You really do get by with a little help from your friends. Recent US research indicates that if you have lots of friends, you are likely to live much longer. The study tracked more than 300,000 people over seven years and concluded that having a good support network is as influential a factor in your long-term health as whether or not you're a smoker. The researchers involved are calling for doctors to add patients' social networks to the health-care checklist.

If your doctor asking about your friends seems unlikely (more unlikely, at least, than insurance companies pushing up premiums for loners), perhaps that's because this finding seems entirely unsurprising. Friends keep you happy, which helps keep you healthy. They also assist with practical issues, such as getting to a hospital check-up. But research into life expectancy can throw out surprises. Take its relation to teenage pregnancy, for example: people in low life-expectancy groups become parents earlier.

It has been known for a while that, in the face of deprivation, stress or direct threats, many animals embark on reproduction much quicker. It is now becoming clear that the animal we know as Homo sapiens is no different. Young teenagers' experiences of being threatened, getting into fights or being offered drugs provoke a biological reaction: they become parents much sooner.

Teenage pregnancies aren't just another fact of life in a difficult upbringing. When biologists use controls for all the social factors, an early exposure to menace or mortality seems to be a biological trigger for reproduction. A brush with death makes you broody.

Perhaps it shouldn't really come as a surprise. If your primary goal is to live for ever by passing your genes on to the next generation, there is always a balance to be struck. You want to maximise your progeny's chances of survival by providing a good, stable upbringing. But you also don't want to die childless while waiting for the right conditions. Poverty brings low life expectancy, a low chance of future prosperity, an increased chance of debilitating illness and greater exposure to life-threatening crime. The genes' response? Get on with getting it on before it's too late.

Anyone hoping for a quick fix to teenage pregnancy would do well to step back and look at this bigger picture; the recent announcement that 28 per cent of all teenage girls who are eligible for free school meals have been pregnant at least once (and 7 per cent twice) makes perfect evolutionary sense.

It's not always a good idea to draw social policy directly from the lab; the eugenics movement was hardly biology's finest hour. Nonetheless, cutting benefits would not seem to be the scientifically recommended course of action against teenage pregnancy. The genes won't like it one bit - and you really don't want to make the genes jumpy. That's what got us into this mess in the first place.