Australasian Science: Australia's authority on science since 1938

Being Physically Active Became Twice as Hard

By Tim Olds

How much exercise do we need to remain healthy? A group of experts has now upped the ante dramatically.

The last Australian Physical Activity Guidelines were in 2004, when it was recommended that adults get 30 minutes of exercise five times per week. In February the new guidelines were published, with the Australian government now recommending that we get 60 minutes of exercise five times per week. We also need two muscle strengthening sessions, and we need to reduce our sitting.

A 100% increase seems a bit beyond the normal rate of inflation, especially given that about half of all Australian adults didn’t get even 30 minutes per day. So how do experts decide on these guidelines?

One thing they normally do, or should do, is to look at dose–response curves. The idea of a dose–response curve started with drug studies: it showed the relationship between a drug dosage and the effect it had on some aspect of health (e.g. X dosage of statins reduces cholesterol by Y). But it doesn’t have to be a drug: the “dosage” could be a certain number of cigarettes, a given intake of dairy, sun exposure, years of schooling, or an amount of physical activity.

The accompanying figure shows a typical dose–response curve relating all causes of mortality to hours of physical activity. You can see that the more physical activity you do, the lower your risk of dying in any given year (in the long run we all have a 100% risk of dying).

You can also see that the biggest life bang for your exercise buck comes from moving from doing nothing at all to doing a little. Investments after that have a lower rate of return. So if you move from zero physical activity to 2.5 hours per week (the old guidelines) you reduce your risk of dying by about 13%. If you move from zero to 5 hours per week (the new guidelines) you get a 20% reduction. Another 5 hours per week only buys you a further 10% reduction, so you’d have to ask whether it’s really worth it.

Most dose–response curves are like this – smooth and continuous, with decreasing marginal returns. You rarely see curves with sudden bends and turns, with discontinuities in the first derivative (as the mathematicians like to say). Why not?

It may be that most outcomes are the result of many, many inputs that can either happen or not (e.g. a large number of genes that can be either switched on or not). When this happens, distributions tend to be smoothed out – the normal distribution is a limiting case of the binomial distribution.

So how do we know where to draw the line? Well, we don’t. It’s completely arbitrary. We have to balance benefit with plausibility. Clearly the more physical activity we do, the better. But let’s face it, we’re never going to get people to exercise for 10 hours per week. Essentially it’s a matter of, “Well, what do you think guys? Last time we said 30 minutes per day. How about we try 60 this time?”

What makes matters more complex is that there are a lot of different dose–response curves. The curve relating physical activity to the risk of diabetes would look quite different (same shape, but different scale), and the Alzheimer’s curve would be different again. And even different types of physical activity have different relationships to the same health outcomes. Ninety minutes of leisure-time physical activity buys you a 50% reduction in the risk of depression; the same amount of physical activity at work gets you nothing at all (

Most guidelines are a line in the sand, and they sink or swim according to their street credibility. Some time ago, the National Health and Medical Research Council planned to define “binge drinking” as three glasses of wine or four middies of beer per night, to which the then-Opposition Leader, Dr Brendan Nelson, responded: “I mean the average fair dinkum Australian would have to ask himself how it is that four middies of beer can be classified as binge drinking.” How indeed.

For physical activity all we can really say is: “A little is better than none. More is better than a little.”

Professor Tim Olds leads the Health and Use of Time Group at the Sansom Institute for Health Research, University of South Australia.