap between feeling ill and being infected is a useful place to start.
Where the "open window" idea came from
The "open window" hypothesis says the hours after hard exercise leave the immune system briefly compromised, and that opportunistic infections walk through.
Neil Walsh's 2018 mini-review in the European Journal of Sport Science notes that innate and acquired immunity are often reported to fall transiently after heavy exertion, typically by 15–70%, while adding that whether this explains more respiratory symptoms "remains contentious". The classic exhibit is a survey of the 1987 Los Angeles Marathon: of 2,311 respondents who finished and had been well the week before, 12.9% reported an infection in the week after, against 2.2% of entrants who withdrew for reasons other than illness. John Campbell and James Turner of the University of Bath reread that literature in Frontiers in Immunology in 2018 and objected on several grounds. None of the self-reported infections in those marathon studies was laboratory-confirmed, they argued, the post-race drop in circulating lymphocytes can be read as cells relocating to tissues on patrol rather than as suppression, and a mass-participation race is also a crowd. In a 2020 debate paper in Exercise Immunology Review, eight researchers arguing opposite sides, Campbell, Turner and Walsh among them, agreed that infection susceptibility has more than one cause. They left open whether exercise by itself is one of those causes, and they noted that infection burden is reported to be second only to injury as a source of lost training days in the build-up to major events.
The risk factors nobody disputes
Walsh's review lists the factors that recent studies have tied to respiratory illness in athletes. A single hard session is not on it:
- intensified training in the winter
- long-haul travel
- low energy availability, meaning too little food for the work being done
- high psychological stress and anxiety, and depression
Campbell and Turner add that air travel was a significant predictor of illness symptoms in athletes. Olli Ruuskanen and colleagues in Finland, writing in the Journal of Sport and Health Science, report that during a two-week major winter sports competition 38% of athletes developed a common cold, against 6% of normally exercising controls. A major competition packages hard exercise together with the other exposures on that list, which is the confounding problem the debate paper describes. For a club runner or cyclist the practical reading is plain: the fortnight when a training-load jump, a flight and a stressful month at work all land together is the fortnight to protect.
Sleep, the best-documented lever
Two studies of adult volunteers, neither of them athletes, give sleep the clearest numbers in this field. In a 2009 paper in Archives of Internal Medicine, Sheldon Cohen's group at Carnegie Mellon found that people averaging under seven hours of sleep were 2.94 times more likely to develop a cold than those averaging eight or more (95% confidence interval 1.18 to 7.30). Aric Prather and colleagues tested it again in 2015, in the journal Sleep, this time measuring sleep with wrist actigraphy instead of questionnaires: 164 volunteers wore the device for a week, were then given rhinovirus nasal drops and were monitored for five days. Those who slept under five hours had 4.50 times the odds of a clinical cold compared with people sleeping more than seven hours, and those sleeping five to six hours had 4.24 times the odds. People in the 6.01-to-7-hour band were not at significantly higher risk, though the confidence intervals throughout are wide. Sleep fragmentation, notably, did not predict who got ill; duration did. The UK and EU clocks go back on 25 October, which hands you one extra hour on a single night and does nothing for a week of five-and-a-half-hour nights. Treat seven hours as the floor in any week when training load goes up.
Training through a cold
Sports medicine has long used the "neck check": symptoms above the neck, such as a runny nose or sneezing, allow easy exercise, while symptoms below it, such as chest congestion, body aches or fever, mean rest. The Mayo Clinic's version of the advice says mild to moderate activity is usually fine with a common cold and no fever. The evidence behind it is thinner than its popularity suggests. Ruuskanen's group states that return-to-sport guidelines after a viral respiratory illness rest on expert opinion and calls the neck check "nonscientific but may be partly useful".
The best direct test is old and small. In 1998, Weidner and colleagues in Medicine & Science in Sports & Exercise gave rhinovirus 16 to 50 moderately fit people aged 18 to 29. Thirty-four of them then exercised for 40 minutes every other day at 70% of heart-rate reserve for 10 days, and 16 did not. Symptom scores and the weight of used tissues did not differ between the groups. That result covers moderate sessions in young volunteers with an experimentally induced cold. It says nothing about a threshold session in an elite athlete, or about an illness that turns out not to be a cold.
Cardiac symptoms are the exception to everything above. Ruuskanen's paper treats chest pain, fainting, shortness of breath and palpitations as absolute reasons not to exercise, and lists suspected SARS-CoV-2, adenovirus, enterovirus or influenza as reasons for a break in training. Skip the session on any day with a fever or a symptom below the neck, and get a test if flu or COVID-19 is circulating where you train.
What the autumn evidence says not to do
Do not stop training to dodge colds. In a 2011 study in the British Journal of Sports Medicine, David Nieman's group followed 1,002 adults aged 18 to 85 through 12 weeks of autumn and winter. Those who reported aerobic exercise on five or more days a week had 43% fewer days with respiratory symptoms than those who exercised one day a week or less, and the top third for self-rated fitness had 46% fewer than the bottom third. The design is observational and the symptoms self-reported, so the numbers show an association, but they point away from the idea that regular training is what opens the window.
On supplements, the NHS website gives UK government advice that everyone should consider a daily vitamin D supplement of 10 micrograms (400 IU) during autumn and winter, when sunlight is too weak for the body to make it. The advice is framed around bones, teeth and muscles, not around colds, and it is guidance for the general population rather than a finding about athletes. Stay at that dose and leave megadoses alone; the same page sets upper limits.
Before the next block starts, open the diary and mark the fortnights where a flight, a race and a heavy week fall together. Those are the stretches where the factors on Walsh's list stack up.