Watching television for an average of six hours a day could shorten life expectancy by almost five years, according to a study we published today in the British Journal of Sports Medicine.

What’s more, our research found the impact of sedentary behaviour linked to television viewing rivals that of other major risk factors.

In the 1950s, Professor Jerry Morris reported that conductors on London’s double-decker buses had a lower risk of coronary heart disease than the drivers of those buses, who sat all day.

As the conductors ran up and down the stairs at work, the inference from the observation was that physical exercise was good for health.

Much research followed but surprisingly, it took more than 50 years to discover what might have been obvious from the start: the reverse is also true – too much sitting is bad for health.

Australian study

We used previously published data on the impact of television viewing on death from analyses of the Australian Diabetes, Obesity and Lifestyle Study (AusDiab) and national population and mortality figures to ascertain the degree of risk posed by television viewing.

We wanted to answer questions such as how much higher would the life expectancy of Australians be if they watched no television at all, and how does the effect of television viewing compare with other major risk factors, such as a lack of exercise, smoking and obesity.

The results are striking: our study indicates that if Australian men never watched television, they could expect to live nearly two years longer than they do now. For women, the figure was one and a half years.

Six hours of television a day amounts to almost five years of lost life. That means every single hour of television viewed after the age of 25 reduces life by 22 minutes.

And because Australians are exposed to an average of about two hours of television a day, the associated disease burden may be bigger than that of a lack of exercise, being overweight or obese, or even smoking.

So it appears that many Australians are hastened to their death by their television sets.

How could this be?

Television viewing is a marker for time spent passively sitting.

It’s likely that other sedentary activities such as sitting in front of a computer carry an equivalent risk. But it’s much more difficult to measure these compared with asking people how much television they watched in recent memory.

Theoretically, viewing time also comes with exposure to advertising for unhealthy foods, so perhaps people who watch a lot of television have other habits that kill them.

Our study adjusted for a long list of these including age, gender, leisure-time exercise, waist circumference, smoking, education, total energy intake, alcohol intake, diet quality, hypertension, total cholesterol, medication use, previously reported cardiovascular disease, and glucose tolerance status, among others.

Statistical correction for such factors is never perfect because they can only be measured with limited accuracy, but much of their effect was removed by our adjustment.

And the effects of dietary factors and physical activity are estimated in exactly the same way, so if our results are biased, then much of our knowledge about risk factors for chronic disease is also called into question.

Because it is self-reported, measurement of television viewing is imperfect, and this could in theory have led to an underestimation of its effect on mortality.

And what if we’ve got it all the wrong way around? Perhaps people who are in bad shape and close to death watch more television?

We can’t rule that possibility out although we did exclude persons with a known history of cardiovascular disease.

Other studies

An optimist might think it is all simply due to chance but that’s unlikely. Since the AusDiab paper came out in 2010, two other studies on the same topic have reached similar conclusions.

A study in England found a 4% risk of mortality for every hour of television (but the participants in that study were a bit older and relative risks tend to go down with age because people increasingly die of unrelated causes).

And Scottish data showed a 7% increase in mortality.

A recent meta-analysis of thee three studies in the Journal of the American Medical Association reports a weighted average effect of 6%, with a range of 3% to 9%.

That means each hour of television shortens the viewer’s life by at least eight minutes and as much as 25 minutes.

In all, it seems likely that television viewing does increase the risk of chronic disease and death. Future research can lead to better estimates of the extent of that risk.

What this means for you?

Watching television is a form of sedentary behaviour – the absence of physical activity, as it were.

Until recently, physical activity was measured only with questionnaires. Such surveys are reasonable at picking up deliberate physical exercise, but bad at measuring light physical activity.

This led to recommendations for exercise, and talk of a level of physical activity so low that it confers no health benefit.

These days, pedometers and accelerometers can measure almost every step, and it seems increasingly likely that every step counts.

Health recommendations for children include limiting viewing time. Our study suggests that adults would do well to follow the same advice.