People are better at estimating their own risk of an illness if they stick to personal experience rather than number crunching, say researchers.
Thinking about personal experience can help us judge risk
By asking yourself how many of your friends and relatives have had a disease, you will reach a better judgement of personal risk, they say.
In comparison, those who rely on media reports of statistics can be way off mark, the German and Swiss team found.
But there are pitfalls with the Journal of Experimental Psychology method.
People who are over-anxious about their health or have had an unusually unfortunate experience may overestimate risk, for example.
Doctors know it can be difficult for patients to interpret risk.
For example, one person's understanding of "likely" to get a disease may be a chance of one in 10, whereas another may think it means a chance of one in two.
Official data listing numbers and risk can be daunting and difficult to interpret and media reports about disease outbreaks can cause undue concern.
Dr Ralph Herwig from the University of Basel in Switzerland, working with German colleagues from the Max Planck Institute for Human Development in Berlin, decided to look at how individuals judged risk and which method used was best.
They asked 110 students at their universities to rank pairs of illnesses according to risk - which one of the pair would kill or affect more people.
Overall, the students were correct with their estimate between 70% and 80% of the time.
It appeared that the students were using two methods to come to their estimates, 'availability by recall' and 'regressed frequency mechanism'.
With availability by recall, the individual assess the odds of an event by the frequency of episodes within their social network - i.e. how many of their friends and family had the disease.
The regressed frequency mechanism is when people base their risk on data from a multitude of sources, such as news reports, health campaigns, doctors' warnings.
But because it is hard to take in all this information at the same time, people's estimates shift towards an average value.
As a result, the risk of diseases that are actually rarer but have been frequently mentioned by media tend to be overestimated by the person while more illnesses are underestimated.
Dr Hertwig said: "People can arrive at relatively accurate estimates...by thinking of how many of their relatives, friends and acquaintances died from these causes.
"However, when they start sampling from the virtual world as created by the mass media, they are more likely to arrive at distorted estimates of likelihood."
For example, some might conclude from media coverage that more people die from mad cow disease than asthma.
Dr Christine Bundy, a health psychologist from the University of Manchester who helps counsel well and sick people about their risks of disease, said the findings made sense.
Help or hindrance
"When dealing with health risks that we have not had we tie it into personal experience and try to think of people who have had it to make sense of it."
However, she said the availability by recall method did not always work.
Some people have difficulty of equating a risk, for example a one in five risk of heart attack, to themselves.
"We can't say this one in five people living in that household will get it, we can only apply it to a general population, so people can find that hard."
Also, personal experience, such as illness of a loved one, can skew judgement.
"It also depends on the person's outlook. Most people underestimate their risk of developing a common disorder and overestimate their chances of remaining healthy."
In the extreme, people can think they are invincible and take excessive risks with their health.
At the other end of the spectrum are people who have a morbid belief that they are going to become ill.
People also put off thinking about their health and health risks until tomorrow, particularly when they are young.
She said it was helpful for doctors to know how an individual judges risk when giving health advice.
For example, urging a heavy smoker to quit by outlining their risks of dying from cancer and heart disease might backfire because they may just go off and smoke more because they are now anxious, she said.