People who sleep fewer than six hours a night are more likely to die early, researchers have found in a study they claim provides "unequivocal evidence" of a link between sleep deprivation and premature death. They discovered that people who slept for less than six hours each night were 12 percent more likely to die prematurely—before the age of 65—than those who slept the recommended six to eight hours a night.

The team from the University of Warwick and Federico II University Medical School in Naples analysed 16 studies involving a total of 1.3 million people before reaching their conclusions. They pointed out that previous studies had shown that sleep deprivation was associated with heart disease, high blood pressure, obesity, type II diabetes, and high cholesterol.

However, the researchers also found that sleeping too much was linked to an early death. Those who slept for more than nine hours a night were 30 percent more likely to die early, the research published in the journal Sleep found. That directly contradicts research published in the same journal last week which suggested that people who slept for ten hours or longer a night were more likely to live to 100. This was thought to be because people who lived into extreme old age were healthier and therefore slept better.

However, the authors of the latest research contradicted this and suggested that long sleep was a sign of underlying illnesses such as depression and low levels of physical activity. Some cancer is also associated with sleeping for longer. Professor Francesco Cappuccio, leader of the Sleep, Health and Society Programme at the University of Warwick and Consultant Physician at the University Hospitals Coventry and Warwickshire NHS Trust, said: "Whilst short sleep may represent a cause of ill-health, long sleep is believed to represent more than an indicator of ill-health."

"Consistently sleeping six to eight hours per night may be optimal for health. The duration of sleep should be regarded as an additional behavioural risk factor, or risk marker, influenced by the environment and possibly amenable to change through both education and counselling as well as through measures of public health aimed at favourable modifications of the physical and working environment."