"I go to the gym, so I'm covered"

Some readers will have reached this point feeling that the book isn't about them. You swim twice a week, or you go to a class, or you do a parkrun on Saturdays. You've done what the health advice asks. The tally of thirteen hours is just the background to a life that includes proper exercise.

You're better off than someone who does nothing, by a wide margin, and I don't want to blur that. But the two things turn out to be more separate than they look, and the evidence on how they interact is worth knowing.

In 2016 Ulf Ekelund and colleagues pooled data from studies that had followed more than a million men and women for up to eighteen years. For each person they had two pieces of information, which were how long they sat each day and how physically active they were. That let the researchers ask the question directly. Does activity cancel out sitting?

The answer was yes, at a price. Among people in the most active quarter, sitting for more than eight hours a day carried no detectable extra risk of dying during the follow-up. That most active quarter, though, was doing about sixty to seventy-five minutes of moderate activity every day. That's an hour or more of something like brisk walking, daily. It's four or five times the 150 minutes a week that the guidelines ask for, and far more than two classes and a swim.

Below that level, sitting time mattered. Among the less active groups, the more people sat, the higher their death rate over the follow-up. The worst placed were the people who were both the least active and sat for more than eight hours. Their death rate was 59 per cent higher than that of people who sat little and moved a lot.

I should be careful about what kind of evidence this is. These are observational studies. People reported their own sitting and activity, which they do inaccurately. People who sit a lot may differ from people who don't in ways the researchers couldn't fully adjust for, and some people sit a lot because they're already unwell. Nobody can run the ideal experiment, in which thousands of people are made to sit for twenty years. So treat the figures as the best available estimate.

Two further results from the same study are worth having. The first is about which matters more. People who sat for less than four hours a day but were in the least active group had a death rate 27 per cent higher than the most active group. People who sat more than eight hours but were in the most active group had almost no extra risk. So if you had to choose between the two, moving a lot beats sitting little. Your swimming counts for a good deal.

The second concerns the sofa. When the researchers looked at hours of television in place of total sitting, the picture was worse. Three or more hours a day in front of the television went with a higher death rate in every group except the most active, and even in that group the rate rose at five hours or more. Chapter 11 looks at why evenings might be different.

The NHS draws the practical conclusion on its page about sitting. It says that you can hit your weekly activity target and still be at risk of ill health if you spend the rest of the time sitting or lying down. Its guidelines for adults ask for two separate things. One is 150 minutes of moderate activity a week, and the other is to reduce the time you spend sitting and break up long periods of not moving.

So a Tuesday with an evening class in it still has a morning, an afternoon and a long sit from two until half past three. The class doesn't reach back into those hours. If you exercise, keep exercising, because on this evidence it's the most protective thing you can do. But the noise at twenty to seven on a Wednesday is a result of what you did between five and twenty to seven, and Tuesday's class doesn't change that.